Work Team Effectiveness In Organizational Contexts : Recent Research And Application In Spain And Portugal 9781845442798, 9781845441173

Citation preview

jmp cover (i).qxd

19/04/2005

13:46

Page 1

ISBN 1-84544-117-6

ISSN 0268-3946

Volume 20 Number 3/4 2005

Journal of

Managerial Psychology Work team effectiveness in organizational contexts: recent research and applications in Spain and Portugal Guest Editors: Francisco Gil, Carlos-María Alcover and José-María Peiró

www.emeraldinsight.com

Journal of Managerial Psychology

ISSN 0268-3946 Volume 20 Number 3/4 2005

Work team effectiveness in organizational contexts: recent research and applications in Spain and Portugal Guest Editors Francisco Gil, Carlos-Marı´a Alcover and Jose´-Marı´a Peiro´

Access this journal online _________________________

191

Editorial advisory board __________________________

192

INTRODUCTION Work team effectiveness in organizational contexts: recent research and applications in Spain and Portugal Francisco Gil, Carlos-Marı´a Alcover and Jose´-Marı´a Peiro´______________

193

Types of intragroup conflict and affective reactions Francisco J. Medina, Lourdes Munduate, Miguel A. Dorado, Ine´s Martı´nez and Jose´ M. Guerra_________________________________

219

Exploring the effects of intragroup conflict and past performance feedback on team effectiveness Ana Margarida Passos and Anto´nio Caetano _______________________

231

Affective responses to work process and outcomes in virtual teams: effects of communication media and time pressure Amparo Caballer, Francisco Gracia and Jose´-Marı´a Peiro´ ______________

245

Effects of task interdependence and type of communication on performance in virtual teams Ramo´n Rico and Susan G. Cohen _________________________________

Access this journal electronically The current and past volumes of this journal are available at:

www.emeraldinsight.com/0268-3946.htm You can also search more than 100 additional Emerald journals in Emerald Fulltext (www.emeraldinsight.com/ft) and Emerald Management Xtra (www.emeraldinsight.com/emx) See page following contents for full details of what your access includes.

261

CONTENTS

CONTENTS continued

Idea generation through computer-mediated communication: the effects of anonymity Joa˜o Pissarra and Jorge C. Jesuino ________________________________

275

Organisational support and group efficacy: a longitudinal study of main and buffer effects Amparo Osca, Begon˜a Urien, Genoveva Gonza´lez-Camino, M. Dolores Martı´nez-Pe´rez and Nuria Martı´nez-Pe´rez _________________

292

Change-oriented leadership, satisfaction and performance in work groups: effects of team climate and group potency ´ ngel Barrasa ____ Francisco Gil, Ramo´n Rico, Carlos-Marı´a Alcover and A

312

www.emeraldinsight.com/jmp.htm As a subscriber to this journal, you can benefit from instant, electronic access to this title via Emerald Fulltext and Emerald Management Xtra. Your access includes a variety of features that increase the value of your journal subscription.

Additional complimentary services available

How to access this journal electronically

E-mail alert services These services allow you to be kept up to date with the latest additions to the journal via e-mail, as soon as new material enters the database. Further information about the services available can be found at www.emeraldinsight.com/alerts

To benefit from electronic access to this journal you first need to register via the internet. Registration is simple and full instructions are available online at www.emeraldinsight.com/admin Once registration is completed, your institution will have instant access to all articles through the journal’s Table of Contents page at www.emeraldinsight.com/0268-3946.htm More information about the journal is also available at www.emeraldinsight.com/ jmp.htm Our liberal institution-wide licence allows everyone within your institution to access your journal electronically, making your subscription more cost-effective. Our web site has been designed to provide you with a comprehensive, simple system that needs only minimum administration. Access is available via IP authentication or username and password.

Key features of Emerald electronic journals Automatic permission to make up to 25 copies of individual articles This facility can be used for training purposes, course notes, seminars etc. This only applies to articles of which Emerald owns copyright. For further details visit www.emeraldinsight.com/ copyright Online publishing and archiving As well as current volumes of the journal, you can also gain access to past volumes on the internet via Emerald Fulltext and Emerald Management Xtra. You can browse or search these databases for relevant articles. Key readings This feature provides abstracts of related articles chosen by the journal editor, selected to provide readers with current awareness of interesting articles from other publications in the field. Non-article content Material in our journals such as product information, industry trends, company news, conferences, etc. is available online and can be accessed by users. Reference linking Direct links from the journal article references to abstracts of the most influential articles cited. Where possible, this link is to the full text of the article. E-mail an article Allows users to e-mail links to relevant and interesting articles to another computer for later use, reference or printing purposes. Emerald structured abstracts New for 2005, Emerald structured abstracts provide consistent, clear and informative summaries of the content of the articles, allowing faster evaluation of papers.

Your access includes a variety of features that add to the functionality and value of your journal subscription:

Research register A web-based research forum that provides insider information on research activity world-wide located at www.emeraldinsight.com/researchregister You can also register your research activity here. User services Comprehensive librarian and user toolkits have been created to help you get the most from your journal subscription. For further information about what is available visit www.emeraldinsight.com/usagetoolkit

Choice of access Electronic access to this journal is available via a number of channels. Our web site www.emeraldinsight.com is the recommended means of electronic access, as it provides fully searchable and value added access to the complete content of the journal. However, you can also access and search the article content of this journal through the following journal delivery services: EBSCOHost Electronic Journals Service ejournals.ebsco.com Huber E-Journals e-journals.hanshuber.com/english/index.htm Informatics J-Gate www.j-gate.informindia.co.in Ingenta www.ingenta.com Minerva Electronic Online Services www.minerva.at OCLC FirstSearch www.oclc.org/firstsearch SilverLinker www.ovid.com SwetsWise www.swetswise.com TDnet www.tdnet.com

Emerald Customer Support For customer support and technical help contact: E-mail [email protected] Web www.emeraldinsight.com/customercharter Tel +44 (0) 1274 785278 Fax +44 (0) 1274 785204

JMP 20,3/4

EDITORIAL ADVISORY BOARD

EDITORIAL REVIEW BOARD

Professor Neil Anderson Department of Work and Organization Psychology, University of Amsterdam

Dr Dean Bartlett University of North London, UK

Professor Chris Argyris Harvard Graduate School of Business Administration, USA

192

Professor Yehuda Baruch School of Management, University of East Anglia, UK Professor Frank Bournois Universite´ Panthe´on-Assas, Paris II, France Professor Cary Cooper Lancaster University Management School, Lancaster Dr Frank Heller Tavistock Institute, UK Professor Geert Hofstede Institute for Research on Intercultural Co-operation, The Netherlands Professor Paul Iles Teesside Business School, UK Professor Andrew Kakabadse Cranfield School of Management, UK, Founding Editor of Journal of Managerial Psychology Dr Bruce Kirkcaldy International Centre for the Study of Occupational and Mental Health, Du¨sseldorf, Germany

Dr Gayle Baugh University of West Florida, USA Dr Adrian Carr School of Applied Social and Human Sciences, University of Western Sydney, NSW Alf Crossman The University of Surrey, UK Dr Patricia Hind Ashridge Management College, UK Professor Henry S.R. Kao University of Hong Kong, Hong Kong Dr Ute-Christine Klehe Programmagroep A&O Psychologie, Amsterdam Dr Robert Kovach and Brett Seamons RHR International Co., London, UK Dr Peter Liu Verity International Ltd, Toronto Dr Tuvia Melamed Quo Consulting Ltd, Redditch, UK Dr Michael Morley University of Limerick, Ireland Dr Chris Rees University of Manchester, UK Dr Ramon Rico Universidad Autonoma de Madrid, Spain

Professor Harold J. Leavitt Stanford University, USA

Alain M. Roger IAE de Lyon, Universite´ Jean Moulin, Lyon

Professor Manuel London State University of New York, Stony Brook, USA Professor Dr Wolfgang Mayrhofer Vienna University of Economics and Business Administration, Austria

Dr Raymond Saner and Dr Lichia Yiu Centre for Socio-Economic Development, Geneva, Switzerland Dr Rene´ Schalk Tilburg University, The Netherlands

Professor Greg Northcraft Graduate School of Business, Stanford University, USA

Dr Ruth Simpson School of Business and Management, Brunel University, UK

Dr Francisco Gil Rodriguez Universidad Complutense de Madrid, Spain

Dr Sherry E. Sullivan Bowling Green State University, USA

Professor Zhong-Ming Wang Hangzhou University, China

Professor Gladys L. Symons Ecole Nationale d’Administration Publique, Universite´ de Quebec, Canada Dr Shay Tzafrir University of Haifa, Israel Dr Daniel Vloeberghs University of Antwerp, Belgium

Journal of Managerial Psychology Vol. 20 No. 3/4, 2005 p. 192 # Emerald Group Publishing Limited 0268-3946

Dr Jacob (Yaacov) Weisberg Bar-Ilan University, Israel Professor Jack Wood Monash University, Australia

The Emerald Research Register for this journal is available at www.emeraldinsight.com/researchregister

The current issue and full text archive of this journal is available at www.emeraldinsight.com/0268-3946.htm

INTRODUCTION

Work team effectiveness in organizational contexts Recent research and applications in Spain and Portugal

Work team effectiveness

193

Francisco Gil Universidad Complutense de Madrid, Madrid, Spain

Carlos-Marı´a Alcover Universidad Rey Juan Carlos, Madrid, Spain, and

Jose´-Marı´a Peiro´ Universidad de Valencia, Valencia, Spain Abstract Purpose – This introductory paper aims to provide a contextualization of recent research and applications on work team effectiveness in organizational contexts carried out in Spain and Portugal and to describe connections between this research and the main trends in the international scene. Design/methodology/approach – Since the 1990s, new occupational and organizational realities have deepened scientific interest in work teams in both Spain and Portugal. A range of recently published (1992-2004) works in this area are reviewed. The selected sources are papers published in Spanish, Portuguese and international journals. Findings – Reviewing this work, four major trends are identified that synthesize the key concerns of researches in both countries: work teams and new information/communication technologies; intra- and inter-group conflicts in organizational contexts; definition, dimensions and measurement criteria for work team effectiveness; and teams in innovation and change processes. Originality/value – This paper fulfils a panoramic vision of this research area in both Spain and Portugal, and provides an overview of the papers included in this special issue and an outlook for the future. Keywords Team working, Spain, Portugal Paper type General review

1. Introduction Current research on work teams in organizational contexts stems, in the main, from the development of psycho-social theory and inquiry in the field of group psychology since the 1930s. This body of learning was largely generated in the laboratory using experimental methods (McGrath et al., 2000), and comparatively few studies were performed in real contexts, mostly in the 1950s (Sundstrom et al., 2000). It is no surprise, then, that scholars are divided in their evaluations of the results obtained. Some (e.g. West, 1996) hold that the tradition provides a considerable source of useful findings for our understanding of the way in which teams function, while others (e.g. Bramel and Friend, 1987; Cannon-Bowers et al., 1992; Ilgen et al., 1993) have criticized the data as fragmentary, incomplete, confused and, at times, contradictory, although they recognize its value.

Journal of Managerial Psychology Vol. 20 No. 3/4, 2005 pp. 193-218 q Emerald Group Publishing Limited 0268-3946 DOI 10.1108/02683940510589000

JMP 20,3/4

194

Over the past 25 years research on groups and work teams has shifted away from the field of social psychology into the domain of work and organizational psychology (Levine and Moreland, 1990; Simpson and Wood, 1992). As Steiner (1986) remarks: The group is too important to an understanding of human behavior and the knowledge of society to be forever neglected. If social psychologists do not research the group, someone else surely will (p. 283).

This resurgence of interest has ushered in a golden age, in terms of both the volume and quality of work. Nevertheless, the limited scope of theory and the absence of any comprehensive conceptual framework, problems traditionally associated with psycho-social inquiry, continue to hinder the accumulation of findings that would represent real progress in our knowledge. The reawakening of interest in groups and work teams is related with wider changes in the world of work and organizations driven by economic, strategic and technological imperatives. The pressures of global competition, the need to consolidate business models in complex and shifting environments and the pursuit of continuous innovation have led to a reappraisal of the team as a key element of the basic organizational architecture (Kozlowski and Bell, 2003). In this context, the joint action of individuals working together in a cooperative manner to attain shared goals through the differentiation of roles and functions, and the use of elaborate communication and coordination systems, are now viewed as essential to effectiveness and competitive advantage. As West (2001) points out, the current enthusiasm for work groups and team working: Reflects a deeper, perhaps unconscious, recognition that this way of working offers the promise of greater progress than can be achieved through individual endeavor or through mechanistic approaches to work (p. 270).

The mechanistic approach may, of course, be appropriate when activities are stable, planned and predictable, but the modern organization must respond ever more quickly, flexibly and adaptively to shifting circumstances and demands that are often difficult to foresee. Such situations can only be successfully addressed by combining diverse but interrelated competencies, skills, knowledge and experiences, and through cooperation between the members of the organization (Tjosvold et al., 2003). Teams respond to these needs (Kozlowski and Bell, 2003). Unsurprisingly, research in the past two decades has focused almost exclusively on the effectiveness of work teams in organizational contexts. Given the prevalence and expanding role of teams in organizations of all kinds (Devine et al., 1999) the key research issue is to enable individually competent and skilled people to synergize as a team. To put it another way, the question is how to turn a team of experts into an expert team (Salas et al., 1997). The urgent need to discover the factors involved in work team performance and effectiveness, and to develop powerful training strategies, has, however, thrown up a myriad of theoretical models and applications, which are often insufficiently integrated to be of much practical use (Salas et al., 2004). 2. Models of work team effectiveness in organizational contexts With the aim mapping research into work team effectiveness in organizational contexts, Salas et al. (2004) group the dozens of models into two major integrative frameworks.

The first comprises models that follow the “input-process-output” structure initially applied to groups by McGrath (1964) and subsequently refined by Hackman and Morris (1975). This has been dubbed the functional approach (Wittenbaum et al., 2004, p. 19), and it is defined by three basic assumptions: (1) Groups are goal-oriented. (2) Group performance varies in quality and quantity, and can be evaluated. (3) Internal and external factors influence group performance via the interaction process. It is currently represented by the integrated model of Tannenbaum et al. (1992). The model identifies four types of input variable, comprising: (1) Members’ attributes (e.g. knowledge and skills, motivation, attitudes, mental models and so forth). (2) Work structure (task assignment, team rules and communication structure). (3) Team attributes (distribution of power, similarity of members, team resources, climate and cohesion). (4) The nature of the task (organization, type and complexity of tasks). These variables relate not only to each other but also to the processes performed by the team over time. Processes stand at the center of the structure and include coordination, communication, conflict resolution, decision-making and problem solving, and boundary spanning. Such processes are, in turn, directly related with performance results or output variables, which are usually classified into three groups: (1) Changes in the team (rules, roles, communication patterns and new processes). (2) Team performance (quantity, quality, timing, errors and costs). (3) Individual changes (changes in the input variables represented by individual attributes such as attitudes, motivation, mental models, etc.). Team results provide feedback, upholding or changing the system. One key aspect of this model is that it accounts for the organizational and situational factors that may impact the structure as a whole, affecting input variables, processes and outputs. These factors include reward systems, resource scarcity, management control, environmental stress, organizational climate, competence, inter-group relations and environmental uncertainty. It also addresses the role of team interventions, basically individual and group training programs, and techniques such as team building and development. Structural input-process-output models usually measure team effectiveness in the wide sense on the basis of triple criteria that coincide with the three types of results obtained (Guzzo and Dickson, 1996). These are: (1) Productive results (quantity-quality, cost savings, shorter waiting times, customer complaints, etc.). (2) Member outcomes (acquisition of knowledge, skills, etc., changes in attitudes and personal development). (3) Viability, or the capacity of the team to continue functioning effectively in the future.

Work team effectiveness

195

JMP 20,3/4

196

A fourth measure sometimes applied to assess effectiveness is the degree of innovation in terms either of outcomes or work processes (e.g. Unsworth and West, 2000; West et al., 1998). As Ilgen (1999) notes, research on work teams in organizational contexts has led to a shift in interest from the center (processes) to the right (results or effectiveness) and/or to the left (input variables). As a consequence, the focus has switched to the role of input and context variables as determining factors of team effectiveness, although the importance of processes is accepted. The second group of models mapped by Salas et al. (2004) includes the theoretical “meta” framework for work team effectiveness proposed by Campion and colleagues (Campion et al., 1993; Campion et al., 1996), which integrates and extends numerous earlier models. These scholars define five major categories of team effectiveness variables, comprising job design, interdependence, composition, context and process. Job design refers to self-management, participation and the variety, significance and identity of tasks. Interdependence subsumes tasks and goals, and feedback/rewards. Composition refers to member heterogeneity, flexibility, relative team size and the preference for team work. Context variables include training, managerial support and communication/cooperation between groups. Finally, processes comprise group potency, social support, workload sharing and communication/cooperation between group members. The variables subsumed in both theoretical frameworks are clearly similar. In fact, the only difference is the significance assigned to each and the relationships established. Thus, while input-process-output models structure variables in some kind of order, the theoretical model proposed by Campion and colleagues treats the links between variables and effectiveness more directly. Both approaches highlight the numerous variables influencing team effectiveness, the multitude of interrelationships and, above all, the complex dynamics inherent in the way teams function in contemporary organizational contexts (Cohen and Bailey, 1997; Harris and Barnes-Farrell, 1997; Hyatt and Ruddy, 1997; Jordan et al., 2002; Levi and Slem, 1995; Salas et al., 2004; Zalesny et al., 1995). The two frameworks also provide an overview of the methods used to measure team effectiveness (Gibson et al., 2000; Sawyer et al., 1999; Whiteoak et al., 2004). It seems clear, then, that future research will be based on the variables identified in these models, whatever the approach taken, while the results obtained should help clarify and integrate the determining factors of team effectiveness. 3. Spanish and Portuguese research on work team effectiveness The recent histories of Spain and Portugal are in many respects very similar. During the twentieth century both were underdeveloped compared to their more advanced European peers, a legacy in both cases of the slow decay of the great empires established in the sixteenth and seventeenth centuries. At the same time, long-lasting fascist dictatorships aggravated both countries’ marginal role in the modern international order (Telo and de la Torre, 2003) by persisting with protectionist and isolationist policies towards neighboring democracies, which reacted by curtailing political and economic involvement beyond the Pyrenees. After 1970, however, the dictatorships were toppled, first in Portugal and later in Spain, setting in train a transition toward democratic government and a swift process

of social, political, economic, cultural and technological modernization. While these developments did not entirely eliminate material backwardness, both countries were accepted into the European Union in the mid-1980s and have since succeeded in closing the economic gap with the richer European states (Pinto and Teixeira, 2002). In a little over two decades Portugal and Spain have seen far-reaching change at every level of society and are now much closer to their true potential than could ever have been achieved under the former regimes. The organizational and social contexts in both countries are now comparable to the conditions found in the UK, France, Germany and Italy, which are similar to those of other advanced nations. These outcomes are, of course, traceable to economic and financial globalization and, particularly, the adoption and spread of the new technologies. At the same time, other cultural and social factors have tended to bring the realities of a large part of the world closer together, at least in the somewhat fuzzily defined West. Scientific research in the fields of work and organizational psychology in both Spain and Portugal thus goes ahead in a similar context to Western Europe, North America and the developed countries of Asia. It is hardly surprising, then, that our concerns have much in common with the lines of inquiry pursued by colleagues abroad. Among these issues, work teams have attracted enormous attention since the 1990s, and a particular interest has emerged in the study of the variables and processes we describe below. Meanwhile, let us briefly explain the current orientation of Spanish and Portuguese research. We shall begin by reflecting on scholarly interest in teams and then go on to discuss the variables (following the models identified by Salas et al. (2004)) that have claimed researchers’ attention, as well as the connections between their work and the main trends in international science. Finally, we shall try to explain local preferences by placing team research in its social and cultural context in Spain and Portugal.

3.1 The Spanish and Portuguese interest in teams Spain and Portugal have seen major changes in the conceptualization and organization of work in line with the business and occupational contexts created in general by the global economy. Meanwhile, organizations have had to restructure and innovate to share in the European market (Cunha, J.V. and Cunha, M.P., 2001a; Peiro´, 2001a, 2002; Peiro´ and Munduate, 2001). A key consequence of this transformation (Peiro´ and Munduate, 1999) has been to drive the development of team working through a process of conceptual change in the organization of work. Thus, the notion of a set of discrete tasks clustered together as a job has increasingly given way to conceptualization in terms of flexible areas of responsibility linked to individuals and teams, who cooperate to achieve specific goals. This new system devolves greater responsibility for decision-making upon team members, enhances their commitment to results-based performance, and ensures involvement in the processes of innovation and organizational change. The new occupational and organizational realities have deepened scientific interest in work groups and teams in a movement affecting both Spain (Alcover et al., 2000; Ayestara´n, 1998a; Gil and Alcover, 2002) and Portugal (Cunha, M.P. and Cunha, J.V., 2001) since the 1990s.

Work team effectiveness

197

JMP 20,3/4

198

3.2 Variables and links with international lines of research The following discussion concerns the main lines of Spanish and Portuguese research based on the variables explored. 3.2.1 Work teams and new information/communication technologies. This area has received considerable attention thanks to the efforts of the University of Valencia’s Organizational and Work Psychology Research Unit (Unidad de Investigacio´n de Psicologı´a de las Organizaciones y del Trabajo – UIPOT) directed by Professor Peiro´, and of the team led by Professors Salanova and Grau at the Jaume I University in Castello´n. Both of these research teams are, of course, located in Spain. Spanish and Portuguese scholars were responsible for three out of the six papers published in the monograph issue of Small Group Research given over to European perspectives on electronically mediated communication, which clearly reflects the importance of this line of research in the Peninsula (Jesuino, 2002). In these papers, Zornoza et al. (2002b) from the University of Valencia analyze differences in conflict management between face-to-face teams and in groups communicating through computer networks; Dorado et al. (2002), all from the University of Seville except the last (University of Utrecht, The Netherlands), consider electronically mediated negotiating processes in scaled conflicts; and Amaral and Monteiro (2002) from the Lisbon-based Institute for Work and Business Science (Instituto Superior de Cieˆcias do Trabalho e da Empresa) examine the processes by which members of groups communicating through computer networks create new identities. A number of other significant papers have been published by researchers at the two Spanish universities mentioned above. Among these we may cite Salanova et al. (2003), the first four from Jaume I University in Castello´n and the last from the University of Utrecht, who consider the impact of electronic working practices in teams on well-being and performance, finding a moderating effect on perceived collective efficacy. Thus, perceived low performance had a negative impact on well-being (anxiety), while high performance had a positive influence (engagement). Meanwhile, Orengo et al. (2000) have studied the influence of member familiarity, group atmosphere and assertiveness on uninhibited behavior in teams working in situations of face-to-face communication, videoconferencing and computer mediated communication (e-mail). Taking the same three group communication media, Gracia et al. (2002) analyzed the effect of time pressure on group cohesion in teams involved in three different task types (creative, intellectual and mixed-motive tasks), while Orengo et al. (1996) studied group interaction (measured in terms of conflict management and uninhibited behavior) in a multi-decision task. Observations were measured using the SYMLOG technique (Orengo et al., 1998). Martı´nez et al. (2002) consider the modulating effect of information technology on affective psychological well-being in task groups, and Martı´nez and Mejı´as (2003) examine the influence of anonymity and gender upon consensus levels, work group cohesion, and the satisfaction of individuals working in group decision-making environments supported by information technology (e.g. group support systems). Ripoll et al. (1998) analyzed differences in the results achieved by work teams depending on the communication channel employed and the experience gained in the use of technological media. Finally, Zornoza et al. (1992) examined the influence of previous attitudes to and experiences with new technologies on team performance, while Zornoza et al. (1993)

analyzed the effects of different communication media (face-to-face, videoconferencing and e-mail) as a moderating variable for work team performance. Zornoza et al. (2002a) consider how group processes develop and adapt to the communication media employed (face-to-face, videoconferencing and e-mail) depending on time pressures and the task type (creative or intellectual tasks, and conflict). Another key line of inquiry has been developed by Joa˜o Vieira da Cunha of the MIT Sloan School of Management in Cambridge, Massachusetts (USA) and Miguel Pina e Cunha of the New University of Lisbon. This focuses on improvisation in virtual teams taking into account their structure (Cunha, J.V. and Cunha, M.P., 2001b) and multicultural diversity (Cunha, M.P. and Cunha, J.V., 2001). This work is of particular interest in that the authors apply a qualitative methodology (grounded theory), in contrast to the other papers cited in this section, which mainly employ quantitative methodologies. Finally, a team at the University of Malaga (Spain) has studied the factors associated with virtual communication and group efficacy (Canto and Moral, 2003), and the in-group minority influence of computer-mediated communication on different group processes (Moral et al., 2003). Scholars from the three Madrid universities, Gil et al. (2003) have recently opened up a line of research concerning leadership in virtual work teams. 3.2.2 Work team climate. Another fruitful line of research developed by the University of Valencia’s Organizational and Work Psychology Research Unit (Spain) focuses on the work team climate. In outline, these contributions refer to climate measurement, the background and consequences of individual climate perceptions, and the formation of an aggregate climate in terms of shared group perceptions, their correlations and influence (for a review, see Gonza´lez-Roma´ and Peiro´, 1999). Inquiry has centered on the validation of collective (Gonza´lez-Roma´ et al., 1999) and affective climates in work teams (Gonza´lez-Roma´ et al., 2000); climate formation within teams as an outcome of social interaction (Gonza´lez-Roma´ et al., 1995); the evolution of climate in work teams and job satisfaction (Man˜as et al., 1998); climate history and its consequences for team outcomes (Man˜as et al., 1999); the influence of team climate on role stress, tension, job satisfaction and leadership (Peiro´ et al., 1992); and the antecedents and moderating influence of the strength of team climate (Gonza´lez-Roma´ et al., 2002). 3.2.3 Cultural factors in work teams. Let us begin this discussion by distinguishing between research into the variables inherent in the culture of a society or nation (e.g. individualism, collectivism, power distance, etc.) and organizational or team culture. Also, many such studies seek to elucidate the relationship between cultural dimensions and conflict management, an issue that is dealt with in a separate section of this review. For this reason they are mentioned only in passing here. The research group directed by Professor Sabino Ayestara´n at the University of the Basque Country and the team led by Professor Lourdes Munduate at the University of Seville (Spain) have published a number of important papers in the field of national cultural differences. For example, Arro´spide et al. (1995) analyzed group members’ perceptions of hierarchy and conflict management. Ayestara´n (1998b) studied the functioning of work teams based on the variables of power distance, individualism/ collectivism and identification with the group’s goals. Finally, Ayestara´n and Reoyo (Ayestara´n and Reoyo, 1998; Reoyo, 1998) consider relationships between culture, conflict management style and identification in work teams.

Work team effectiveness

199

JMP 20,3/4

200

Meanwhile, Cruces et al. (1999) have studied the influence of cultural (individualism/collectivism) and socio-emotional factors on negotiation processes, focusing particularly on lying in negotiation teams, and Munduate and Dorado (1999) have sought to explain conflict in work teams in terms of cultural factors (e.g. cooperative versus competitive cultures). In the field of organizational and team culture, we may refer to Ayestara´n’s important contributions and various studies of work team training, conflict management and cultural change within organizations by the Spanish Distance Learning University (Universidad Nacional de Educacio´n a Distancia or UNED). The UNED scholars Alonso and Sa´nchez (1998) analyzed organizational commitment on the basis of the predominant culture of work teams and the relationship between culture and group cohesion and identification (Sa´nchez and Alonso, 2003), while Osca et al. (1994) studied organizational culture in primary healthcare and multi-professional teams working in drug dependence centers. 3.2.4 Intra- and inter-group conflicts in organizational contexts. Research into intra-group conflicts has been closely linked to the work on cultural factors described in the preceding section (see Arro´spide et al., 1995; Ayestara´n, 1999; Ayestara´n and Reoyo, 1998; Munduate and Dorado, 1999) and to the study of the new information and communication technologies in work teams (see, Dorado et al., 2002; Orengo et al., 1996; Peiro´ and Munduate, 1999); Zornoza et al., 2002b). In addition to these lines of inquiry, we may also mention work on the dimensions of conflict within work teams. Passos and Caetano (n.d.) have found empirical evidence to support a three-dimensional model of intra-group conflict, distinguishing between relational conflict, task conflict and process conflict. Medina et al. (2003), meanwhile, confirm that these conflict types may actually enhance efficacy when team members can enter into constructive debate about their work in a job climate that enables group goal attainment, provided tasks are relatively simple and routine. The same authors (Medina et al., 2004) examine the relationship between conflict types – task and relationship conflict – and work group climate, finding that relationship conflict worsens work group climate, while task conflict increases innovation and goal climate. Also, worker satisfaction increases in teams if high levels of task conflict combine with a high goal climate. These results suggest that constructive controversy may be productive in a climate where teams have compatible goals and are offered contingent rewards. Finally, Monteiro (2003) proposes using the dimensions of identity and diversity inherent in social interaction in psycho-social research applied to inter-group conflicts in organizational contexts. 3.2.5 Group processes in work teams. This section discusses a series of papers concerned with the group processes involved in the functioning and effectiveness of work teams. Let us begin with a mention to research on the processes related with innovation in teams and their role in organizational change. Curral and Chambel (1999) at the University of Lisbon examined relationships between group processes (following the model proposed by Anderson and West (1996) who define four types: (1) Vision. (2) Participation.

(3) Task orientation and support for innovation. (4) Innovation. Such group processes were found to be more prevalent in teams producing more and better innovations than in less innovative teams in terms of both quantity and quality. Curral et al. (2001) subsequently confirmed that teams scored higher for group participation and innovation support processes where the demand for innovation was greater. Bigger teams, meanwhile, exhibited poorer group processes, and large teams working under intense pressure to innovate showed weaker group processes than peers with less pressure to innovate. Ayestara´n (1993) examined organizational innovation processes implemented through work teams, as well as the variables influencing their effectiveness as vehicles of change. Costa et al. (1997) have analyzed the role of quality circles as tools to gain acceptance for organizational change, while Gonza´lez-Roma´ et al. (1992) have studied the factors influencing participation in QCs. Finally, Gonza´lez-Roma´ et al. (1987) examined the impact of QCs on climate, performance and absenteeism in organizational contexts. On the issue of organizational change, Aritzeta (2001) has considered the effects of self-managing work teams on group characteristics related with effectiveness, job satisfaction, commitment, self-esteem and personal autonomy. In general SMWTs were found to have generally beneficial effects compared to traditional individual working. Looking at the effects of certain organizational flexibility strategies, Alcover and Gil studied the impact of member continuity and change over time on the functioning and effectiveness of work teams (Alcover and Gil, 1998), as well as the mediating role of group members’ anticipation of change (Alcover and Gil, 1999a) and the influence of different modes of organizational flexibility (temporal, functional, numerical and technological) on work team processes and outcomes (Alcover and Gil, 1999b). Meanwhile, Wood and Tabernero (2000) have sought to explain how beliefs about ability influence behavior through their effect on perceived efficacy and individual and group reactions to performance. Aritzeta (2003) has shown how organizational support for work teams and the interdependence of goals positively predict members’ organizational commitment, while raising teams’ workloads has a negative impact on commitment. Finally, Ripoll et al. (2004) have studied the modulating role of communications media in the relationship between the communication process (coordination, commitment to the solution and socio-emotional behavior) and group effectiveness (performance and cohesion) when groups are involved in a variety of different tasks. Their results suggest that the influence of interaction processes on group outcomes varies depending on the task participants are asked to perform, and on the communication media used. Thus, it may be important to train work teams to handle group interaction processes, especially when tasks are complex. 3.2.6 Effectiveness criteria and dimensions. The last line of inquiry we shall discuss here focuses on the criteria for and dimensions of work team effectiveness. Concerned with the definition of the phenomenon, Lourenc¸o and colleagues have defended the need to analyze the multiple facets of work team effectiveness in order to identify the forms it may take (Lourenc¸o et al., 2000; Lourenc¸o et al., 2004). These authors have also

Work team effectiveness

201

JMP 20,3/4

202

established the two-dimensional nature of effectiveness, distinguishing between social or affective factors and technical or task factors to propose a coherent structure for a socio-technical paradigm (Lourenc¸o and Gomes, 2003). In another paper, Lourenc¸o (2000) proposes an alternative approach to the relationship between effectiveness and leadership. This treats effectiveness as a condition for the exercise of leadership in an effort to expand the explanatory capacity of both phenomena in work teams and organizations. Meanwhile, Segurado et al. (2004) review the different methodologies used to measure work team effectiveness. A number of studies also consider the effects of training and intervention on work teams (Reoyo and Lo´pez Canas, 2003). Such actions are found to boost collaboration and goal identification, as well as increasing competition between members to improve tasks, reducing conformity and raising satisfaction, all of which enhance the effectiveness of work teams. 3.3 Research on work team effectiveness and the organizational context in Spain and Portugal Spanish and Portuguese research work teams closely follows the main trends of international science, as shown by the above discussion. Nevertheless, we may identify four major areas that synthesize the key concerns of researchers in the Iberian Peninsula. These can be represented graphically using the systemic model (Figure 1). In the first place, interest centers on the multiple effects of the new information and communication technologies (ICTs) on both input variables (e.g. members’ attitudes and expectations, task structure and interdependence, levels of training) and on team processes (e.g. decision making, communication, leadership, conflict), which in turn will have an impact on team outcomes or effectiveness. The second focus is defined by the analysis of intra- and inter-group conflict and its effect on outcomes. Thirdly, there is considerable concern for the definition and measurement of work team effectiveness. Finally, researchers have devoted considerable efforts to innovation and change processes, and the role of teams within them. In our opinion, this structure follows a logic derived from organizational (and wider social, cultural and economic) contexts that are characteristic of both countries, despite local differences. Over the past decade Spain and Portugal have been immersed in a process of rapid economic and technological change, the pace of which has been faster than in neighboring countries, if not in absolute then in relative terms. This has obliged firms, organizations and institutions throughout the Peninsula to make a bigger

Figure 1. Main trends of Spanish and Portuguese research into work team effectiveness

technological and organizational leap than has been the case in other Western European countries or in North America. Moreover, it has been necessary to set up from scratch the various types of work groups and modes of team working discussed above. Interest in innovation processes is closely related, because the same organizations have been forced to undertake change processes, often starting from frankly anachronistic structures, in order to create dynamic and complex new organizational forms. Once again, the leap is greater here than in other more developed countries, where change processes tend to be more gradual and less traumatic. At the same time, Spanish and Portuguese firms have found it difficult to compete with more advanced foreign rivals, at least in the mass markets for consumer goods and services. One of the opportunities to compete is therefore to innovate and seek out new market niches. The combination of these two factors explains the interest in the role of work teams in innovation processes. Though practically all of Spanish and Portugal research in one way or another concerns the measurement of effectiveness and the impact of different variables, it is nonetheless possible to identify a specific focus on the criteria for and dimensions of effectiveness. In our opinion, this responds to the need for an accurate assessment of the advantages of work teams in the quest for effectiveness, ensuring the viability of organizations and goal attainment. Concern for effectiveness is, of course, shared by the majority of organizations the world over, but in the case of Spain and Portugal the need is greater, because of the pressure for local firms and organizations to achieve continuous productivity gains and compete in a European and global market. If this cannot be done, it is likely that multinational corporations in particular will continue to relocate plant and services to more competitive countries, albeit the causes of delocalization are more complex. Finally, let us consider intra- and inter-group conflicts in organizational contexts. As mentioned in the introduction, work teams and team working have become increasingly common, indeed essential, in organizational contexts, and this implies a mix of cooperation and conflict that depends, above all, on the type of task. In short: The study of conflict takes on a new relevance for our understanding of work groups, and new research is needed for us to understand its processes and results on new conditions and forms of work (Peiro´ et al., 1999, p. 48).

This important line of inquiry perhaps requires rather more explanation than the other three. We believe it is rooted in cultural and socio-historical factors. Both Spain and Portugal tend to be rated in cultural terms as intermediate countries along the individualism-collectivism dimension. Thus, the results obtained by Triandis (1994, 1995) place the countries of Southern Europe slightly closer to the individualistic end of the scale in relation to a hypothetical center. The research directed by Hofstede (2003, 1996) differs somewhat from Triandis, however, in that the two countries are situated rather further apart. Of the 53 countries and regions included in these studies, Spain was 20th, while Portugal was between the 33rd and 35th (Hofstede, 1996, p. 105), where 1 indicates maximum individualism and 53 maximum collectivism. These scores appear to reflect a slightly stronger tendency towards individualistic values and behavior in Spain than in Portugal, which might result in less frequent or natural cooperation and collaboration, at least in work contexts (social networking and

Work team effectiveness

203

JMP 20,3/4

204

solidarity in times of emergency or catastrophe may be more prevalent in other social milieus). At the same time, socio-historical factors (such as backwardness and isolation from other more economically developed countries) have allowed certain organizational structures based on the concept of the job and on personal responsibility, competence and merit to survive for longer in both countries. This has hindered gradual change towards more cooperative forms of work in which sharing is a key element, while the hurried implementation of new work systems (e.g. information and communication technologies) has caught many on the back foot, These workers need both teamwork training and time to adapt. Both factors may be behind the high levels of conflict in Spanish and Portuguese organizational contexts, whether affecting individual team members or rival teams. It is in this light that scholarly interest of scholars in the origins and solution of conflicts should be viewed. To sum up, Spanish and Portuguese research into work team effectiveness is intertwined with the main international trends and yet displays its own special characteristics, which inevitably refer to organizational contexts and wider political, social, economic, cultural and economic factors peculiar to both countries. Scholars have sought to contribute both to the progress of knowledge and to its transfer in a manner that is applicable precisely in the local contexts where it is most relevant. Returning to the models discussed in section 2, it is clear that Spanish and Portuguese research into work team effectiveness in organizational contexts closely follows the functional approach (Wittenbaum et al., 2004) or input-process-output model (Salas et al., 2004). Similarly, it reflects what Ilgen (1999) has called a displacement of research on teams towards contextual factors (in our case, analysis of technological factors and the contexts of change and innovation) and towards outcomes (interest in team effectiveness), although this does not mean that researchers have ignored the study of group processes (analysis of conflict and innovation). 4. Monograph issue This section provides an overview of the papers included in this monograph issue. 4.1 Objectives The aim of this issue is to reflect the consolidation and maturity of Portuguese and Spanish research into work teams in organizational contexts. We believe the field has come of age in the Iberian Peninsula after two decades of continuous activity. The dominant lines of research are demonstrably linked to the main thrust of scientific research in other European Union countries and in the US, while Portuguese and Spanish researchers collaborate freely with scholars from abroad. In this light, we may conclude that the field has developed its own identity in each country and internationally. 4.2 Trends and lines of research The papers selected for this issue provide a general idea of the most representative lines of research into work team effectiveness in organizational contexts in Portugal and Spain. Inquiry into conflicts in work teams is represented by two papers. Medina et al. (2005) evaluate the relationships between task conflict and relational conflict, and their influence on certain affective responses of team members such as satisfaction,

well-being and propensity to leave a job. The paper also addresses the mediating and moderating role of relational conflict, finding evidence for the existence of a link between the two conflict types and a relationship between task conflict and job satisfaction, and for the mediating role of relational conflict in well-being. Meanwhile, Passos and Caetano (2005) consider the effects of intra-group conflict in three dimensions (relational, task and process conflict), performance feedback and perceived efficacy of decision making processes related with team outcomes and the affective responses of team members. They find that the relationship between process conflict and outcomes is mediated by the perceived efficacy of decision making, while conflicts of this nature directly impact team satisfaction. Feedback is found directly to influence current team outcomes. The second line of research presented here concerns technological and temporal factors affecting the functioning and effectiveness of teams. Caballer et al. (2005) examine the moderating role of time pressure on the influence of the communication media and satisfaction with processes, satisfaction with outcomes, and the commitment of team members to solutions. The group media considered are face-to-face communication, videoconferencing and e-mail. The results obtained point to key effects of time pressure on affective responses and suggest that the interaction between the communication media utilized by groups and time pressure is important to predicting such responses. These results are analyzed in light of the task/technology fit model, while their theoretical implications may recommend a reformulation of the model in certain aspects. Meanwhile, Rico and Cohen (2005) explore the performance of virtual teams based on the fit between task characteristics and the type of communication medium employed by members. Their results show that virtual teams achieve higher levels of performance under conditions of low task interdependence with asynchronous communication, and high task interdependence with synchronous communication. In general, high performance in virtual teams is contingent upon the association between the nature of the task and the type of communication media employed. Finally, Pissarra and Jesuino (2005) analyze the processes by which ideas are generated through computer mediated communication, using “electronic brainstorming”. They find that member anonymity provides greater satisfaction and produces better ideas and enhanced conceptual diversity. Furthermore, technology may mediate the generation of ideas at the group level by structuring the emergence of new conceptual categories. The last two papers are concerned with the analysis of variables. Osca et al. (2005) consider the influence of organizational support for the implementation of team based systems through longitudinal design. Their results reveal that organizational support has a positive influence on satisfaction and member commitment to the new work system, and on team productivity appraised through objective measures. Meanwhile, Gil et al. (2005) investigate the impact of change-oriented leadership on team results. Their findings suggest that the team climate (especially in relation to innovation) represents the nexus mediating between change-oriented leadership and group outcomes. This relationship is reinforced by group potency. This sample of Portuguese and Spanish research includes each of the main lines of inquiry discussed in the preceding sections and highlights the links between our concerns and international research on work team effectiveness in organizational contexts.

Work team effectiveness

205

JMP 20,3/4

206

4.3 Outlook for the future As we have seen, Spanish and Portuguese research on work groups and teams has made considerable progress in recent decades, contributing both to theoretical development and practical applications. We believe that the study of organizations and work will continue to advance over the coming decade. The need to perfect the functioning of work teams poses an enormous new challenge for the world of work and organizations. Fresh theoretical models, empirical knowledge, intervention and change methodologies and technologies will therefore be required to improve the creation and development of work teams. Emerging work and organizational realities demand progress and innovation in the investigation of work groups and teams. Recent methodological and theoretical developments should provide a boost for basic research and work-group and team-based R&D. We refer here to the methodology for multilevel and cross-level analysis of key variables in group studies. These new methodologies will provide potent research strategies to address long-standing issues. Group studies are likely to benefit enormously from multilevel and cross-level analysis (Klein and Kozlowski, 2000). This technique makes it possible to address the constructs incorporated within theoretical models at different levels, permitting an exploration of relationships across levels. The aggregation strategies used to obtain group measures based on observations or information obtained at the individual level opens up a range of possibilities with regard to the processes contributing to the emergence of group constructs based on individual data and perceptions. Furthermore, dispersion theory suggests the need to consider the diversity of group variables and their theoretical significance. Meanwhile, considerable progress has been made with the measures obtained directly at the group level and the significance of relationships between the various individual measures. A good example of this is relational demographics. Thus, the significance of an individual datum may vary depending on the stance of other subjects within the same group. Studies combining individual characteristics as a gestalt that influences other group constructs and processes are highly relevant to research seeking to obtain a clearer understanding of group effectiveness and outcomes. The appropriate combination of diverse individual, group and contextual features, and the ideal level of diversity depending on the criteria employed (group effectiveness, member satisfaction, etc.) are key to clarifying strategies for the design, development and improvement of work teams. The dynamics of basic psycho-social processes are another area of group studies that will require further development in the future. This concerns inquiry into how shared perceptions, the affective climate and the judgments and beliefs generated by groups under certain conditions emerge and become consolidated through social cognition processes, because these experiences form the basis for cognitions, emotions, affective ties and collective beliefs, though this does not mean the reification of these phenomena. New, far-reaching approaches have also emerged in the study of outcomes. Insofar as effective working is something more than the successful completion of previously standardized and defined tasks, but also requires innovation, problem solving, role extension and extra-role behaviors, the issue of work team effectiveness suddenly widens and becomes complex. It is therefore important to expand the spectrum of outcomes considered.

Drawing the classic distinction between instrumental, and expressive and socio-emotional outcomes, the groundwork has been laid for the analysis of numerous phenomena making up groups’ “psycho-social production”. Thus, one of the classic themes in the area of performance and effectiveness is moral satisfaction, cohesion and team climate or culture. However, new issues have begun to emerge, including collective emotions and the processes they produce (e.g. emotional contagion), stress and collective burnout in work teams. An ever increasing body of evidence has in fact been amassed regarding the “collective reality” of certain phenomena, which conventional research has only been able to address at the individual level. Thus, new collective approaches to stress have made it possible to re-conceptualize the constructs involved in group terms. These approaches include the analysis of primary and secondary perceptions of the collective nature of the phenomenon within the work team or unit, and the study of the emotions and affective ties shared by members when the team is under stress. Groups frequently employ collective strategies to combat stress and, indeed, must do so if they are to resolve stressors effectively, while the consequences of team stress may also have collective components. In this light, the analysis of stress at the collective level requires us to question numerous implicit assumptions of the individualistic approach taken in traditional studies, which has inspired research for the last 50 years (Peiro´, 2001b). The study of leadership has also enjoyed considerable development, and future progress seems certain both here and in the related, though separate, field of management. Beyond functions and roles studies, which monopolized a good part of classical management research, interest nowadays centers on the study of the managerial skills that allow quick, flexible and adaptive responses to shifting conditions. In the field of leadership, new approaches and paradigms have enriched inquiry, linking up with new team scenarios and needs such as handling change, orientation to innovation, the management of virtual teams, and the management of diversity, all of which require new forms of management and leadership (delegation and empowerment, transformational leadership, knowledge management, etc.). The increasing integration of the new information and communication technologies in the world of work has had profound repercussions for the study of groups. In recent decades, the enormous spread of computer use and telecommunications has driven a wide range of telematic technologies. To name but a few, we may mention computer mediated technologies (e.g. e-mail, chats, forums and instant messaging), audio-conferencing and video-conferencing. Some not only facilitate but also structure communication depending on the design criteria employed to enhance effectiveness, usability, etc. Among these, we may mention decision support systems and other forms of “groupware”. One of the most direct and well-known effects of these technologies has been to lower the barriers and the time and spatial constraints upon group working. This has enormously increased asynchronous and non-location specific working. These transformations have been deep, and debate now centers ever less on virtual teams and groups and more on the degree of virtualization, suggesting that the phenomenon could be described in terms of a series of relevant dimensions. All of these technological changes have opened up a new and wide-ranging work group research scenarios. Traditional studies of groups and teams have, of course,

Work team effectiveness

207

JMP 20,3/4

208

assumed that work is carried on face to face. The new technologies, however, have changed this setting, affecting not only the composition and design of groups, but also processes, dynamics, evolution and outcomes. All of these transformations open up new fields for research that were hardly imaginable a few decades ago and remain to a great extent unexplored. Moreover, work teams mediated by the ICTs frequently discover that effective technologies, rules and construction and development procedures formulated to improve “conventional” work teams are no longer applicable because they fall short in situations mediated by ICTs and in virtual teams. This heralds an immense R&D field to revise existing and formulate new technologies for the creation and development of teams. The new technologies also offer new solutions for numerous services. One example that has seen significant development is open and distance learning. In “traditional” learning situations, teamwork was regarded as a fundamental strategy to develop the competencies in demand in the workplace. The new open and distance learning approach also stresses teamwork, but under virtual conditions. The analysis of these “reformulated” and new group processes in terms of the enabling technologies is a matter of considerable theoretical and practical interest. The development of psycho-social technologies permitting the integration and functioning of technical and social subsystems also represents a challenge of unquestionable interest and utility. A further area of enormous theoretical and practical interest concerns the multiple contextual factors of work teams, which are key contingencies for composition, design, structure, processes, evolutionary dynamics and development. Groups, and especially work teams, are created, act and develop in a range of organizational, business and socio-economic contexts. It is by no means easy to identify relevant taxonomies or dimensions that would better our understanding of these environments and their implications for the functioning of groups. Organizations may take a wide range of structural forms, while the phases and changes they undergo are increasingly intense and dynamic in a global environment where competition has become the basis for economic and social development. Conventional analyses that seek to establish bi-variate or multi-variate relations (treating certain variables jointly) have proved insufficient to capture the complexity of these environments and the dynamics of change. In this context, the theoretical models and frameworks underlying research have tended to take into account and incorporate a wide range of constructs and relationships, but empirical research lags far behind and has failed to verify such relations overall, remaining highly fragmented and incomplete. The development of epistemologies, models and methodologies that would permit scientific progress in the validation of theoretical proposals designed to advance our understanding of work groups and teams in more integrated terms is therefore a challenge of no small import. This list of challenges could of course be extended considerably, and we shall therefore end by mentioning the various temporal aspects of groups and teams. Groups are artificial phenomena, and are therefore part and parcel of the historical and temporal situation of human reality. The Spanish philosopher Zubiri insisted on the temporal and historical dimension of humankind, holding that collectivity was a basic element of our nature. It was in these terms that he propounded the concept of “historicity”. Groups are necessarily temporal and historical, and this forms a part of their “essence”. Nevertheless, the temporal nature of work groups has been only

partially investigated. Groups develop while they work. Consequently, all of their processes are of a dynamic nature, the formation, operation and evolution of which is largely unknown. Groups frequently work subject to time constraints or deadlines set, whether implicitly or explicitly, by team members or by external agents or audiences. These time constraints condition the life and work of the group. Thus, “opportunity” becomes a key concept in the operation and development of groups, and in the course of certain processes because outcomes may vary widely depending on the moment. Opportunity thus depends on numerous factors (e.g. the presence or absence of other processes). All of these complex realities are scarcely understood and require further research. Groups of course undergo processes of formation, growth, development, and ageing or decline, and many models have sought to describe stages or states in the evolution of work groups. However, further research on phenomena of this kind is called for in the face of new contexts, realities and demands, and the transformations we have discussed. Many of these issues require a longitudinal approach. For some decades now, there has been increasing insistence on the need for this kind of approach to the study of groups, but a long road remains. Longitudinal research involves one major complication. The number, reiteration and content of the observations to be made, the lag between observations and evaluation of their theoretical significance, control of relevant variables and their accessibility, among other factors, mean that longitudinal research is both costly and difficult. We believe that this is one of the fields in which new theoretical, methodological and practical advances could be made. This would not only allow raise new research questions, but would also clarify numerous existing issues. To sum up, work team effectiveness in organizational contexts is a vigorous field of research, which has a promising future. There is no shortage of issues, while new theoretical models and research techniques continue to emerge. Portuguese and Spanish researchers have contributed to this development to the full extent possible not only within their own cultural and national context, and in response to local social demand, but also taking into account general theory and issues. This balance between the general and the local means researchers in our two countries will continue to make contributions to the development of science and technology related with work teams and units in close collaboration with the international scientific community. References Alcover, C.M. and Gil, F. (1998), “La anticipacio´n del cambio como estrategia para afrontar la flexibilidad en la composicio´n de los equipos de trabajo” (“Change anticipation like strategy for flexibility coping in the composition of work teams”, II Congreso Iberoamericano de Psicologı´a, Madrid, 13-17 July, CD-ROM, ISBN 84-923717-0-6. Alcover, C.M. and Gil, F. (1999a), “The effects of member change and continuity on the productive efficiency of work teams”, Psychology in Spain, Vol. 3, pp. 88-97. Alcover, C.M. and Gil, F. (1999b), “Influencias de las nuevas formas de flexibilidad organizacional sobre los equipos de trabajo” (“Influences of new kinds of organizational flexibility on work teams”), Revista de Psicologı´a General y Aplicada, Vol. 52, pp. 177-201. Alcover, C.M., Martı´nez-I´n˜igo, D., Rodrı´guez-Mazo, F., Gil, F. and Caldero´n, N. (2000), “La investigacio´n sobre equipos de trabajo en Espan˜a en la segunda mitad del siglo XX”

Work team effectiveness

209

JMP 20,3/4

210

(“Spanish work teams research in the second half of 20th century”), in Agullo´, E., Remeseiro, C. and Ferna´ndez, J.A. (Eds), Psicologı´a del Trabajo, de las Organizaciones y de los Recursos Humanos, Vol. 86-102, Biblioteca Nueva, Madrid, pp. 86-102. Alonso, E. and Sa´nchez, J.C. (1998), “El compromiso organizacional de los grupos de trabajo en funcio´n de su cultura” (“Work groups’ organizational commitment depending on his culture”), II Congreso Iberoamericano de Psicologı´a, Madrid, 13-17 July, CD-ROM, ISBN 84-923717-0-6. Amaral, M.J. and Monteiro, M.B. (2002), “To be without being seen: computer-mediated communication and social identity management”, Small Group Research, Vol. 33, pp. 575-89. Anderson, N.R. and West, M.A. (1996), “The team climate inventory: the development of the TCI and its applications in teambuilding for innovativeness”, European Journal of Work and Organizational Psychology, Vol. 5, pp. 53-66. Aritzeta, A. (2001), “Efectos de los Equipos de Trabajo Autogestionados (EQTA) y del trabajo individual sobre caracterı´sticas grupales e individuales” (“Effects of self-managed work teams (SMWT) and individual work on group and individual characteristics”), Revista de Psicologı´a del Trabajo y de las Organizaciones, Vol. 17, pp. 197-218. Aritzeta, A. (2003), “Compromiso organizacional en los equipos de trabajo: un ana´lisis multinivel” (“Organizational commitment and work teams: a multilevel analysis”), Encuentros en Psicologı´a Social, Vol. 1 No. 2, pp. 114-19. Arro´spide, J.J., Martı´nez-Taboada, C. and Ayestara´n, S. (1995), “Estudio longitudinal de grupo: focalizado en la percepcio´n de jerarquı´a y el manejo del conflicto” (“Group longitudinal analysis: focus on hierarchy perception and conflict management”), in Sa´nchez, J.C. and Ulla´n, A.M. (Eds), Procesos Psicosociales Ba´sicos y Grupales, Eudema, Salamanca, pp. 531-40. Ayestara´n, S. (1993), “La innovacio´n a trave´s de los equipos de trabajo” (“Innovation through work teams”), in Munduate, L. and Baro´n, M. (Eds), Gestio´n de Recursos Humanos y Calidad de Vida Laboral, Eudema, Sevilla, pp. 69-72. Ayestara´n, S. (1998a), “La Psicologı´a y los grupos” (“Psychology and the groups”), Papeles del Psico´logo, Vol. 70, pp. 13-16. Ayestara´n, S. (1998b), “Cultura y manejo de conflictos en equipos de trabajo” (“Culture and conflict management in work teams”), II Congreso Iberoamericano de Psicologı´a, Madrid, 13-17 July, CD-ROM, ISBN 84-923717-0-6. Ayestara´n, S. (1999), “Formacio´n de equipos de trabajo, conductas de manejo de conflicto y cambio cultural en las organizaciones” (“Team building, conflict management behaviors and cultural change in organizations”), Revista de Psicologı´a General y Aplicada, Vol. 52, pp. 203-17. Ayestara´n, S. and Reoyo, A. (1998), “Culture, styles of conflict management and identification in work teams”, paper presented at the 11th Annual Conference of the International Association for Conflict Management, Madrid, 7-10 June. Bramel, D. and Friend, R. (1987), “The work group and its vicissitudes in social and industrial psychology”, Journal of Applied Behavioral Science, Vol. 23, pp. 233-53. Caballer, A., Gracia, F. and Peiro´, J.-M. (2005), “Affective responses to work process and outcomes in virtual teams: effects of communication media and time pressure”, Journal of Managerial Psychology, Vol. 20 Nos 3/4, pp. 245-60. Campion, M.A., Medsker, G.J. and Higgs, A.C. (1993), “Relations between work group characteristics and effectiveness: implications for designing effective work groups”, Personnel Psychology, Vol. 46, pp. 823-50.

Campion, M.A., Papper, E.M. and Medsker, G.J. (1996), “Relations between work team characteristics and effectiveness: a replication and extension”, Personnel Psychology, Vol. 49, pp. 429-52. Cannon-Bowers, J.A., Oser, R. and Flanagan, L. (1992), “Work teams in industry: a selected review and proposed framework”, in Sweezy, R.W. and Salas, E. (Eds), Teams: Their Training and Performance, Ablex Pub., Norwood, NJ, pp. 355-77. Canto, J.M. and Moral, F. (2003), “Comunicacio´n virtual, grupos y eficacia” (“Virtual communication, groups and efficacy”), Encuentros en Psicologı´a Social, Vol. 1 No. 4, pp. 109-12. Cohen, S.G. and Bailey, D.E. (1997), “What makes teams work? Group effectiveness research from the shopfloor to the executive suite”, Journal of Management, Vol. 23, pp. 239-90. Costa, M., Rumbo, A. and Pe´rez, M. (1997), “Cı´rculos de calidad para la aceptacio´n del cambio organizativo?” (“Quality circles for organizacional change acceptance?”), Cadernos de Psicologı´a, Vol. 17, pp. 41-5. Cruces, S.J., Cisneros, I.F.J., Munduate, L., Go´mez, T. and Dorado, M.A. (1999), “Factores culturales y socioemocionales en los procesos de negociacio´n: la importancia de la mentira en una muestra espan˜ola” (“Cultural and socio-emotional factors in negotiation processes: the importance of the lie in a Spanish sample”), Revista de Psicologı´a General y Aplicada, Vol. 52, pp. 255-68. Cunha, J.V. and Cunha, M.P. (2001a), “Round the clock: collaborative work in the international moulding industry”, International Journal of Innovation Management, Vol. 5, pp. 49-71. Cunha, J.V. and Cunha, M.P. (2001b), “Brave new (paradoxical) world: structure and improvisation in virtual teams”, Strategic Change, Vol. 10, pp. 337-47. Cunha, M.P. and Cunha, J.V. (2001), “Managing improvisation in cross-cultural virtual teams”, International Journal of Cross-Cultural Management, Vol. 1, pp. 187-208. Curral, L. and Chambel, M.J. (1999), “Processos de grupo em equipas de inovac¸a˜o” (“Group processes in innovation teams”), Psicologı´a, Vol. 13, pp. 163-92. Curral, L., Forrester, R.H., Dawson, J.F. and West, M.A. (2001), “It’s what you do and the way that you do it: team task, team size, and innovation-related group processes”, European Journal of Work and Organizational Psychology, Vol. 10, pp. 187-204. Devine, D.J., Clayton, L.D., Philips, J.L., Dunford, B.B. and Melner, S.B. (1999), “Teams in organizations: prevalence, characteristics, and effectiveness”, Small Group Research, Vol. 30, pp. 678-711. Dorado, M.A., Medina, F.J., Munduate, L., Cisneros, I.F.J. and Euwema, M. (2002), “Computer-mediated negotiation of an escalated conflict”, Small Group Research, Vol. 33, pp. 509-24. Gibson, C.B., Randel, A.E. and Earley, P.C. (2000), “Understanding group efficacy: an empirical test of multiple assessment methods”, Group and Organization Management, Vol. 25, pp. 67-97. Gil, F. and Alcover, C.M. (2002), “Small group research in Europe: contributions to the field from Spanish social psychology 1955-2000”, European Psychologist, Vol. 7, pp. 265-74. Gil, F., Rico, R., Alcover, C.M. and Barrasa, A. (2005), “Change-oriented leadership, satisfaction and effectiveness in work groups: effects of team climate and group potency”, Journal of Managerial Psychology, Vol. 20 Nos 3/4, pp. 312-28. Gil, F., Rico, R., Ferna´ndez Rı´os, M., Barrasa, A., Alcover, C.M. and Rodrı´guez Mazo, F. (2003), “Liderazgo en equipos virtuales” (“Leadership in virtual teams”), Encuentros en Psicologı´a Social, Vol. 1 No. 5, pp. 101-6.

Work team effectiveness

211

JMP 20,3/4

212

Gonza´lez-Roma´, V. and Peiro´, J.-M. (1999), “Clima en las organizaciones laborales y en los equipos de trabajo” (“Climate in work organizations and work teams”), Revista de Psicologı´a General y Aplicada, Vol. 52, pp. 269-85. Gonza´lez-Roma´, V., Peiro´, J.-M. and Prieto, F. (1992), “Un ana´lisis de la participacio´n en los cı´rculos de calidad” (“An analysis of participation in quality circles”), Revista de Psicologı´a del Trabajo y de las Organizaciones, Vol. 10, pp. 189-97. Gonza´lez-Roma´, V., Peiro´, J.-M. and Tordera, N. (2002), “An examination of the antecedents and moderador influences of climate strength”, Journal of Applied Psychology, Vol. 87, pp. 465-73. Gonza´lez-Roma´, V., Peiro´, J.-M., Lloret, S. and Zornoza, A. (1999), “The validity of collective climates”, Journal of Occupational and Organizational Psychology, Vol. 72, pp. 25-40. Gonza´lez-Roma´, V., Peiro´, J.-M., Subirats, M. and Man˜as, M.A. (2000), “The validity of affective work team climates”, in Vartiainen, M. and Avallone, F. (Eds), Innovative Theories, Tools, and Practices in Work and Organizational Psychology, Hogrefe and Huber Publishers, Kirkland, WA, pp. 97-109. Gonza´lez-Roma´, V., Can˜ero, J., Melia´, J.L., Peiro´, J.-M. and Prieto, F. (1987), “Quality circles and its influence on management perception, organizational climate, work performance and absenteeism in an organizational context”, in de Witte, K. (Ed.), The Challenge of Technological Change for Work and Organization: Tools and Strategies for the 1990s, Proceedings of the 3rd Wets European Congress on the Psychology of Work and Organization, Vol. II, Amberes, 13-15 April, pp. 651-4. Gonza´lez-Roma´, V., Ramos, J., Peiro´, J.-M., Rodrı´guez, I. and Mun˜oz, P. (1995), “Formacio´n del clima de los equipos de trabajo e interaccio´n social” (“Formation of work team climate and social interaction”), in Zurriaga, R. and Sancerni, M.D. (Eds), Experiencias Laborales en Organizaciones de Trabajo, Nau Llibres, Valencia, pp. 65-78. Gracia, F., Caballer, A. and Peiro´, J.-M. (2002), “Efectos de la presio´n temporal sobre la cohesio´n grupal en diferentes tipos de tareas y en diferentes canales de comunicacio´n” (“Effects of time pressure on group cohesiveness in different task types and communication media”), Psicothema, Vol. 14, pp. 434-9. Guzzo, R.A. and Dickson, M.W. (1996), “Teams in organizations: recent research on performance and effectiveness”, Annual Review of Psychology, Vol. 47, pp. 307-28. Hackman, J.R. and Morris, C.G. (1975), “Group tasks, group interaction process, and group performance: a review and proposed integration”, in Berkowitz, L. (Ed.), Advances in Experimental Social Psychology, Vol. 8, Academic Press, New York, NY, pp. 45-99. Harris, T.C. and Barnes-Farrell, J.L. (1997), “Components of teamwork: impact of evaluations of contributions to work team effectiveness”, Journal of Applied Social Psychology, Vol. 27, pp. 1694-715. Hofstede, G. (1996), Culturas y Organizaciones. El Software Mental. La Cooperacio´n Internacional y su Importancia Para la Supervivencia (Cultures and Organizations. Software of the Mind. Intercultural Cooperation and its Importance for Survival), Alianza Editorial, Madrid. Hofstede, G. (2003), Culture’s Consequences: Comparing Values, Behaviors, Institutions and Organizations across Nations, 2nd ed., Sage, Thousand Oaks, CA. Hyatt, D.E. and Ruddy, T.M. (1997), “An examination of the relationship between work group characteristics and performance: once more into the breach”, Personnel Psychology, Vol. 50, pp. 553-85. Ilgen, D.R. (1999), “Teams embedded in organizations: some implications”, American Psychologist, Vol. 54, pp. 129-39.

Ilgen, D.R., Major, D.A., Hollenbeck, J.R. and Sego, D.J. (1993), “Team research in the 1990s”, in Chemers, M.M. and Ayman, R. (eds), Leadership Theory and Research: Perspectives and Directions, Academic Press, San Diego, CA, pp. 245-70.

Work team effectiveness

Jesuino, J.C. (2002), “Introduction: European views of computer-mediated communication”, Small Group Research, Vol. 33, pp. 475-80. Jordan, M.H., Feild, H.S. and Armenakis, A.A. (2002), “The relationship of group process variables and team performance”, Small Group Research, Vol. 33, pp. 121-50. Klein, K.J. and Kozlowski, S.W.J. (Eds) (2000), Multilevel Theory, Research, and Methods in Organizations: Foundations, Extensions, and New Directions, Jossey-Bass, San Francisco, CA. Kozlowski, S.W.J. and Bell, B.S. (2003), “Work groups and teams in organizations”, in Borman, W.C., Ilgen, D.R. and Klimoski, R.J. (Eds), Industrial and Organizational Psychology, Vol. 12, John Wiley & Sons, Hoboken, NJ, pp. 333-75. Levi, D. and Slem, C. (1995), “Team work in research and development organizations: the characteristics of successful teams”, International Journal of Industrial Ergonomics, Vol. 16, pp. 29-42. Levine, J.M. and Moreland, R.L. (1990), “Progress in small group research”, Annual Review of Psychology, Vol. 41, pp. 585-634. Lourenc¸o, P.R. (2000), “Lideranc¸a e eficacia: uma relac¸a˜o revisitada” (“Leadership and efficacy: a revisited relation”), Revista Psicolo´gica, Vol. 23, pp. 119-30. Lourenc¸o, P.R. and Gomes, A.D. (2003), “Da pluralidade a` bidimensionalidade da eficacia dos grupos/equipas de trabalho” (“From plurality to bidimensionality of work groups/teams efficacy”), Revista Psicolo´gica, Vol. 33, pp. 7-32. Lourenc¸o, P.R., Miguez, J., Gomes, A.D. and Carvalho, C. (2004), “Efica´cia grupal: ana´lise e discussa˜o de um modelo multidimensional” (“Group efficacy: an analysis and discussion of a multidimensionality model”), Psicolo´gica, Extra-Se´rie, pp. 611-21. Lourenc¸o, P.R., Miguez, J., Gomes, A.D. and Freire, P. (2000), “Equipas de trabalho: efica´cia ou eficacias?” (“Work teams: efficacy or efficacies?”), in Gomes, A.D., Caetano, A., Keating, J. and Cunha, M.P. (Eds), Organizac¸o˜es em transic¸a˜o. Contributos da Psicologia do Trabalho e das Organizac¸o˜es, Imprensa da Universidade, Coimbra. McGrath, J.E. (1964), Social Psychology: A Brief Introduction, Holt, Rinehart & Winston, New York, NY. McGrath, J.E., Arrow, H. and Berdahl, J.L. (2000), “The study of groups: past, present, and future”, Personality and Social Psychology Review, Vol. 4, pp. 95-105. Man˜as, M.A., Gonza´lez-Roma´, V. and Peiro´, J.-M. (1998), “Clima de los equipos de trabajo y satisfaccio´n laboral: un estudio longitudinal” (“Work team climate and work satisfaction: a longitudinal study”), in de Diego, R. and Valdivieso, J.D. (Eds), Psicologı´a del Trabajo. Nuevos Conceptos, Controversias y Aplicaciones, Pira´mide, Madrid, pp. 43-9. Man˜as, M.A., Gonza´lez-Roma´, V. and Peiro´, J.-M. (1999), El Clima de los Equipos de Trabajo: Determinantes y Consecuencias (Work Team Climate: Determinants and Consequences), Servicio de Publicaciones de la Universidad de Almerı´a/Instituto de Estudios Almerienses, Almerı´a. Martı´nez, I. and Mejı´as, R.J. (2003), “Efectos del anonimato y el ge´nero sobre el nivel de consenso, cohesio´n y satisfaccio´n en grupos con apoyo de sistemas informa´ticos” (“Effects of anonymity and gender on consensus, cohesion and satisfaction in group support systems”), Revista de Psicologı´a Social, Vol. 18, pp. 107-20.

213

JMP 20,3/4

Martı´nez, I., Cifre, E., Llorens, S. and Salanova, M. (2002), “Efectos de la tecnologı´a asistida por ordenador en el bienestar psicolo´gico afectivo” (“Effects of computer-aided technology on affective psychological wellbeing”), Psicothema, Vol. 14, pp. 118-23. Medina, F.J., Munduate, L., Dorado, M.A., Martı´nez, I. and Cisneros, I.F.J. (2005), “Types of intragroup conflict and affective reactions”, Journal of Managerial Psychology, Vol. 20 Nos 3/4, pp. 219-30.

214

Medina, F.J., Munduate, L., Martı´nez, I., Dorado, M.A. and Man˜as, M.A. (2003), “Una perspectiva contingente en el estudio de la efectividad de los tipos de conflicto organizacional. El rol moderador del clima de los equipos de trabajo” (“A contingent perspective in the study of organizational conflict types effectiveness. The moderational role of work group climate”), Encuentros en Psicologı´a Social, Vol. 1 No. 3, pp. 341-6. Medina, F.J., Munduate, L., Martı´nez, I., Dorado, M.A. and Man˜as, M.A. (2004), “Efectos positivos de la activacio´n del conflicto de tarea sobre el clima de los equipos de trabajo” (“Positive effects of task-conflict stimulation on work-team climate”), Revista de Psicologı´a Social, Vol. 19, pp. 3-15. Monteiro, M.B. (2003), “Dealing with intergroup conflicts in organizations: a challenge to the right of diversity”, Portuguese Journal of Social Science, Vol. 2, pp. 79-88. Moral, F., Canto, J.M. and Go´mez Jacinto, L. (2003), “Comunicacio´n mediada por ordenador: influencia minoritaria endogrupal” (“Computer-mediated communication: In-group minority influence”), Encuentros en Psicologı´a Social, Vol. 1 No. 1, pp. 153-7. Munduate, L. and Dorado, M.A. (1999), “El conflicto en los grupos de trabajo” (“Conflict in work groups”), in Morales, J.F. and Yubero, S. (Eds), El Conflicto en los Grupos de Trabajo, Universidad de Castilla-La Mancha, Cuenca, pp. 101-16. Orengo, V., Ripoll, P., Gosa´lvez, I. and Solanes, A. (1998), “Ana´lisis de la interaccio´n grupal a partir de una adaptacio´n de la te´cnica SYMLOG en funcio´n del canal de comunicacio´n y del tipo de tarea” (“Analysis of group interaction through an adaptation of the SYMLOG technique as a function of channel communication and task type”), Revista de Psicologı´a Social, Vol. 13, pp. 241-50. Orengo, V., Zornoza, A., Acı´n, C., Prieto, F. and Peiro´, J.-M. (1996), “Ana´lisis de la interaccio´n grupal a trave´s de medidas de observacio´n en comunicacio´n mediada” (“Group interaction analysis in mediated communication through observational measures”), Revista de Psicologı´a Social, Vol. 11, pp. 151-62. Orengo, V., Zornoza, A.M., Prieto, F. and Peiro´, J.-M. (2000), “The influence of familiarity among group members, group atmosphere and assertiveness on uninhibited behavior through three different communication media”, Computers in Human Behavior, Vol. 16, pp. 141-59. Osca, A., Alonso, E. and Palacı´, F. (1994), “La cultura organizacional en los Equipos de Atencio´n Primaria (EAPS) y de los Centros de Atencio´n a Drogodependientes (CADS)” (“Organizacional culture in primary health-care teams and in drugs dependence centers”), in Zurriaga, R. and Sancerni, M.D. (Eds), Experiencias Laborales en Organizaciones de Trabajo, Nau Llibres, Valencia, pp. 49-64. Osca, A., Urien, B., Gonza´lez-Camino, G. and Martı´nez-Pe´rez, M.D. (2005), “Organizational support and performance in teamwork systems: a longitudinal study”, Journal of Managerial Psychology, Vol. 20 Nos 3/4, pp. 292-311. Passos, A.M. and Caetano, A. (n.d), “Tipos de conflito intragrupal: um modelo tridimensional” (“Intergroup conflict types: a tridimensional model”), in Vala, J. and Garrido, M. (Eds), Percursos de investigac¸a˜o em psicologia social e organizacional (in press), Colibri, Lisboa.

Passos, A.M. and Caetano, A. (2005), “Exploring the effects of intragroup conflict and past performance feedback on team effectiveness”, Journal of Managerial Psychology, Vol. 20 Nos 3/4, pp. 231-44. Peiro´, J.-M. (2001a), “Perspectiva histo´rica y desarrollos recientes de la psicologı´a del trabajo y de las organizaciones en Espan˜a” (“Historical overview and recent developments of work and organizational psychology in Spain”), Revista de Psicologı´a del Trabajo y de las Organizaciones, Vol. 17, pp. 255-71. Peiro´, J.-M. (2001b), “Stressed teams in organizations: a multilevel approach to the study of stress in work units”, in Pryce, J., Weilkert, C. and Torkelson, E. (Eds), Occupational Health Psychology: Europe 2001, European Academy of Occupational Health, Nottingham, pp. 9-13. Peiro´, J.-M. (2002), “Perspectiva histo´rica y tendencias actuales de la psicologı´a del trabajo, las organizaciones y del personal en Espan˜a. Un modelo para su ana´lisis” (“Historic perspective and present tendencies of the work, organizational and personnel psychology in Spain. A model for its analysis”), Revista de Historia de la Psicologı´a, Vol. 23, pp. 223-47. Peiro´, J.-M. and Munduate, L. (1999), “Psicologı´a del trabajo y de las organizaciones en Espan˜a en la de´cada de los noventa” (“Work and organizational psychology in Spain in the 1990s”), Revista de Psicologı´a General y Aplicada, Vol. 52, pp. 371-428. Peiro´, J.-M. and Munduate, L. (2001), “Research developments on work and organizational psychology in Spain: an overview”, Revista de Psicologı´a Social Aplicada, Vol. 11, pp. 31-48. Peiro´, J.-M., Gonza´lez-Roma´, V. and Ramos, J. (1992), “The influence of work-team climate on role stress, tension, satisfaction, and leadership perceptions”, European Review of Applied Psychology/Revue Europe´enne de Psychologie Applique´e, Vol. 42, pp. 49-56. Peiro´, J.-M., Prieto, F., Zornoza, A. and Ripoll, P. (1999), “El conflicto y su gestio´n en el trabajo en equipo mediado por nuevas tecnologı´as de la informacio´n” (“Conflict and its management in team work mediated for new information technologies”), in Morales, J.F. and Yubero, S. (Eds), El Grupo y Sus Conflictos, Universidad de Castilla-La Mancha, Cuenca, pp. 47-99. Pinto, A.C. and Teixeira, N.S. (Eds) (2002), Southern Europe and the Making of the European Union, Columbia University Press, New York, NY. Pissarra, J. and Jesuino, J.C. (2005), “Idea generation through computer-mediated communication: the effects of anonymity”, Journal of Managerial Psychology, Vol. 20 Nos 3/4, pp. 275-91. Reoyo, A.M. (1998), “Cultura, manejo del conflicto e identificacio´n grupal en los grupos de trabajo” (“Culture, conflict management and identification in work teams”), Revista de Psicologı´a Social, Vol. 13, pp. 261-8. Reoyo, A.M. and Lo´pez Canas, R. (2003), “Efectividad de la intervencio´n psicosocial en los equipos de trabajo: comparacio´n de muestras universitarias y de empresa” (“Psychosocial intervention effectiveness in work teams: comparison between university students and company workers samples”), Encuentros en Psicologı´a Social, Vol. 1 No. 4, pp. 113-18. Rico, R. and Cohen, S. (2005), “Effects of task interdependence and type of communication on performance in virtual teams”, Journal of Managerial Psychology, Vol. 20 Nos. 3/4, pp. 261-74. Ripoll, M.P., Gonza´lez-Navarro, P., Zornoza, A. and Orengo, V. (2004), “La influencia de los procesos de interaccio´n grupal y el medio de comunicacio´n sobre la eficacia de los grupos de trabajo” (“Influences of group interaction processes and communication media on work group effectiveness”), Revista Latinoamericana de Psicologı´a, Vol. 36 No. 2, pp. 195-208. Ripoll, M.P., Subirats, M., Torres, M.J. and Marzo, J.C. (1998), “Nuevas tecnologı´as telema´ticas y trabajo en grupo: un estudio de las diferencias en los resultados grupales en funcio´n del canal de comunicacio´n utilizado y de la experiencia adquirida” (“New telematic

Work team effectiveness

215

JMP 20,3/4

216

technologies and work team: a study of group outcomes differences depending on communication channel and gained experience”), in de Diego, R. and Valdivieso, J.D. (Eds), Psicologı´a del Trabajo. Nuevos Conceptos, Controversias y Aplicaciones, Pira´mide, Madrid, pp. 159-71. Salanova, M., Llorens, S., Cifre, E., Martı´nez, I. and Schaufeli, W.B. (2003), “Perceived collective efficacy, subjective wellbeing and task performance among electronic work groups: an experimental study”, Small Group Research, Vol. 34, pp. 43-73. Salas, E., Cannon-Bowers, J.A. and Johnston, J.H. (1997), “How can you turn a team of experts into an expert team? Emerging training strategies”, in Zsambok, C.E. and Klein, G. (Eds), Naturalistic Decision Making, Erlbaum and Associates, Mahwah, NJ, pp. 359-70. Salas, E., Stagl, K.C. and Burke, C.S. (2004), “25 years of team effectiveness in organizations: research themes and emerging needs”, in Cooper, C.L. and Robertson, I.T. (Eds), International Review of Industrial and Organizational Psychology, Vol. 19, John Wiley & Sons, Chichester, pp. 47-91. Sa´nchez, J.C. and Alonso, E. (2003), “La cultura de los equipos de trabajo y su relacio´n con la cohesio´n e identificacio´n grupal” (“The culture of work teams and their relationship with group cohesion and identification”), Encuentros en Psicologı´a Social, Vol. 1 No. 4, pp. 22-5. Sawyer, J.E., Latham, W.R., Pritchard, R.D. and Bennett, W.R. (1999), “Analysis of work group productivity in an applied setting: application of a time series panel design”, Personnel Psychology, Vol. 52, pp. 927-67. Segurado, A., Mier, L. and Ferna´ndez, M.A. (2004), “El estudio de la eficacia en los equipos de trabajo: avances metodolo´gicos” (“Effectiveness teamwork: methodological advances”), Encuentros en Psicologı´a Social, Vol. 2 No. 1, pp. 157-62. Simpson, J.A. and Wood, W. (1992), “Introduction: where is the group in social psychology? An historical overview”, in Worchel, S. and Simpson, J.A. (Eds), Group Process and Productivity, Sage, Newbury Park, CA, pp. 1-12. Steiner, I.V. (1986), “Paradigms and groups”, in Berkowitz, L. (Ed.), Advances in Experimental Social Psychology, Vol. 19, Academic Press, New York, NY, pp. 251-89. Sundstrom, E., McIntyre, M., Halfhill, T. and Richards, H. (2000), “Work groups: from the Hawthorne studies to work teams of the 1990s and beyond”, Group Dynamics: Theory, Research, and Practice, Vol. 4, pp. 44-67. Tannenbaum, S.I., Beard, R.L. and Salas, E. (1992), “Team building and its influence on team effectiveness: an examination of conceptual and empirical developments”, in Kelley, K. (Ed.), Issues, Theory, and Research in Industrial/Organizational Psychology, Elsevier Science, Amsterdam, pp. 117-53. Telo, A.J. and de la Torre, H. (2003), Portugal y Espan˜a en los Sistemas Internacionales Contempora´neos (Portugal and Spain in Comtemporary International Systems), Gabinete de Iniciativas Transfronterizas de la Junta de Andalucı´a, Me´rida. Tjosvold, D., West, M.A. and Smith, K.G. (2003), “Teamwork and cooperation: fundamentals of organizational effectiveness”, in West, M.A., Tjosvold, D. and Smith, K.G. (Eds), International Handbook of Organizational Teamwork and Cooperative Working, John Wiley & Sons, Chichester, pp. 3-8. Triandis, H.C. (1994), Culture and Social Behavior, McGraw-Hill, New York, NY. Triandis, H.C. (1995), Individualism and Collectivism, Westview Press, Boulder, CO. Unsworth, K.L. and West, M.A. (2000), “Teams: the challenges of cooperative work”, in Chmiel, N. (Ed.), Introduction to Work and Organizational Psychology: A European Perspective, Blackwell, Oxford, pp. 327-46.

West, M.A. (1996), “Preface: introducing work group psychology”, in West, M.A. (Ed.), Handbook of Work Group Psychology, John Wiley & Sons, Chichester, pp. xxv-xxxiii. West, M.A. (2001), “The human team: basic motivations and innovations”, in Anderson, N., Ones, D.S., Sinangil, H. and Viswesvaran, C. (Eds), Handbook of Industrial, Work and Organizational Psychology, Vol. 2, Sage, London, pp. 270-88. West, M.A., Borrill, C.S. and Unsworth, K.L. (1998), “Team effectiveness in organizations”, in Cooper, C.L. and Robertson, I.T. (Eds), International Review of Industrial and Organizational Psychology, Vol. 13, John Wiley & Sons, Chichester, pp. 1-48. Whiteoak, J.W., Chalip, L. and Hort, L.K. (2004), “Assessing group efficacy: comparing three methods of measurement”, Small Group Research, Vol. 35, pp. 158-73. Wittenbaum, G.M., Hollingshead, A.B., Paulus, P.B., Hirokawa, R.Y., Ancona, D.H., Peterson, R.S., Jehn, K.A. and Yoon, K. (2004), “The functional perspective as a lens for understanding groups”, Small Group Research, Vol. 35, pp. 17-43. Wood, R.E. and Tabernero, C. (2000), “Concepcio´n de capacidad como determinante de autoeficacia y eficacia grupal percibida” (“Conceptions of ability as determinants of self-efficacy and group efficacy”), Apuntes de Psicologı´a, Vol. 18, pp. 39-56. Zalesny, M.D., Salas, E. and Prince, C. (1995), “Conceptual and measurement issues in coordination: implications for team behavior and performance”, in Ferris, G.R. (Ed.), Research in Personnel and Human Resources Management, Vol. 13, JAI Press, Greenwich, CT, pp. 81-115. Zornoza, A., Ripoll, P. and Peiro´, J.M. (2002b), “Conflict management in groups that work in two different communication contexts: face-to-face and computer-mediated communication”, Small Group Research, Vol. 33, pp. 481-508. Zornoza, A., Orengo, V., Gosa´lvez, I. and Gonza´lez, P. (2002a), “Teamwork in different communication contexts: a longitudinal study”, Psychology in Spain, Vol. 6, pp. 41-55. Zornoza, A., Prieto, F., Martı´, C. and Peiro´, J.M. (1993), “Group productivity and telematic communication”, European Work and Organizational Psychologist, Vol. 3, pp. 117-27. Zornoza, A., Marı´n, F., Ramos, J., Prieto, F. and Martı´, C. (1992), “Medio de comunicacio´n y rendimiento en grupo: influencia de las actitudes y experiencias con nuevas tecnologı´as” (“Communication media and group performance: influences of attitudes and experience with new technologies”), Revista de Psicologı´a Social Aplicada, Vol. 2, pp. 73-88. (Francisco Gil is Professor of Group and Organizational Psychology at the Complutense University of Madrid (Spain). His current research interests include work teams, leadership and organizational learning. He has written several books and academic papers about these topics (Introduction to Group Psychology, Introduction to Organizational Psychology, Group Techniques in Organizational Contexts), he has coordinated international publications (Knowledge Management) and congress (Competencies Management; Coaching and Mentoring; Leadership), and he has directed and taken part in international research, as UIM projects – about formation and modernization of local administration in Ibero-American countries – and Globe Project (Global Leadership and Organizational Behavior Effectiveness). E-mail: [email protected]) (Carlos Marı´a Alcover is a Tenured Professor of Work and Organizational Psychology at Social Psychology Area of the University Rey Juan Carlos, Madrid, Spain. He earned his PhD in Social Psychology from the University Complutense of Madrid. His current research is focused on work groups – specifically on teamwork under temporal, functional, and technological flexible conditions – psychological contracts in contemporary employment relations, emotional labor in service workers, and psychological and social consequences of the early retirement. He

Work team effectiveness

217

JMP 20,3/4

218

has published a number of articles and books within these fields. His latest books written in collaboration with other authors are Group Techniques in Organizational Contexts and Introduction to Work Psychology, both published in 2004. He coordinates the Research Group “In-Psitro” (research in work and organizational psychology) at his University. E-mail: [email protected]) (Jose-M. Peiro´ is Professor of Work and Organizational Psychology at the University of Valencia (Spain) and senior researcher of the Economic Research Institute in Valencia (IVIE). He is also President elect of the International Association of Applied Psychology. Division 1 of Work and Organizational Psychology. He is associate editor of the European Journal of Work and Organizational Psychology and has been president of the European Association of Work and Organizational Psychology (1997-1999). His work focuses on stress, climate, team work at work and customer satisfaction and quality of services. E-mail: [email protected])

The Emerald Research Register for this journal is available at www.emeraldinsight.com/researchregister

The current issue and full text archive of this journal is available at www.emeraldinsight.com/0268-3946.htm

Types of intragroup conflict and affective reactions

Types of intragroup conflict

Francisco J. Medina, Lourdes Munduate and Miguel A. Dorado Department of Social Psychology, University of Seville, Seville, Spain

219

Ine´s Martı´nez Department of Psychology, University Pablo de Olavide, Seville, Spain, and

Jose´ M. Guerra Department of Social Psychology, University of Seville, Seville, Spain Abstract Purpose – Seeks to evaluate the link between task and relationship conflict, and their influence on some employees’ affective reactions such as satisfaction, wellbeing, and propensity to leave a job; and to analyse the mediated and moderated role of relationship conflict. Design/methodology/approach – The study involved 169 employees from six service organizations (hotels) in Andalusia (Spain). A questionnaire was used containing different measures: task and relationship conflict, wellbeing, job satisfaction, and propensity to leave the job. Findings – The two types of conflict have different consequences. Data show that relationship conflict is negatively associated with affective reactions, while task conflict does not relate directly to affective reactions in a predictable way; relationship conflict has a positive influence on the desire to leave the current job, while task conflict does not affect it negatively; the interactive effect of relationships and task conflict shows that this interaction contributes substantially to predict the propensity to leave the current job; and relationship conflict mediates in the link between task conflict and affective reactions. Research limitations/implications – A high level of task conflict may backfire by boosting relationship conflict as well, thus having a negative effect on affective reactions. Thus some conclusions can be drawn with a view to improving conflict management in teams. First an attempt must be made to understand the type of conflict that is taking place. Second, managers should encourage open discussion of task-related issues. Third, special attention should be paid to the level of each conflict because of its interactive effects on some affective outcomes. Thus, in spite of the generally beneficial effects associated with task conflict, the intensification of task-related conflict may backfire when interacting with dysfunctional affective-dissent. Originality/value – Serves too analyze the mediated and moderated role of relationship conflict and to test the role of types of conflict on affective reactions such as wellbeing and propensity to leave the job. Keywords Conflict, Organizational conflict, Job satisfaction, Spain Paper type Research paper

One of the most outstanding aspects of conflict is that it is practically intrinsic to the life and dynamics of teams. Conflict is present in interpersonal relations (Pruitt and Carnevale, 1993), in intragroup and intergroup relations (Jehn, 1995), in strategic decision-making (Amason, 1996), and other organizational episodes. As many authors The authors would like to thank Miguel A Man˜as for facilitating access to the sample. The authors are especially grateful for the valuable feedback provided by Carsten De Dreu. This research was sponsored by the Ministerio de Educacio´n y Ciencia (DGES), grant number: SEC-2000-0531.

Journal of Managerial Psychology Vol. 20 No. 3/4, 2005 pp. 219-230 q Emerald Group Publishing Limited 0268-3946 DOI 10.1108/02683940510589019

JMP 20,3/4

220

have pointed out (De Dreu and Van de Vliert, 1997; Pondy, 1967) conflict is a phenomenon that may give rise to both beneficial and dysfunctional effects on individuals, groups and organizations. Early group theorists have focuses on the negative consequences of conflict for teams. Conflict difficulties communications between individuals, breaks personal and professional relationships, and reduces effectiveness, because it produces tension and distracts team members from performing the task (Hackman and Morris, 1975; Wall and Callister, 1995). Thus, it is no surprise that today’s managers and employees still overwhelmingly view conflict as negative and something to be avoided or resolved as soon as possible (Stone, 1995). Indeed, growing evidence suggests that conflict may be beneficial to team performance. Suppressing conflict could reduce creativity, innovation, performance, quality of decisions, and communication between group’s members (see De Dreu and Van de Vliert, 1997). Research by Jehn (1994, 1995), Amason (1996) or De Dreu and Weingart (2003) provides evidence that this double-edged effect is attributable to different dimensions of conflict. Research has shown conflict to be multidimensional (e.g. Amason, 1996; Cosier and Schwenck, 1990; Jehn, 1995; Van de Vliert and De Dreu, 1994). Thus, it is possible for one dimension of conflict to enhance effectiveness whereas another hinders consensus and commitment between group members. Jehn (1995) distinguished between two kinds of intragroup conflict: task conflict and relationship conflict. Task conflict is a perception of disagreement among group members or individuals about the content of their decisions, and involves differences in viewpoints, ideas and opinions. Examples of task conflict are conflicts about the distribution of resources, about procedures or guidelines, and about the interpretation of facts. Relationship conflict is a perception of interpersonal incompatibility, and includes annoyance and animosity among individuals. Examples of relationship conflict are disagreements about values, personal or family norms, or about personal taste. The two types of intragroup conflict have different personal and organizational consequences. Several studies have investigated the relationships between these types of conflict and several outcomes – such as satisfaction, tension or group commitment. Relationship conflict is negatively associated with employees’ affective reactions such as satisfaction (for a review, see De Dreu and Van Viannen, 2001) and climate, and it reduces team effectiveness (Jehn, 1997). In contrast, task conflict appears to be positively related to the quality of ideas and innovation (West and Anderson, 1996), the increase of constructive debate (Jehn et al., 1999), the affective acceptance of group decisions (Amason, 1996), and the prevention of groupthink (Turner and Pratkanis, 1994). These conclusions about the positive function of task conflict and the negative function of relationship conflict are based on research that only examined how one type of conflict affects team performance regardless of the other type (e.g. Amason, 1996; Jehn, 1994, 1995). Consistent with this perspective, scholars tend to recommend management teams to stimulate task conflict and mitigate relationship conflict during team decision-making. These recommendations are, however, problematic because both types of conflict are related. Almost all studies – with the exception of Jehn (1995) – that measured task and relationship conflict in groups, have shown positive correlations between the two types of conflict (Amason, 1996; De Dreu, 1997; Friedman et al., 2000; Janssen et al., 1999; Jehn, 1995; Jehn and Mannix, 2001; Jehn and Chatman,

2000; Pelled et al., 1999). Thus, it may be extremely difficult for teams to effectively embrace one type of conflict, while simultaneously resisting the other. Amason (1996) pointed out that cognitive criticism might easily be interpreted as a personal disapproval or a strategy to enhance one’s own position at the expense of someone else’s. Baron (1990) showed that a critical evaluation produced negative affective reactions regardless of performance. Finally, many studies finding positive effects of task conflict on performance, also found negative effects of task conflict on team member satisfaction (e.g. Jehn, 1995). In this sense, outcomes regarding task conflict are open to doubt. Several studies have shown that high task conflict decreases satisfaction (e.g. Jehn, 1995) and cohesion (e.g. Jehn and Mannix, 2001), while increasing stress levels (Friedman et al., 2000). De Dreu and Weingart’s meta-analysis demonstrates that task conflict might be as negative as relationship conflict. Conflict stimulation by discrepancy and constructive criticism could have a pernicious effect, because it increases the intensity and quantity of conflict that teams must manage (De Dreu and Weingart, 2003). The incongruence in the role played by task conflict may possibly be understood by evaluating the link between task and relationship conflict. There are two lines in the study of these relations: types of conflict could play a moderational role, and types of conflict could play a mediational role. The first line of research was tested by Janssen et al. (1999) who analyzed the effect of different combinations of types of conflict (low vs high task and relationship conflict) on employees’ affective reactions such as satisfaction. Janssen et al.’s study did not provide any evidence about the positive or negative effects of these combinations on employees’ satisfaction. However, it is possible that these combinations did not affect satisfaction over the performance of a specific task, but did influence other more general and stable affective reactions, such as job satisfaction, psychological well-being or the propensity to leave the job. As De Dreu and Weingart’s meta-analysis suggests, an increase in conflict intensity could be dysfunctional for teams. For this reason, combinations that involve a high intensity of conflict – like high task and relationship conflict – would have a negative influence on subjects’ affective reactions. The second line of research was suggested in Friedman et al.’s (2000) study which, using a path analysis, revealed that the link between task conflict and affective variables such as stress at work, are mediated by relationship conflict. However, Friedman’s study did not analyze the opposite path – task conflict as mediator in the association between relationship conflict and effectiveness. Several authors have suggested that a team member might try to cause difficulties or sabotage the work of a co-worker for personal motives (e.g. Jehn, 1995). In this case, personal discrepancies (relationship conflict) may play a mediational role by generating conflicts at work (task conflict) that will ultimately decrease team productivity. The existing literature provides strong support for the negative impact that relationship conflict plays in affective reactions in the workplace. Nevertheless, additional evidence is necessary to contrast the role of task conflict and the mediating and moderating role of relationship conflict. As Friedman et al. (2000) pointed out, the connections between task conflict and relationship conflict represents one of the key areas for managing conflict in organizations. For this reason, this study aims to evaluate the relationships between task and relationship conflict, and their influence on affective variables such as satisfaction, well-being, and propensity to leave a job; and

Types of intragroup conflict 221

JMP 20,3/4

evaluate the mediating and moderating role of relationship conflict. We hypothesized that: H1. Relationship conflict will reduce satisfaction and well-being in organization members and increase propensity to leave a job.

222

H2. The combination consisting of a high task conflict and a high relationship conflict will decrease satisfaction and well-being, and will increase propensity to leave a job. H3. Relationship conflict will mediate the link between task conflict and satisfaction, well-being and propensity to leave the job. Method Participants The study involved 169 employees from six hotels in Andalusia (Spain). A total of 46.7 percent were male, 50 percent female, and 3.3 percent of respondents gender was not specified. Participants had different educational levels: elementary school (37.8 percent), high school (24.2 percent), high school graduates (24.2 percent), and university graduates (8 percent). Work experience ranged from 3 months to 44 years, and the average age was 31.45. Employees came from the following departments: 8.61 percent from administration, 21.53 percent from reception, 28.70 percent from cafeteria and restaurant, 29.18 percent from cleaning service and 5.26 percent from maintenance service. In terms of employment status, 43.06 percent had a full-time contract, 38.75 percent had a part-time contract and 16.26 percent had a periodically renewed contract. Instruments Task and relationship conflict. Jehn’s (1995) four-item scale was used to assess task conflict. The scale asks respondents to consider the amount of task or work-based conflict he or she experiences with others in the work place (e.g. how often do people you work with disagree about opinions regarding the work being done). The scale used a five-point format. The higher the score, the higher the level of task conflict experienced. To measure relationship conflict we relied on Cox’s (Cox, 1998; Friedman et al., 2000) organizational conflict scale. Cox’s five-items scale focuses on the active hostility found in relationship conflict and is based on items such as “much plotting takes place behind the scenes” and “one party frequently undermines the other”. The scale uses a five-point response format. The higher the score, the higher the levels of relationship conflict experienced. Job satisfaction. We measured job satisfaction with the 23-item version of Melia´ and Peiro´’s (1989) job satisfaction scale. The respondents were asked to consider how satisfied they were with intrinsic job aspects, supervision, participation, environment and services (e.g. “personal relationships with your superior”). In this study, we used an overall job satisfaction index. The higher the score, the more satisfied the workers. Affective wellbeing. Affective wellbeing was tapped using the scales developed by Warr (1990). These scales measure: anxiety-calm, depression-enthusiasm and contented-discontented. The six items on six-point scales measure the extent to which people are either anxious or calm, depressed or enthusiastic, contented or

discontented with their job. Respondents were asked to think of the past few weeks and indicate how they felt in different ways (e.g. calm, gloomy, cheerful, contented). Higher scores on these scales represent higher levels of affective well-being (contentment, enthusiasm, calm). Propensity to leave a job. The Gonza´lez-Roma´ et al.’s (1992) four-item scale was used to assess propensity to leave a job. The respondents were asked to consider different aspects of their workplace and reflect on whether they would change to another workplace in the same organization. The scale uses a six-point response format. The higher the score, the higher the level of propensity to leave a job.

Types of intragroup conflict 223

Procedure A questionnaire containing the aforementioned measures was used. A cover letter explaining the purpose of the study accompanied the questionnaire. Participants were told that they would be entitled to a free summary report of the study if they returned the completed questionnaire. They were told that the questionnaire was not designed for their superiors or heads of department, but for their subordinates. A researcher was present as they filled in the questionnaire to clarify any doubts. To ensure confidentiality, participants put the completed questionnaire in a sealed box. Results Table I provides the number of items, means, standard deviations and internal consistency reliability of all the behavioral variables, assessed by Cronbach’s alpha. The reliability coefficients were good to excellent. The zero order intercorrelations among the variables used in this study are presented in Table II. As can be seen, relationship conflict is negatively related to satisfaction and well-being; and positively related to propensity to leave a job. Task Variables Relationship conflict Task conflict Job satisfaction Wellbeing Propensity to leave a job

Items

a

Mean

SD

5 4 23 18 3

0.84 0.81 0.93 0.92 0.88

2.36 2.75 4.29 3.90 2.95

1.04 0.93 0.99 0.82 1.64

Notes: a = Cronbach’s alpha coefficient; SD = standard deviation; n=169

1. 2. 3. 4. 5.

Task conflict Relationship conflict Satisfaction Wellbeing Propensity to leave a job

Notes: *p , 0.05;

**

p , 0.01

2

3

4

5

0.35** –

20.30** 20.47** –

20.27** 20.45** 0.65** –

0.19* 0.39** 2 0.30** 2 0.35** –

Table I. Number of items, means, standard deviations and Cronbach’s alpha of behavioral variables presented in this study

Table II. Correlations between types of conflict and satisfaction, wellbeing and the propensity to leave a job

JMP 20,3/4

224

Table III. Regression coefficients between types of conflict and satisfaction, wellbeing and propensity to leave a job

Figure 1. Types of conflict and propensity to leave a job

conflict is negatively related to satisfaction and well-being and positively related to propensity to leave the job. This result agrees with other studies (e.g. Janssen et al., 1999), which conclude that these relations may be due to common variance between task conflict and relationship conflict. To assess this supposition we analyzed partial correlations. When controlling for task conflict, we did find negative correlations between relationship conflict and satisfaction, r ¼ 20:39, p , 0:01; well-being, r ¼ 20:37, p , 0:01 and positive correlations with the propensity to leave a job, r ¼ 20:33, p , 0:01. When controlling for relationship conflict, we found no correlations between task conflict and these variables. Data suggests that relationship conflict have a negative influence on affective reactions and that task conflict does not have influence on affective reactions. We used hierarchical multiple regression analysis to test H1 and H2 (Cohen and Cohen, 1983). First of all, the main effect of task-related conflict and relationship conflict was introduced into the equation, after which the interaction terms were included. To prevent problems of multicollinearity, these analyses were conducted with centered variables (Aiken and West, 1991). Table III shows the regression coefficients between task and relationship conflict and the different reactions to conflict. As can be seen in Table III, relationship conflict appears to decrease satisfaction ( b ¼ 20:40; p , 0:01) and psychological well-being ( b ¼ 20:39; p , 0:01). Relationship conflict also increases the predisposition to leave a job (b ¼ 0:40; p , 0:01). Introducing the two-way interaction term into the regression equation contributes substantially to the prediction of the propensity to leave a job (b ¼ 20:18; p , 0:05) As can be seen in Figure 1 the propensity to leave a job increases when high task conflict is combined with high relationship conflict; whereas it decreases when

Satisfaction DR 2 b

Block

R2

1. Task conflict Relationship conflict 2. TCXRC General model

0.24

0.24**

0.24 0.24

0.00

Notes: *p , 0.05;

**

p , 0.01

2 0.14 2 0.40** 2 0.0.2

R2

Wellbeing DR 2 b

Propensity to leave a job R2 DR 2 b

0.21

0.21**

0.11

0.16**

0.22 0.22

0.01

0.19 0.19

0.03*

2 0.12 2 0.39** 2 0.07

2 0.05 0.40** 2 0.18*

low relationship conflict is combined with high task conflict. Data suggest support for H1 and partial support for H2. H3 predicts that relationship conflict would mediate the link between task conflict and affective variables. This hypothesis was tested through regression analyses. Baron and Kenny (1986) suggest a three-step process for testing mediation using regression: (1) Regressing the mediator on the independent variable. (2) Regressing the dependent variable on the independent variable. (3) Regressing the dependent variable on both the independent variable and the mediator. For mediation to exist, the result of the first and second steps must be significant. Furthermore, if the relationship between the mediator and dependent variable in the regression equation (step 3) is also significant, but the relationship between the independent and dependent variables is not significant, then the mediator may be said to account for the relationship between the independent and dependent variables. This process was used to test relationship conflict as a mediator between task conflict and satisfaction and well-being. Two of these analyses satisfy all of the conditions for mediation. Task conflict was significantly related to relationship conflict, b ¼ 0:34; p , 0:01 (step 1) and to some affective variables such as satisfaction, b ¼ 20:34; p , 0:01 and well-being, b ¼ 20:26; p , 0:01 (step 2). The third regression can be seen in Table III. As we can see, when both types of conflict are introduced in the regression, the effect of task conflict disappears. This result suggests that the relationships between task conflict and some affective variables are fully mediated by relationship conflict. To confirm these results with a different methodology and replicate Friedman et al.’s study, comparing the different paths proposed in literature (task conflict ! relationship conflict ! affective outcomes and relationship conflict ! task conflict ! affective outcomes), we used LISREL analysis. We test two alternative models: in the first model (M1) we consider relationship conflict as a mediator variable and in the second (M2), we consider task conflict as a mediator variable (see Table IV). Table IV shows that M1 models have a higher adjustment in all indicators than M2 models. This finding confirms that the relationships between task conflict and some affective variables such as satisfaction and well-being are mediated by relationship conflict. Discussion and general conclusions Our objective in this study was to analyze the relationship between different types of intragroup conflict and employees’ affective reactions. We investigated the effects of task and relationship conflicts on several criterion variables such as satisfaction, well-being and propensity to leave the current job. The results indicate that the two types of conflict have different consequences. Data show that: . relationship conflict is negatively associated with affective reactions, while task conflict does not relate directly to affective reactions in a predictable way; . relationship conflict has a positive influence on the desire to leave the current job, while task conflict does not affect it negatively;

Types of intragroup conflict 225

RC! TC TC! SAT

0.34** 20.46**

4.20 0.04 1 0.138 0.98 0.04

TC ! RC RC!SAT

Re´sume´ x2 p d.f. RMSEA GFI RMR 26.6 0.00 1 0.39 0.90 0.14

0.34** 2 0.29**

Coef. path

Re´sume´ x2 p d.f. RMSEA GFI RMR

TC! RC RC! WB

Path

M1

2.84 0.09 1 0.10 0.99 0.03

0.34** 20.43**

Coef. path

Path

Re´sume´ x2 p d.f. RMSEA GFI RMR

RC! TC TC! WB

Wellbeing M2

23.98 0.00 1 0.37 0.91 0.11

0.34** 20.26**

Coef. path

Notes: TC – task conflict; RC – relationship conflict; SAT – satisfaction; WB – wellbeing; x2 – chi square; GFI – goodness of fit index; RMR – root mean square residuals; RMSEA – root mean square error of approximation; *p , 0.05; **p , 0.01

M2

226

Re´sume´ x2 p d.f. RMSEA GFI RMR

Path

Satisfaction

Coef. path

Table IV. LISREL analysis between types of conflict and affective variables

Path

M1

JMP 20,3/4

.

.

the interactive effect of relationships and task conflict shows that this interaction contributes substantially to predict the propensity to leave the current job; and relationship conflict mediates in the link between task conflict and affective reactions.

As expected, relationship conflict decreases employees’ satisfaction and psychological well-being. These results are similar to the ones found in previous research (e.g. Jehn, 1995; Jehn and Mannix, 2001; Surra and Longstreeth, 1990). These data allow for meaningful interpretations for dealing with emotional conflict and avoiding dysfunctional affective reactions from employees in the work place. An interesting result refers to the interactive effects of relationships and task conflict. It has been found that propensity to leave the current job increases when both types of conflict are present at high levels and decreases when high task conflict is combined with low relationship conflict. Given this situation, task conflict has functional effects when there are low levels of relationship conflict but it tends to become dysfunctional as the relationship conflict increases. The combination of high task and relationship conflict is dysfunctional because relationship dissents produce intolerance and antagonistic attributions concerning each other’s intentions and behaviors (Baron, 1990; Janssen et al., 1999). This mixture increases the desire to leave the job. In contrast, employees not affected by relationship conflict are able to stimulate team members and make them discuss, propose new ideas and integrate varying opinions in order to improve the job. This condition does not negatively affect the employees’ desire to leave his/her current job. As Amason (1996) and Korsgaard et al. (1995) state, when subjects are able to express their opinions, they are more involved with group decisions and their desire to remain in the work group is enhanced. The absence of interactive effects between task and relationship conflict and some affective variables such as satisfaction and well-being agree with previous findings that suggest no moderating effects between the two types of conflict. Neither did Janssen et al.’s (1999) study find any two-way interaction patterns between task and relationship conflict, conflict behavior, decision quality or affective acceptance. Future studies are needed to confirm these interactive effects in other samples and organizations. The present study provides strong evidence for the link between task and relationship conflict. Thus, relationship conflict mediates the link between task conflict and affective reactions. Therefore the benefits from task conflict may sometimes disappear as the level of relationship conflict increase. Some incongruent results that appear in literature can be clarified by this meditational role. A possible explanation is that task conflict may develop a relationship conflict in the future. This is in consonance with Jehn (1997), De Dreu (1997), Friedman et al. (2000) or Simons and Peterson’s (2000) studies, which pointed out the transformation of task conflict into relationship conflict. These results suggest that enhancing task conflict may backfire, as the transformation of task into relationship conflict may be counterproductive. Future longitudinal studies are needed to analyze this transformation and to search for the means and instruments to break this link and prevent the negative consequences of relationship conflict in teams. The practical implications of this study are important. As Dechurch and Marks (2001) found, subjects react negatively to a badly handled conflict even if the results are

Types of intragroup conflict 227

JMP 20,3/4

228

favorable. As a consequence, a model of the mechanisms that link task and relationship conflict is essential to provide practical guidance on how to manage intragroup conflicts. Our results suggest that it is important to consider the task conflict’s intensity in teams, since a high level of task conflict may backfire by boosting relationship conflict as well, thus having a negative effect on affective reactions. On the basis of these results, some conclusions can be drawn with a view to improving conflict management in teams. First of all, before planning the intervention, an attempt must be made to understand the type of conflict that is taking place. Secondly, managers should encourage open discussion of task related issues, in certain limits, in order to improve the quality of decisions as well as their acceptance by the group members. However, managers should mitigate or resolve relationship conflicts as soon as possible. Thirdly, special attention should be paid to the level of each conflict because of its interactive effects on some affective outcomes. Thus, in spite of the generally beneficial effects associated with task conflict, the intensification of task related conflict may backfire when interacting with dysfunctional affective-dissent. Some limitations in our study should be noted. Firstly, the correlational evidence between the different variables of the study does not necessarily reflect causal links between them. Future research is needed using independent variables of conflict types to analyze links between conflict issues and affective variables. However, the dynamic and interactive nature of conflict complicates the validity of this kind of study. Secondly, our results have been obtained using self-report measures and as a consequence the threat of common method variance exists. However, this risk is reduced by using standardized instruments (Spector, 1987) as the present study does. In conclusion, this article clarifies important aspects of the contradictory findings concerning the benefits of task conflict in teams. It is possible that incongruous evidence could be due to the meditational role of relationship conflict and interactions between conflict types. Consideration of conflict as a dynamic and multifaceted process is necessary to guarantee organizational effectiveness. References Aiken, L.S. and West, S.G. (1991), Multiple Regression: Testing and Interpreting Interactions, Sage, Newbury Park, CA. Amason, A.C. (1996), “Distinguishing the effect of functional and dysfunctional conflict on strategic decision making: resolving a paradox for top management teams”, Academy of Management Journal, Vol. 39, pp. 123-48. Baron, R.A. (1990), “Countering the effects of destructive criticism: the relative efficacy of four interventions”, Journal of Applied Psychology, Vol. 75, pp. 235-45. Baron, R.M. and Kenny, D.A. (1986), “The moderator-mediator variable distinction in social psychological research: conceptual, strategic, and statistical considerations”, Journal of Personality and Social Psychology, Vol. 51, pp. 1173-82. Cohen, J. and Cohen, P. (1983), Applied Multiple Regression/Correlation Analysis of the Behavioral Sciences, Erlbaum, Hillsdale, NJ. Cosier, R.A. and Schwenck, C.R. (1990), “Agreement and thinking alike: ingredients for poor decisions”, Academy of Management Executive, Vol. 4, pp. 69-74. Cox, K.B. (1998), “Antecedents and effects of intergroup conflict in the nursing unit”, unpublished doctoral dissertation, Virginia Commonwealth University, Richmond, VA.

DeChurch, L.A. and Marks, M.A. (2001), “Maximizing the benefits of task conflict: the role of conflict management”, International Journal of Conflict Management, Vol. 12, pp. 4-22. De Dreu, C.K.W. (1997), “Productive conflict: the importance of conflict management and conflict issue”, in De Dreu, C K.W. and Van de Vliert, E. (Eds), Using Conflict in Organizations, Sage, London, pp. 9-22. De Dreu, C.K.W. and Van de Vliert, E. (Eds) (1997), Using Conflict in Organizations, Sage, London. De Dreu, C.K.W. and Van Vianen, A.E.M. (2001), “Managing relationship conflict and the effectiveness of organizational teams”, Journal of Organizational Behavior, Vol. 22, pp. 309-28. De Dreu, C.K.W. and Weingart, L.R. (2003), “Task versus relationship conflict: a meta-analysis”, Journal of Applied Psychology, Vol. 88, pp. 741-9. Friedman, R.A., Tidd, S.T., Currall, S.C. and Tsai, J.C. (2000), “What goes around comes around: the impact of personal conflict styles on work conflict and stress”, International Journal of Conflict Management, Vol. 11, pp. 32-55. Gonza´lez-Roma´, V., Merı´, S., Luna, R. and Lloret, S. (1992), “Propiedades psicome´tricas de un cuestionario para medir la propensio´n al abandono del puesto de trabajo y de la organizacio´n, (“Psychometrical proprieties of the questionnaire to measure propensity to leave a job”), Revista de Psicologı´a Social Aplicada, Vol. 2, pp. 25-42. Hackman, J.R. and Morris, C.G. (1975), “Group task, group interaction process, and group performance effectiveness: a review and proposed integration”, in Berkowitz, L. (Ed.), Advances in Experimental Social Psychology, Vol. 8, Academic Press, San Diego, CA, pp. 45-99. Janssen, O., Van de Vliert, E. and Veenstra, Ch. (1999), “How task and person conflict shape the role of positive interdependence in management teams”, Journal of Management, Vol. 25, pp. 117-41. Jehn, K.A. (1994), “Enhancing effectiveness: an investigation of advantages and disadvantages of value-based intragroup conflict”, International Journal of Conflict Management, Vol. 5, pp. 223-38. Jehn, K.A. (1995), “A multimethod examination of the benefits and detriments of intragroup conflict”, Administrative Science Quarterly, Vol. 40, pp. 256-82. Jehn, K.A. (1997), “A qualitative analysis of conflict types and dimensions in organizational groups”, Administrative Science Quarterly, Vol. 42, pp. 530-57. Jehn, K.A. and Chatman, J.A. (2000), “The influence of proportional and perceptual conflict composition on team performance”, International Journal of Conflict Management, Vol. 11, pp. 56-73. Jehn, K.A. and Mannix, E. (2001), “The dynamic nature of conflict: a longitudinal study of intragroup conflict and group performance”, Academy of Management Journal, Vol. 44, pp. 238-51. Jehn, K., Northcraft, G. and Neale, M.A. (1999), “Why difference makes a difference: a field study of diversity, conflict and performance in work group”, Administrative Science Quarterly, Vol. 44, pp. 741-63. Korsgaard, M.A., Schweiger, D.M. and Sapienza, H.J. (1995), “Building commitment, attachment, and trust in strategic decision-making teams: the role of procedural justice”, Academy of Management Journal, Vol. 38, pp. 60-84. Melia´, J.L. and Peiro´, J.M. (1989), “La medida de la satisfaccio´n laboral en contextos organizacionales: el cuestionario de satisfaccio´n S20/23, (“The assessment of job

Types of intragroup conflict 229

JMP 20,3/4

230

satisfaction in organizational environments: the S20/23 questionnaire”), Psicologema, Vol. 3, pp. 55-74. Pelled, L.H., Eisenhardt, K.M. and Xin, K.R. (1999), “Exploring the black box: an analysis of work group diversity, conflict and performance”, Administrative Science Quarterly, Vol. 44, pp. 1-28. Pondy, L. (1967), “Organizational conflict: concepts and models”, Administrative Science Quarterly, Vol. 17, pp. 296-320. Pruitt, D.G. and Carnevale, P.J. (1993), Negotiation in Social Conflict, Open University Press, Milton Keynes. Simons, T.L. and Peterson, R.S. (2000), “Task conflict and relationship conflict in top management teams: the pivotal role of intragroup trust”, Journal of Applied Psychology, Vol. 85, pp. 102-11. Spector, P.E. (1987), “Method variance as an artifact in self-reported affect and perceptions at work: myth or significant problems?”, Journal of Applied Psychology, Vol. 72, pp. 438-43. Stone, R.A. (1995), “Workplace homicide: a time for action”, Business Horizon, Vol. 34, pp. 17-20. Surra, C. and Longstreeth, M. (1990), “Similarity of outcomes, interdependence and conflict in dating relationships”, Journal of Personality and Social Psychology, Vol. 59, pp. 501-16. Turner, M.E. and Pratkanis, A. (1994), “Social identity maintenance prescriptions for preventing groupthink: reducing identity protection and enhancing intellectual conflict”, International Journal of Conflict Management, Vol. 5, pp. 254-70. Van de Vliert, E. and De Dreu, C. (1994), “Optimizing performance by conflict stimulation”, The International Journal of Conflict Management, Vol. 5, pp. 211-22. Wall, J.A. and Callister, R.R. (1995), “Conflict and its management”, Journal of Management, Vol. 21, pp. 515-58. Warr, P. (1990), “The measurement of wellbeing and other aspects of mental health”, Journal of Occupational Psychology, Vol. 63, pp. 193-210. West, M.A. and Anderson, N.R. (1996), “Innovation in top management teams”, Journal of Applied Psychology, Vol. 81, pp. 680-93.

The Emerald Research Register for this journal is available at www.emeraldinsight.com/researchregister

The current issue and full text archive of this journal is available at www.emeraldinsight.com/0268-3946.htm

Exploring the effects of intragroup conflict and past performance feedback on team effectiveness

Effects of intragroup conflict 231

Ana Margarida Passos and Anto´nio Caetano Instituto Superior de Cieˆncias do Trabalho e da Empresa, Lisboa, Portugal Abstract Purpose – The purpose of this paper is to test a model of the effects of intragroup conflict (relationship conflict, task conflict and process conflict), past performance feedback and perceptions of team decision-making effectiveness on team performance and affective responses. Design/methodology/approach – A sample of 183 individuals, working in 47 different teams, participated in this study. All the teams were involved in a national management challenge for a five-week period. Three questionnaires were sent directly to team members by e-mail at different times of the challenge period to collect data concerning demographic data (questionnaire 1), perceptions of team functioning (questionnaire 2) and perceptions of team decision-making effectiveness as well as the affective responses (questionnaire 3). The level of analysis in this study was the group. Thus, all individual survey responses were aggregated to the team level for statistical analysis. Findings – Results showed a full mediation effect of perceptions of team decision in the relationship between process conflict and team performance. Task and relationship conflict showed no significant relationships with team performance and satisfaction with the team. The result that effective past performance feedback directly influences team performance, in a positive way, suggests that past effective decisions may reinforce the decision-making processes previously used by team members. Research limitations/implications – One possible limitation of this study is the fact that measurements were taken at different times of the management challenge. In fact, while intragroup conflict was measured two weeks after the beginning of the challenge, the other variables were measured at the end of the challenge. This time measurement difference could raise some questions concerning the stability of the intragroup conflict over time in work teams. Future research should address this hypothesis. Future research should also elucidate the influence of contextual variables, such as cultural values, on the relationship between intragroup conflict and performance outcomes. Practical implications – This study helps managers to understand how to benefit from conflict. In a highly competitive environment, disagreement among team members about “how to do it” seems to decrease decision-making effectiveness. Originality/value – This study fills a gap in the conflict literature concerning the impact of intragroup conflict in the team members’ perceptions of decision-making effectiveness and how it affects the overall performance. Moreover, this study also clarifies the importance of past performance to the actual team outcomes. Keywords Conflict, Team working, Performance management, Team performance Paper type Research paper

Introduction Teams and teamwork have received considerable attention from social and organizational psychologists over the last decades (Guzzo and Shea, 1992). Despite the fact that teams are usually expected to improve organizational effectiveness, there are several challenges to teamwork. One challenge is intragroup conflict. De Dreu and

Journal of Managerial Psychology Vol. 20 No. 3/4, 2005 pp. 231-244 q Emerald Group Publishing Limited 0268-3946 DOI 10.1108/02683940510589028

JMP 20,3/4

232

Weingart (2002) define intragroup conflict as “the process resulting from tension between team members due to real or perceived differences” (p. 3). The traditional approach to conflict, dominant in the literature of the 1930s and 1940s of the last century, assumed that all conflicts were dysfunctional for group dynamics. Therefore, the main research focus in this field was to study the causes of conflict in order to prevent it and improve the team and organizational performance. In contrast to this view, the human relations’ approach argued that conflict was a natural feature in all groups and organizations. In fact, proponents of that approach considered that there were times when conflict could benefit group performance. In alternative to those approaches, the interactionist approach considers that conflict is necessary and it should be encouraged in order to maximize the team and the organizational performance. According to the interactionist view, a minimum level of conflict is necessary to keep the team viable, self-critical, and innovative (Lewicki et al., 1992). This last approach of conceptualizing conflict has encouraged the development of several empirical research studies designed to evaluate in which circumstances a conflict is functional or dysfunctional to team performance (Jehn, 1994, 1995; Amason, 1996; Simons and Peterson, 2000). However, the empirical findings concerning the effects of intragroup conflict on performance are highly inconsistent. Despite the contradictory results, one helpful contribution to the development of a conceptual framework of intragroup conflict, within the interactionist approach, was the distinction between different types of conflict (Jehn, 1994, 1995; Amason, 1996). Initially, Jehn (1994, 1995) defined two types of intragroup conflict: the task conflict and the relationship conflict. Task conflict, or cognitive conflict, is a perception of disagreements among team members about the content of their decisions and involves differences of opinions, ideas and viewpoints. Relationship conflict, or emotional conflict, is a perception of interpersonal incompatibility and typically includes tension, irritation and hostility among team members. It is a kind of conflict with a strong personal and emotional component, characterized by friction and personality differences among team members. Jehn (1994) distinction was based on the work of Guetzkow and Gyr (1954), who first identified these two dimensions of intragroup conflict. These authors recognized the potential value of conflict but made an important distinction between substantive conflict and affective conflict. While the substantive conflict is rooted in the substance of the task and is more intellectual in origin, the affective conflict derives from the emotional aspects of the group interpersonal relations. The distinction between affective and cognitive aspects of conflict is critical to understand in which circumstances a conflict can be beneficial to team performance (De Dreu, 1997). Jehn conducted several empirical research studies to investigate the influence of task versus relationship conflict on teams. The results showed that task conflict enhances team performance, while relationship conflict has a negative influence on team performance as well as on the satisfaction with the team. Amason (1996) investigated the influence of task and relationship conflict in strategic decision making teams and reached similar conclusions. While the relationship conflict reduced the quality of the decisions of top management teams, task conflict improved the quality of the decision as well as the team members’ implication with their decisions. In a recent meta-analysis on team conflict literature, De Dreu and Weingart (2002) examined the associations found in empirical research between task and relationship

conflict and team outcomes (team performance and team members’ satisfaction). Concerning the effects of intragroup conflict on task performance, the results showed that both task and relationship conflict are equally disruptive. In fact, even when the authors investigated possible moderators (task type, group type, country and performance measure) of the relationship between intragroup conflict and team performance, they found no single positive correlation. In regards to team members’ satisfaction, De Dreu and Weingart’s (2002) results showed that relationship conflict is more dysfunctional than task conflict. The authors concluded that there are no differential associations between task and relationship conflict, on the one hand, and team performance on the other, and questioned the functionality of these processes to work teams. The research on intragroup conflict has neglected a conflict type that was identified later by Jehn (1997) – the process conflict. The author defined process conflict as the conflict about how tasks should be accomplished by the team, including the distribution of responsibilities and the delegation of tasks among their members. Thus, process conflict is concerned with disagreements about assignments of duties and distribution norms of resources. Unlike the distinction between task conflict and relationship conflict, which was based on a theoretical reflection, the distinction between task conflict and process conflict came out of the empirical data analysis. According to Jehn (1997), process conflict is similar to the concepts of distributive conflict (Kabanoff, 1991) and procedural complexity (Kramer, 1991) and therefore it is possible to differentiate it from the concept of task conflict. There is no consensus in the empirical literature concerning the impact of the process conflict on team performance and on the team members’ affective responses. Jehn (1997) found that high levels of process conflict had a dysfunctional effect in the team’s satisfaction and performance. Jehn and Chatman (2000) put forth the hypothesis that this type of conflict tends to increase team performance and the positive responses of team members, as it is the case of satisfaction and team commitment. However, this hypothesis was not supported by their data. In fact, the authors found opposite results. Another relevant aspect for intragroup conflict analysis is the possible influence of cultural values on this process. De Dreu and Weingart (2002) studied the moderator role of the country on the relationship between intragroup conflict and team outcomes (team performance and team member satisfaction). Based on studies performed in the US and in The Netherlands, they found interesting results. The negative correlations between task conflict and team performance as well as the negative correlations between task conflict and team member satisfaction were weaker in the US studies. On the other hand, the negative correlation between relationship conflict and team member satisfaction was stronger in the US studies than the ones performed in The Netherlands. Sosik and Jung (2002) conducted a cross-cultural study to investigate the effects of individualistic versus collectivistic values on group performance of work groups from US and Korea. The US work teams (individualistic) attained higher levels of group performance than did the Korean ones (collectivistic). To explain the obtained results, the authors suggested that collectivistic groups may have focused more effort on building social and interpersonal relations among members than on the task they had to perform. Moreover, collectivistic groups tended to come to consensus prematurely in decision-making tasks.

Effects of intragroup conflict 233

JMP 20,3/4

Therefore, these findings suggest that the impact of the intragroup conflict on team outcomes may be influenced, to a certain degree, by the cultural values of a country. Due to the fact that the present study was conducted in a Latin country, with a dominance of collectivistic values (Hofstede, 1991), it is interesting to find out if the results concerning to intragroup conflict are similar to the ones reports by the Anglo-American literature.

234 Intragroup conflict and team decision-making effectiveness Most studies of team intragroup conflict have focused on the immediate outcomes of the conflict such as decision quality, and satisfaction with the team (e.g. Amason, 1996; Jehn and Chatman, 2000), or have investigated moderator variables that affect the relationship between intragroup conflict and team outcomes (e.g. Jehn, 1995; Jehn et al., 1999). A more recent focus of research has investigated how conflict management styles used to deal with the intragroup conflict influence team outcomes (e.g. Kuhn and Poole, 2000; De Dreu and Van Vianen, 2001). A substantial amount of work has been done in order to understand the influence of intragroup conflict on team outcomes. However, there is still a lack in the literature concerning the impact of intragroup conflict on the decision making process itself. Some authors consider that conflict can increase the scrutiny of the team decision options (Putnam, 1994), and encourage a cognitive understanding of the issues involved in the decision process (Simons and Peterson, 2000). Other authors consider that conflict is time consuming and that it often appears frustrating for team members (Kuhn and Poole, 2000). Schwenk (1990) summarized the literature that has dealt with the effects of conflict on the decision process (e.g. Cosier, 1980; Schwenk and Cosier, 1980; Schwenk, 1982) and suggested that conflict can improve prediction of performance, leads to increased understanding of alternative views, and improves the evaluation of these alternatives. According to this author, conflict may stimulate systematic questioning, and therefore, may lead to a greater cognitive effort in decision making than the one that would be reached without it. In that way, conflict “ensures that the available information is used effectively” (Schwenk, 1990, p. 438). However, there is an important contrast when we look to the team members’ perceptions of the effects of conflict. Team members usually consider that conflict is harmful to team performance and attempt to avoid it. Schweiger et al. (1986) found that group members perceive conflict as a process that reduces the potential for future decision-making. The functional theory of group decision-making (Gouran and Hirokawa, 1996) posits that group interaction affects performance by shaping the team’s critical thinking that leads to a final team choice. Critical thinking depends on the team ability to analyze the problem, to establish goals, and to evaluate the positive and negative outcomes of solutions. Hence, team interaction patterns fostered by intragroup conflict can either promote or inhibit these functions. Based on the intragroup conflict literature we consider that task conflict promotes team members’ ability to consider multiple viewpoints enhancing critical thinking and therefore the perceptions of decision-making effectiveness. Relationship conflict, on the other hand, may limit the information processing ability of the team due to the fact that members spend their energy focusing on the personal antagonisms rather than on the task (Simons and Peterson, 2000). Thus, relationship conflict reduces the perception of team decision-making effectiveness. The relationship between process conflict and

decision-making has less theoretical and empirical support. We suppose that this effect is strongly influenced by team work contextual constrains. For instance, in a high competition context with high time pressure, the disagreements among team members about how they should accomplish the goal and the strategy to be used may be understood as a process loss (Steiner, 1972). Accordingly, process conflict should reduce the perception of decision-making effectiveness. Therefore, it is relevant to analyze the following hypotheses, concerning the relationship between intragroup conflict and perceptions of decision-making effectiveness in a competitive and time pressure context: H1. Task conflict is positively associated with the perceptions of team decision-making effectiveness. H2. Relationship and process conflicts are negatively associated with the perceptions of team decision-making effectiveness. H3. The perceptions of team decision-making effectiveness are positively associated with the team performance and the satisfaction with the team. Many researchers have concluded that the most effective teams are those that can gain the benefits of task conflict while avoiding the costs of affective conflict (Amason, 1996; Jehn, 1994, 1995). However, as discussed earlier, empirical results are inconsistent. In this study we argue that intragroup conflict influences team performance through the perceptions that team members have concerning the process through which a decision is made. Therefore we formulated the following hypotheses: H4. The relationship between intragroup conflict (task, relationship, and process conflict) and team performance is mediated by the perceptions of team decision making effectiveness. H5. Intragroup conflict (task, relationship, and process conflict) has a direct negative effect on the satisfaction with the team. Past performance feedback and team performance As Bateman and Zeithaml (1989) suggested, the success or failure of past events is an important aspect of the psychological context that constrains individuals’ present or future decisions. The argument that past performance constrains actual team performance is supported by the work of Weldon et al. (1991), who found a positive effect of past team performance on actual performance. Past performance may influence the team’s actual behavior by inducing performance adjustments when the outcome is poor or reinforcing the processes and strategies that revealed to be effective. Marks et al. (2001) proposed a time-based conceptual framework to study team processes and performance. According to the authors, team performance should be understood in temporal cycles of goal-directed activity called episodes. Each episode is a distinguishable period of time over which performance feedback is available and outcomes from initial episodes become inputs for the next cycle. So, we argue that past team performance sets the stage for future decision making and consequently to future team performance. Therefore we formulated the following hypothesis:

Effects of intragroup conflict 235

JMP 20,3/4

236

H6. Effective past performance feedback is positively associated with the actual team performance. Hence, the purpose of this paper is to test a model of the effects of intragroup conflict on team performance and affective responses. It is proposed that the relationship between intragroup conflict (relationship conflict, task conflict and process conflict) and team performance is mediated by team members’ perception of decision-making effectiveness. However, the model also states that the intragroup conflict directly affects the team affective outcomes. Concerning the impact of past performance feedback, it is assumed a direct positive influence between this variable and the actual team performance (Figure 1). Method Sample A sample of 183 individuals, working in 47 different teams, participated in this study. All the teams were involved in a national management challenge for a five-week period. The average age of team members was 28.92 years and 56.8 percent were male. Average team size was 4.93 members (SD ¼ 0.716). Procedure The management challenge is based on a realistic simulation of business life as seen through managing a fictitious company. In this simulation, a number of companies, represented by the various teams, compete in a common business environment in order to achieve the highest share price for their company. Teams have to manage the company making decisions about marketing, production, personnel, purchasing and finance subjects. This particular challenge is very well known in the country, and the majority of team members come from important national and multinational companies as well as from universities.

Figure 1. Proposed model of the effects of intragroup conflict and past performance on actual team performance and satisfaction with the team

Three questionnaires were sent directly to team members by e-mail at different times of the challenge period. Questionnaire 1 was sent just before the challenge started and was intended to collect demographic data about the teams. Questionnaire 2 was sent two weeks after the beginning of the challenge. At this time we collected data about team functioning, namely intragroup conflict. At the end of the challenge, and just before the results were announced, we sent the last questionnaire to collect data about the perceptions of team decision-making effectiveness as well as the affective responses regarding the perceptions of team functioning effectiveness. Participants sent back the questionnaires by e-mail to the researchers.

Measures Intragroup conflict. Intragroup conflict was measured by the 12 items of the intragroup conflict scale developed by Jehn (1995). Participants answered on a seven-points rating-scale (1 ¼ completely disagree; 7 ¼ completely agree) reflecting the extent of their agreement with each statement. Four items measured task conflict (e.g. “there are differences of opinions regarding the task in my team”), four items measured relationship conflict (e.g. “friction is present in my team”) and four items measured process conflict (e.g. “members of my team disagree about who should do what”). Alpha coefficients for these scales were 0.79 for task conflict, 0.84 for relationship conflict and 0.62 for process conflict. Higher scores reflect more intragroup conflict. Decision-making effectiveness. Team decision-making effectiveness was operationalized through four items of the group reaction questionnaire (Kuhn and Poole, 2000). Examples of the four items include “the team thoroughly considered and evaluate all the relevant information to make the decisions” and “the team made good use of time to make the decisions”. Participants answered on a seven-points rating-scale (1 ¼ completely disagree; 7 ¼ completely agree) reflecting the extent of their agreement with each statement. Higher scores indicated greater perception of decision-making effectiveness. Alfa coefficient for this scale was 0.81. Past performance feedback. During the challenge teams received feedback about their former decisions. A computer model analyzed the marketing, production, personnel, purchasing and finance decisions made by the various competing teams and calculated the ranking of all teams. The ranking position was used as a measure of the objective team performance feedback. Lower scores indicated better performance feedback. Satisfaction with the team. Satisfaction with the team was measured at the end of the challenge through a scale with five items. Participants indicated on a seven points rating-scale (1 ¼ totally unsatisfied; 7 ¼ totally satisfied) the extent of their satisfaction with each of the following items: team, team functioning, communication among team members, team leadership and relationship climate. Examples of the five items include “in what extent are you satisfied with your team” and “in what extent are you satisfied with your team functioning”. Alfa coefficient for this scale was 0.86. Higher scores indicated greater satisfaction with the team. Team performance. The final share price obtained by each team was used to measure team performance. The share price reflects the results obtained by each team and the general financial condition of the company. Higher scores indicated better performance.

Effects of intragroup conflict 237

JMP 20,3/4

238

Aggregation The level of analysis in this study was the group. Thus, all individual survey responses were aggregated to the group level for further analysis. The aggregation is conceived to reduce the impact of individual differences within each team. To compute aggregation responses from team members have to be more similar to each other than would be expected by chance (Bliese, 2001). To justify aggregation of the studied variables we computed Rwg(j) (James et al., 1984, 1993) designed for multiple-item scales. The Rwg for relationship conflict, task conflict, process conflict, perceptions of decision-making effectiveness and satisfaction with the team averaged 0.92, 0.75, 0.85, 0.89, and 0.93, respectively. Because the values exceeded the minimum criterion of 0.70 (Cohen et al., 2001), it was deemed appropriate to aggregate the variables to team level for further analysis. Results Table I presents the correlation matrix for the variables included in this study. This table includes the means, standards deviations, and reliabilities, where appropriate. Multiple regression analyses were used to test the hypothesized relationship between intragroup conflict and performance feedback, and the outcome measures. It was not possible to perform structural equation modeling due to the sample size (n ¼ 47). In order to test the mediation model we followed Baron and Kenny’s (1986) three-step procedure. Step 1 of the procedure states that when the mediator variable is regressed on the predicting variables, the regression weights for the predictors should be statistically significant. Step 2 states that when the criterion variable is regressed on the predicting variables, the regression weights for the predictors should be statistically significant. Finally, in the third step of this procedure, the criterion variable is regressed on the mediator and the predictor variables. If full mediation exists, the mediator should be statistically significant but the effects of the predictor variables should not be statistically significant. Step 1 of the test for mediation is the same for both team performance and the satisfaction with the team and is reported in Table II. Team size was used as a control Variables

Table I. Means, standard deviations, intercorrelations and scale reliabilities

1. Team size 2. Relationship conflict 3. Task conflict 4. Process conflict 5. Past performance feedback 6. Team decision making 7. Team performance 8. Satisfaction with the team

Mean SD

1

2

3

4

4.93

0.72

1.36 3.68 2.22

0.56 0.99 0.75

0.08 0.25 0.24

(0.84) 0.26 0.54**

(0.79) 0.68**

(0.62)

4.09

2.52

0.28

0.09

0.13

0.28

5.50 1.35

0.81 20.17 0.16 0.13 20.19 20.15

6.09

0.68 20.20 20.41** 20.18

5

6

7

8

20.29* 2 0.59** 2 0.27 (0.81) 20.08 2 0.34* 2 0.54** 0.55** 2 0.50** 2 0.06

0.40* 0.20 (0.86)

Notes: Coefficient alpha reliabilities are reported on the main diagonal where appropriate; *p , 0.05; p , 0.01

**

variable and entered in the equation as the first step. A statistically significant negative relationship of process conflict (b ¼ 2 0.93, p , 0.01) was found when perceptions of decision making effectiveness were regressed on the predictor variables. Process conflict predicted 36 percent ( p , 0.01) of the variance in perceptions of decision making effectiveness. These results provide some support for H2 that states that process conflict is negatively associated with the perceptions of team decision making effectiveness. However, we found no support for the relationship between task and relationship conflict and the perceptions of team decision-making effectiveness. In fact, the association between task conflict and perceptions of team decision-making effectiveness is not significant. Therefore, the results do not support H1. The first half of Table III reports step 2 of the mediation test for the team performance. In this analysis, team performance was regressed on team size, relationship conflict, task conflict, process conflict and past performance feedback. This analysis revealed statistically significant negative effects for process conflict (b ¼ 2 0.42, p , 0.05) and past performance feedback (b ¼ 2 0.46, p , 0.01). Predictor variables

DR 2 (block)

b

1. Team size 2. Relationship conflict Task conflict Process conflict Past performance feedback Overall adjusted R 2

2 0.17 0.28 0.32 2 0.93** 2 0.06

0.01

0.36** 0.36**

Notes: Standardized betas are reported; *p , 0.05;

Variables 1. Team size 2. Team size Relationship conflict Task conflict Process conflict Past performance feedback 3. Team size Relationship conflict Task conflict Process conflict Past performance feedback Decision-making effectiveness Overall adjusted R 2

**

Team performance DR 2 (block) F

b 20.19 20.03 0.06 0.23 20.42* 20.46 20.01 20.08 0.08 0.02

**

20.43

**

0.47

**

0.01

0.28**

5.34

dfs

1.23

1,45

7.11 6.07

4,41 5,41

p , 0.01

dfs

1.66

F

1,45

4,41

9.96 6.49

Notes: standardized betas are reported; *p , 0.05;

**

1,40 6,40

p , 0.01

239

Table II. Regression analysis of the perception of decision-making effectiveness on predictor variables

Satisfaction with the team DR 2 b (block) F dfs 2 0.20 2 0.15 2 0.12 0.32 2 0.68 0.12 2 0.15 2 0.16 0.27 2 0.53

0.02

1.76

1,45

0.27*

5.04

4,41

**

**

0.14 0.13** 0.42**

Effects of intragroup conflict

0.17

0.00 0.29*

1.05 3.94

1,40 6,40

Table III. Regression analysis of team performance and satisfaction with the team on the predictor and the mediator variables

JMP 20,3/4

240

Together, these two variables explained 28 percent ( p , 0.01) of the variance in team performance. Regarding the satisfaction with the team, only process conflict predicted this variable (b ¼ 2 0.68, p , 0.01), as it is shown in Table III. Process conflict accounted for 27 percent ( p , 0.01) of the variance in satisfaction with the team. These results allowed us to continue with the mediation analysis. In fact, we wanted to explore if the process conflict affected the team performance and the satisfaction with the team through the perception of decision-making effectiveness. Table III also reports the third step of the mediation test for the team performance and for the satisfaction with team. Team performance was regressed on the control variable, followed by intragroup conflict (relationship conflict, task conflict and process conflict) and past performance feedback (entered as a block) and by the perceptions of decision-making effectiveness as the third step. While past performance feedback (b ¼ 2 0.43, p , 0.01) continued to influence team performance, the relationship between process conflict and team performance became no significant (b ¼ 0.02, p ¼ 0.63) when the perceptions of decision making effectiveness entered in the equation. Moreover, perceptions of decision making effectiveness (b ¼ 0.47, p , 0.01) predicted team performance, accounting for 13 percent ( p , 0.01) of the variance in team performance. As predicted by H3, this result indicates that perceptions of decision-making effectiveness positively influenced team performance. This finding indicates that the effect of process conflict on team performance was completely mediated by the perception of decision-making effectiveness, while past performance feedback had a direct effect on team performance. These results provide some support to H4 and completely support H6. Concerning the satisfaction with the team, results are also presented in Table III. Satisfaction with the team was regressed on the control variable, followed by intragroup conflict (relationship conflict, task conflict, and process conflict) and past performance feedback. Perceptions of decision-making effectiveness entered in the equation on the third step. The predictor variables accounted for 27 percent ( p , 0.01) of the variance in the satisfaction with the team. As predicted in H5, the results showed that process conflict influences the satisfaction with the team directly, and in a negative way. There was no full or partial mediation of the perception of decision-making effectiveness on the relationship between process conflict and satisfaction with the team. In fact, contrary to our expectation, perceptions of decision-making effectiveness did not predicted satisfaction with the team. Therefore, we found no support to H3, concerning the positive relationship between the perceptions of team decision-making effectiveness and satisfaction with the team.

Discussion The present study aimed to analyze the effects of intragroup conflict on team performance and team affective responses. More specifically, it was proposed that the perceptions of team decision-making effectiveness mediate the relationship between intragroup conflict and team performance. On the other hand, intragroup conflict was expected to have a direct effect on affective responses. Results showed a full mediation effect of perceptions of team decision in the relation between process conflict and team performance.

Task and relationship conflict showed no significant relationships with team performance and satisfaction with the team. Concerning the influence of task conflict on team performance and on satisfaction with the team, these results were similar to the ones found by Jehn (1995) and De Dreu and Weingart (2002). Task conflict did not have an effect on team performance. Regarding relationship conflict, these results need some elaboration. Both Pelled et al. (1999) and Jehn (1995) found no evidence that emotional conflict impaired team performance. One possible reason for the absence of a significant relationship is the fact that, as referred by Jehn (1995), individuals tend to “avoid working with those with whom they experience emotional conflict” (Jehn, 1995, p. 276). Contrary to our expectation, there was no significant association between relationship conflict and satisfaction with the team. To understand this result, it is useful to refer to the type of teams involved in this study. Team members voluntarily chose to participate in this challenge. The majority of them had already experienced working together. The mean of relationship conflict was 1.36 for this sample, which suggests that this type of conflict was not present in work teams. Regarding the negative influence of process conflict on the perceptions of team decision-making effectiveness, it appears to follow the suggestion of Schwenk (1990) that managers tend to perceive conflict as harmful and attempt to avoid it. In fact, disagreements among team members about “how to do it” seems to be aversive to decision making effectiveness. In this particular context of high competition, teams had a clear goal of performance level and time pressure. They may view conflict as something that detracts them from achieving the team’s most important goal: the best decisions that allow them to win the competition. These findings also suggest that one should consider the context in order to understand the relationship between intragroup conflict and team performance. National cultural values may influence the way individuals interpret and react to the different types of intragroup conflict. The collectivistic values that characterize Southern European countries may lead individuals to experience intragroup conflict as something that threatens the interest of their own team. As Hofstede (1983) argued, in collectivistic cultures “everybody is supposed to have no other opinions and beliefs than the opinions and beliefs in their own group” (p. 79). Moreover, collectivistic groups may underestimate the importance of intragroup conflict for group decision-making tasks (Sosik and Jung, 2002). The result that effective past performance feedback directly influences team performance, in a positive way, follows the deterministic view discussed earlier. Past effective decisions may reinforce the decision-making processes previously used by team members. Another result that deserves some attention is the strong correlation found among the conflict types. De Dreu and Weingart (2002) reported an average correlation of 0.52 between task conflict and relationship conflict on their meta-analysis. In the present study, we found a statistically significant positive correlation of 0.54 between process conflict and relationship conflict, and 0.68 between process conflict and task conflict. These strong correlations could suggest a problem with intragroup scale validity. However, these two types of conflict correlate differently with objective team performance, satisfaction with the team and perceptions of team decision-making effectiveness, which indicates that they are different from each other. Nevertheless, it may be useful to reanalyze the measure of process conflict. The items proposed by Jehn

Effects of intragroup conflict 241

JMP 20,3/4

242

(1997) to measure process conflict fail to capture the dynamics of this conflict type. A conflict process measure should capture the disagreements among team members about the means by which they interdependently utilize the various resources to get meaningful outcomes (Marks et al., 2001). It is not just a question of disagreement about distribution of tasks, resources and responsibilities. One possible limitation of this study is the fact that measurements were taken at different times of the management challenge. In fact, while intragroup conflict were measured two weeks after the beginning of the challenge, the others variables were measured at the end of the challenge. This time measurement difference could raise some questions concerning the stability of the intragroup conflict over time in work team. However, we think that team processes, as it is the case of intragroup conflict, have important consequences on the actual and future team performance. Future research should clarity the influence of contextual variables, such as cultural values, on the relationship between intragroup conflict and performance outcomes. In sum, this study examined the effects of intragroup conflict, past performance feedback and perceptions of team decision-making effectiveness on team performance and affective responses in 47 teams working in a competing environment. Results presented here both replicated some past empirical studies as well as fit within a number of existing theoretical perspectives (De Dreu and Weingart, 2002; Peterson and Beharf, 2003). The studied variables accounted for 42 percent of the variance in team performance and for 29 percent in the case of the satisfaction with the team. Process conflict was an important predictor of both team performance and satisfaction with the team. Our results are also consistent with the framework proposed by Marks et al. (2001) suggesting that past performance can have subsequent effects on teamwork. Finally, this research contributes to understand the relationship between intragroup conflict and team performance, suggesting that perceptions of team decision-making effectiveness is an important mediator between these two variables.

References Amason, A.C. (1996), “Distinguishing the effects of functional and dysfunctional conflict on strategic decision making: resolving a paradox for top management groups”, Academy of Management Journal, Vol. 39 No. 1, pp. 123-48. Baron, R.M. and Kenny, D.A. (1986), “The moderator-mediator variable distinction in social psychology research: conceptual, strategic, and statistical considerations”, Journal of Personality and Social Psychology, Vol. 51, pp. 1173-82. Bateman, T.S. and Zeithaml, C.P. (1989), “The psychological context of strategic decisions: a model and convergent experimental findings”, Strategic Management Journal, Vol. 10, pp. 59-74. Bliese, P.D. (2001), “Within-group agreement, non-independence, and reliability”, in Klein, K.J. and Kozlowski, W.J. (Eds), Multilevel Theory, Research, and Methods in Organizations, Jossey-Bass, San Francisco, CA, pp. 349-81. Cohen, A., Doveth, E. and Eick, U. (2001), “Statistical properties of the rwg (J) index of agreement”, Psychological Methods, Vol. 6 No. 3, pp. 297-310. Cosier, R.A. (1980), “Inquiry method, goal difficulty, and context effects on performance”, Decision Science, Vol. 11, pp. 1-16.

De Dreu, C. (1997), “Productive conflict: the importance of conflict management and conflict issue”, in De Dreu, C. and Van de Vliert, E. (Eds), Using Conflict in Organizations, Sage Publications, London, pp. 9-22. De Dreu, C. and Van Vianen, A.E.M. (2001), “Managing relationship conflict and the effectiveness of organizational teams”, Journal of Organizational Behavior, Vol. 22, pp. 309-28. De Dreu, C. and Weingart, L.R. (2002), “Task versus relationship conflict, team performance, and team member satisfaction: a meta-analysis”, unpublished manuscript, University of Amsterdam, Amsterdam. Gouran, D.S. and Hirokawa, R.Y. (1996), “Functional theory and communication in decision-making and problem-solving groups: an expanded view”, in Hirokawa, R.Y. and Poole, M.S. (Eds), Communication and Group Decision Making, Sage Publications, Thousand Oaks, CA, pp. 55-80. Guetzkow, H. and Gyr, J. (1954), “An analysis of conflict in decision-making groups”, Human Relations, Vol. 7, pp. 367-81. Guzzo, R.A. and Shea, G.P. (1992), “Group performance and intergroup relations in organizations”, in Dunnette, M.D. and Hough, L.M. (Eds), Handbook of Industrial and Organizational Psychology, Vol. 3, Consulting Psychology Press, Palo Alto, CA, pp. 269-313. Hofstede, G. (1983), “The cultural relativity of organizational practices and theories”, Journal of International Business Studies, Vol. 14 No. 2, pp. 75-90. Hofstede, G. (1991), Cultures and Organizations: Software of the Mind, McGraw-Hill, London. James, L.R., Demaree, R.J. and Wolf, G. (1984), “Estimating within-group interrater reliability with and without response bias”, Journal of Applied Psychology, Vol. 69, pp. 85-98. James, L.R., Demaree, R.J. and Wolf, G. (1993), “RWG: an assessment of within-group interrater agreement”, Journal of Applied Psychology, Vol. 78, pp. 306-9. Jehn, K.A. (1994), “Enhancing effectiveness: an investigation of advantages and disadvantages of value-based intragroup conflict”, International Journal of Conflict Management, Vol. 5 No. 3, pp. 223-38. Jehn, K.A. (1995), “A multimethod examination of the benefits and detriments of intragroup conflict”, Administrative Science Quarterly, Vol. 40, pp. 256-82. Jehn, K.A. (1997), “A qualitative analysis of conflict types and dimensions in organizational groups”, Administrative Science Quarterly, Vol. 42 No. 3, pp. 530-57. Jehn, K.A. and Chatman, J.A. (2000), “The influence of proportional and perceptual conflict composition on team performance”, International Journal of Conflict Management, Vol. 11 No. 1, pp. 56-73. Jehn, K.A., Northcraft, G.B. and Neale, M.A. (1999), “Why differences make a difference: a field study of diversity, conflict, and performance in workgroups”, Administrative Science Quarterly, Vol. 44 No. 4, pp. 741-63. Kabanoff, B. (1991), “Equity, equality, power and conflict”, Academy of Management Review, Vol. 16 No. 2, pp. 416-41. Kramer, R.M. (1991), “Intergroup relations and organizational dilemmas: the role of categorization processes”, in Cummings, L.L. and Staw, B.M. (Eds), Research in Organizational Behavior, JAI Press, Greenwich, CT, pp. 191-228. Kuhn, T. and Poole, M.S. (2000), “Do conflict management styles affect group decision making? Evidence from a longitudinal study”, Human Communication Research, Vol. 26 No. 4, pp. 558-90.

Effects of intragroup conflict 243

JMP 20,3/4

244

Lewicki, R.J., Weiss, S.E. and Lewin, D. (1992), “Models of conflict, negotiation and third-party intervention: a review and synthesis”, Journal of Organizational Behavior, Vol. 13, pp. 209-52. Marks, M.A., Mathieu, J.E. and Zaccaro, S.J. (2001), “A temporally based framework and taxonomy of team processes”, Academy of Management Review, Vol. 26 No. 3, pp. 356-76. Pelled, L.H., Eisenhardt, K.M. and Xin, K.R. (1999), “Exploring the black box: an analysis of work group diversity, conflict, and performance”, Administrative Science Quarterly, Vol. 44, pp. 1-28. Peterson, R.S. and Beharf, K.J. (2003), “The dynamic relationship between performance feedback, trust, and conflict in groups: a longitudinal study”, Organizational Behavior and Human Decision Processes, Vol. 92, pp. 102-12. Putnam, L.L. (1994), “Productive conflict: negotiation as implicit coordination”, International Journal of Conflict Management, Vol. 5, pp. 285-99. Schweiger, D., Sandberg, W. and Ragan, J. (1986), “Group approaches to improving strategic decision making”, Academy of Management Journal, Vol. 29, pp. 51-71. Schwenk, C.R. (1982), “Effects of inquiry methods and ambiguity tolerance on prediction performance”, Decision Science, Vol. 13, pp. 207-21. Schwenk, C.R. (1990), “Conflict in organizational decision making: an exploratory study of its effects in for-profit and not-for-profit organizations”, Management Science, Vol. 36 No. 4, pp. 436-48. Schwenk, C.R. and Cosier, R.A. (1980), “Effects of the expert, devil’s advocate and dialectical inquiry methods on prediction performance”, Organizational Behavior and Human Performance, Vol. 26, pp. 409-23. Simons, T.L. and Peterson, R.S. (2000), “Task conflict and relationship conflict in top management teams: the pivotal role of intragroup trust”, Journal of Applied Psychology, Vol. 85 No. 1, pp. 102-11. Sosik, J.J. and Jung, D.I. (2002), “Work-group characteristics and performance in collectivistic and individualistic cultures”, The Journal of Social Psychology, Vol. 142 No. 1, pp. 5-23. Steiner, I.D. (1972), Group Processes and Productivity, Academic Press, New York, NY. Weldon, E., Jehn, K.A. and Pradham, P. (1991), “Processes that mediate the relationship between a group goal and improved group performance”, Journal of Personality and Social Psychology, Vol. 61 No. 4, pp. 555-69.

The Emerald Research Register for this journal is available at www.emeraldinsight.com/researchregister

The current issue and full text archive of this journal is available at www.emeraldinsight.com/0268-3946.htm

Affective responses to work process and outcomes in virtual teams Effects of communication media and time pressure

Communication media and time pressure 245

Amparo Caballer, Francisco Gracia and Jose´-Marı´a Peiro´ University of Valencia, Valencia, Spain Abstract Purpose – To analyze the direct and combined effects of the communication media and time pressure in group work on the affective responses of team members while performing intellective tasks Design/methodology/approach – A laboratory experiment was carried out with 124 subjects working in 31 groups. The task performed by the groups was an intellective one. A 2 £ 3 factorial design with three media (face-to-face, video-conference, and e-mail) and time pressure (with and without time pressure) was used to determine the direct and combined effects of these two variables on group members’ satisfaction with the process and with the results, and on members’ commitment with the decision. Findings – Results show a direct effect of communication media on satisfaction with the process, which confirms the prediction of the media-task fit model, and a negative effect of time pressure on satisfaction with group results and commitment to those results. Most interestingly, the interaction effects for the three dependent variables are significant and show that the most deleterious effects of time pressure are produced in groups working face-to-face, while groups mediated by video-conference improve their affective responses under time pressure. Research limitations/implications – Some limitations are the use of a student sample, so generalizability of the findings is limited, and the use of only one task type. Practical implications – It can help one to know how to design work to improve satisfaction and implication of workers. Originality/value – This paper shows some innovations as the combined effects of media and time pressure, controlling for the task type on group members’ affective responses to their work and achievements. Keywords Communication technologies, Team working Paper type Research paper

Introduction Work in organizations is often accomplished in groups. In fact, effective group performance has a clear impact on organizational results and outcomes. In recent times, group work has experienced important changes due to multiple factors (McGrath and Hollingshead, 1994). Information and communication technologies have had a strong influence on group work, by reducing time and space constraints and creating new work settings, demands and ways of operating (Gibson and Cohen, 2003). This research has been founded by the National Research Program on Science and Technology (SEC2001/3509). It has been carried out by members of the group I þ D þ I (Group 03/195) founded by the Regional Government of the Valencia Region (Generalidad Valenciana).

Journal of Managerial Psychology Vol. 20 No. 3/4, 2005 pp. 245-260 q Emerald Group Publishing Limited 0268-3946 DOI 10.1108/02683940510589037

JMP 20,3/4

246

Work performed by groups also involves diverse goals and tasks. Groups often deal with negotiations, quality issues, projects, problem solving and a large array of physical and intellective tasks (Galegher et al., 1990). Finally, time influences work group and team processes in several ways. One pervasive influence is the existence of deadlines and time constraints that often result in time pressure for employees. In fact, time stress is experienced in many organizations where decisions must be made with deadlines in mind (Brown and Miller, 2000). Although research on technology, tasks and time in groups has been carried out during the last few decades, very few studies have analyzed the combined effects of these sets of variables on satisfaction of group members. The aim of the present study is to identify the independent and combined effects of communication media and time pressure, when groups perform intellective tasks, on members’ affective responses to group processes and outcomes. To achieve this goal, first we will review the evidence available on group and teamwork mediated by new information and communication technologies and the influence of these media on group and individual outcomes. We will also review the technology – task fit and its implications for group processes and outcomes. Finally, we will direct our attention to the relationship between time pressure and affective responses of team members working in face-to-face (FTF) and new information and communication-mediated groups. Teamwork mediated by new information and communication technology (NICT) Computer mediated communication (CMC) poses new opportunities and challenges for interpersonal communication and decision-making (Thatcher and De la Cour, 2003). Using asynchronous CMC, “people are no longer restricted to a physical presence requirement to engage in the communication process” (Olaniran, 1996, p. 24). In the past decade, communication media have generated research comparing face-to-face (FTF) interaction with CMC. Several theories, such as social presence (Short et al., 1976), media richness (Daft and Lengel, 1984, 1986), social information theory (Whalter, 1996) and adaptive structuration theory (Poole, 1991; Poole et al., 1991), have attempted to explain differences between media in processes and outcomes. Many empirical studies have identified significant differences. Most have focused on group effectiveness; however, socio-emotional processes and affective responses, such as satisfaction, have also received attention. The study of satisfaction and affective outcomes captures important factors for predicting the use of ICT. Regarding satisfaction, Straus (1997) found that the communication media had a direct effect on the satisfaction of group members when they were working on a judgment task. CMC groups were less satisfied than FTF groups. Other studies also reported that groups using new information technology were less satisfied with the process and the outcomes than FTF groups, despite a lack of differences in performance quality (Hollingshead et al., 1993; Straus, 1996; Thatcher and De la Cour, 2003). Baltes et al. (2002), in a meta-analysis of 27 studies, found that CMC groups reported decreased member satisfaction when compared with FTF groups. Nevertheless, the authors point out that the effect size found was not homogeneous, which may indicate the existence of moderators, such as type of task, anonymity and time pressure. Olson and Olson (2003), in their revision on group decision support systems (GDSS), concluded that their use generated more ideas from participants and helped to ensure equal participation. However, decisions were not rated as satisfying, and the meetings took longer than

normal ones (Kraemer and Pinsonneault, 1990; McLeod, 1992; Hollingshead et al., 1993). A similar conclusion about satisfaction is reached by the authors with regard to video-conferencing. Nevertheless, some studies have found no significant differences across media on either satisfaction with the process or satisfaction with group outcomes (Huang, 2003; Straus and McGrath, 1994).

Communication media and time pressure

Information and communication technology (ICT) – task fit Several authors have pointed out that, in fact, group members’ satisfaction working with different media depends partly on the type of task. They follow the rationale developed by McGrath and Hollingshead (1994), when these authors point out that tasks, as well as communication media, vary in their “richness” (Daft and Lengel, 1986) and in the social presence required/permitted (Short et al., 1976) for an effective performance. McGrath (1984), in his circumplex model, distinguished between four main task types that differ in the degree to which effective performance on them depends only on the transmission of information among members of the group and collaboration, requires cooperation and coordination, or requires transmission of values, interests, personal commitment and the like. McGrath (1984), in his circumplex model, distinguished between four main task types that differ in the degree to which effective performance on them depends only on the transmission of information among members of the group and collaboration. At the same time, collaboration depends on type of information needed, if requires cooperation and coordination, or if requires transmission of values, interests, personal commitment and the like. As tasks differ in their complexity, they require different degrees of media richness for their execution. McGrath and Hollingshead (1994) formulated the task-media fit model. This model presents a four-by-four space defined in terms of the four task types (idea generation tasks, intellective tasks, judgment tasks and mixed-motive tasks) and the four media forms (CMC systems, audio systems, video systems and FTF systems). Therefore, the effectiveness of a media system for a task will vary depending on the fit between the richness of information that can be transmitted via the system’s technology and the information richness requirements for the task’s performance. For intellective and judgmental tasks, the model hypothesizes that audio systems and video systems are better fitted than CMC or FTF. The model has received only limited support as far as group effectiveness is concerned (Menecke et al., 2000; Tan et al., 1999; Zornoza et al., 2002). Empirical evidence is also mixed for group members’ affective responses. Straus and McGrath (1994) found support for the predictive power of the model for tasks that require low coordination. They reported that CMC groups experienced somewhat higher satisfaction than FTF groups did. However, these authors did not find differences in satisfaction between groups performing an intellective task in FTF and those performing it in CMC. Huang (2003), in a study where an intellective task and a preference task were used to elicit a good fit and a poor fit, respectively, did not find support for the model with regard to members’ satisfaction with group process and group results. “The good fit for an intellective task failed to increase group satisfaction, relative to the poor fit for a preference task” (p. 17). Given these non-conclusive results, more research is needed. In the present study, based on the task-fit model, we expect:

247

H1. Satisfaction with the group process and results and members’ commitment to the results when groups perform intellective tasks will be higher in groups mediated by video-conferencing than in those working FTF or via CMC.

JMP 20,3/4

248

Time pressure and affective responses of group members The interplay between task type and media to determine performance effectiveness in groups and the satisfaction of their members is complex, and it is also influenced by the time available for the group to perform the tasks. Straus (1997) pointed out that CMC groups, when they focus more on the tasks, this task focus, rather than indicating greater involvement in the task or reduced social concern, may be due to the fact that the groups communicate slowly and, therefore, are under greater time pressure than FTF groups. Issues regarding time effects should be differentiated in future research by manipulating both communication modes and time constraints. Time constraint has been shown to have deleterious effects on both decision-making processes and outcomes (Chu and Spires, 2001). It is usually regarded as a source of stress, especially in organizations where decisions must be reached with deadlines in mind (Argote et al., 1989; Brown and Miller, 2000; Gladstein and Reilly, 1985). Nevertheless, “time stress is often due not so much to strict deadlines as it is to the potential opportunity-cost of delaying decisions” (Payne et al., 1996, p. 131). Isenberg (1981) and Kelly and McGrath (1985) examined communication patterns among group members under varying conditions of time pressure and stress. They found that, as pressure increased, members’ participation was more disproportionate, and discussion time was more unequally shared. In addition, limited time for processing information needed to perform the task increased the task focus of the group and reduced behaviors oriented toward fulfilling social functions. Kruglanski and Freund (1983) suggested that time pressure induces closing of the mind. In fact, under this condition people seek cognitive closure, they stop considering multiple alternatives, they use poor processing of information, and they refrain from critical probing of a given seemingly adequate solution (De Dreu, 2003). Keeping all this in mind, it can be expected that excessive task focus produced by time pressure is likely to have a negative impact on satisfaction with the group process, as it reflects an imbalance in instrumental and expressive concerns. Furthermore, less equal participation induced by time pressure will be negatively associated with satisfaction with the group process, satisfaction with the result obtained and members’ commitment to these results. As the opportunities to express one’s opinion or to participate in group decision-making are reduced, satisfaction and commitment will deteriorate. In addition, Chu and Spires (2001) found that subjects, when using a decision support system under time constraint, reported less confidence in their decisions than did unconstrained participants. Thus, the commitment to the decision will also be lower. Based on this rationale and empirical evidence we expect: H2. Group members’ satisfaction with the group process, their satisfaction with the results and their commitment to these results will be significantly lower under conditions of time pressure. Combined effects of communication media and time pressure in mediated groups on affective responses As Brown and Miller (2000) have pointed out: Despite the prevalence of time pressure, however, few studies have investigated its effects on group processes . . . and the research done is largely non-cumulative and conceptually unintegrated (p. 133).

This is especially true when time pressure is researched in groups working with different media. Several confounding factors make the results meaningless or difficult to interpret. Baltes et al. (2002) provide a good example. In a meta-analysis of research, the authors compare satisfaction with the group decision process in FTF and CMC groups. Results suggest that CMC groups with unlimited discussion time showed less decrease in member satisfaction (d ¼ 20:36), compared to FTF groups, than computer mediated groups with limited discussion time (d ¼ 20:66). However, these results need to be interpreted with caution. As the authors point out in a footnote: The time limits in many studies seemed to be so long that no group would have a problem meeting the deadline . . . In other words, time limitation may not be an adequate operationalization of the perception of time pressure (pp. 169-170).

In fact, differences in the way and the speed of the communication in different media (e.g. typing) influence the meaning of time limits in each medium. So, appropriate actions are needed to create conditions of time pressure for groups working with a specific medium. Brown and Miller (2000) suggested that groups working under stressful conditions tend to place their decision in the hands of fewer members of the group. Moreover, they operate in a relatively centralized communication network with an unequal allocation of discussion time and participation. Furthermore, they tend to focus more on the tasks than in a no-pressure condition, greatly reducing time available for social functions. On the other hand, according to the theory of social presence, CMC communication contributes to promoting equalization of the participation, and it filters social cues. Thus, mediated communication will mitigate the unequal allocation of discussion time and centralization under time pressure. Therefore, satisfaction with the process and the result, as well as commitment to the result in groups working with ICT, will not experience a strong reduction under these circumstances. Moreover, as the space for social functions is also more limited in CMC, the impact of time pressure on the reduction of this function will also be smaller than in FTF, and time pressure will have a lower negative impact on process satisfaction. Thus, groups mediated by new ICT should experience less negative effects of time pressure on the affective responses of their group members than those working FTF. Following this rationale, we expect: H3. Negative effects of time pressure on affective responses of group members will be more deleterious in FTF than in mediated groups (video-conference and e-mail). Method Participants The sample in this study consisted of 124 subjects, all of whom were graduate students of psychology. The majority of the participants were female (n ¼ 101, 81 percent), and the average age was 21.3 years (SD 2.17). The subjects were randomly placed in 31 groups of four members each. Technology and equipment Three different types of equipment were used, one for each communication channel: face-to-face, e-mail and video-conference.

Communication media and time pressure 249

JMP 20,3/4

250

In the face-to-face condition, members of the group were put together in the same room and could use a computer only to perform the task. Their meetings were recorded by video cameras. E-mail was conducted using the program QUICKMAIL (Apple software). QUICKMAIL was installed in a local network allowing synchronous communication. This software splits a user screen into a message-receiving area at the right portion of the screen and a message-sending area at the left portion. Participants received experimental training to acquire competent use of the technology. The videoconference channel had four different workstations (Silicon Graphics R4600sc, INDICAM). In each room there was a workstation equipped with a camera, a microphone and two loudspeakers. This technology makes it possible for subjects to connect and disconnect from the network whenever they want to. The four participants were in separate rooms connected by a local network. Participants received experimental training to acquire competent use of the technology.

Design and procedure (task and time pressure) The present study is a laboratory experiment with a 2 £ 3 factorial design. The factors are time pressure (with and without time pressure conditions) and communication media (face to face, video-conference and e-mail conditions). In the experiment, ten groups worked in a face-to-face condition, ten groups worked in the e-mail condition, and 11 groups worked in the video-conference condition. Participants working in the e-mail and video-conference conditions received training before the experiment on using the system correctly, and they had several practice sessions in order to become familiar with the work method. Half of the groups in every condition worked under time pressure, and the other half worked without a time limit. Participants were randomly assigned to the experimental conditions (communication media and time pressure). Time pressure. In order to operationalize time pressure in groups working in every medium, it is necessary to significantly reduce the amount of time needed to perform the task in this medium. Benson and Beach (1996) noted that many researchers select time constraint levels arbitrarily, with no rationale for their decision. With this procedure, the time limit is fixed, but time constraint is not assured. One way to prevent this situation is that subjects without time pressure perform the task first. Second, we obtain the mean execution time and finally, we reduce that mean one standard deviation below. The result will be the time assigned to the groups with time pressure. If times are normally distributed, this forces about 84 percent of the subjects to perform the task faster than normal. However: Time pressure indicates that the time constraint induces some feeling of stress and creates a need to cope with the limited time (Ordon˜ez and Benson, 1997, p. 122).

Two conditions appear to be necessary for time constraint to function as a stressor. First, group members must understand that there is some deadline by which a task must be finished. Second, group members must believe that there is some reward to be gained or costs to be avoided. These two conditions indicate that time limitation is considered to be a stressor, due to the perceived pressure of group members rather than to the actual time limit imposed on the group.

In the present study, the groups that worked without time pressure did not have a temporal limit to finish their work, whereas the groups that worked under time pressure had a temporal limit. As suggested by Svenson and Maule (1993), we induced time pressure by giving the groups under this condition 25 percent less time than the average needed by the groups without time pressure. Therefore, the experiment was initiated with the groups in the no time pressure condition. Their average time was reduced by 25 percent, and the remaining time was set as the time limit for the groups in the time pressure condition. A second requirement for inducing time pressure is to offer a reward to the group achieving the best results (Brown and Miller, 2000). As Payne et al. (1996) pointed out, time stress is often due to the potential opportunity-cost of delaying decisions. In this study, a reward of 120 euros was given to the best group in each communication condition. Finally, as time stress is a subjective experience, it was necessary to check whether the manipulation actually produced the time pressure experience in the subjects. In order to do so, we measured the perception of time pressure. Participants used a five-point scale to respond to the questions: “there was enough time to undertake the task” (reversed score) and “I have felt time pressure completing the task”. Cronbach’s alpha is 0.79. Type of task. In each condition, groups had to resolve the same task. An intellective task (logic problem with a correct solution) was used. The task “find out the profession” required subjects to associate four different people to four different pieces of information (name, profession, department they work for and address of the company). Therefore, groups had to fill out 16 blanks, four for each person. The task required the group to develop a joint product. Each participant had only a part of the necessary information to solve the task, and only by sharing it could the task successfully be completed. The work sessions were recorded on video. After resolving the task, subjects had to complete a questionnaire containing scales we describe below.

Measures Process satisfaction. This was measured on a five-point scale by three items that were very similar to the scale from the JEMCO weekly experiment questionnaire (McGrath, 1993). The items used are the following: “how comfortable does the group feel working together?”, “. . . do you feel free enough to express disagreements?”, “. . . were you prepared to learn from others?” (Cronbach’s alpha ¼ 0:77). Result satisfaction. Two items adapted from Gruenfeld et al. (1996) were used to measure result satisfaction: “how satisfied are you with the result achieved?”, “to what extent are you confident about the group solution being adequate?” (Cronbach’s alpha ¼ 0:92). Commitment to the solution achieved. This was measured on a five-point scale by three items: “to what extent have you participated in the final solution?”, “. . . do you feel committed to the solution?”, “. . . do you feel responsible for the quality of the final solution?” (Cronbach’s alpha ¼ 0.95). Manipulation check of time pressure. The manipulation of time pressure was tested by conducting a t-test. The results revealed a significant effect of time pressure: tð1; 122Þ ¼ 10:731, p , 0.001. Participants who worked under the time pressure condition (M ¼ 2:96, SD ¼ 1:08) perceived significantly higher time pressure than participants working without time pressure (M ¼ 1:56, SD ¼ 0:62).

Communication media and time pressure 251

JMP 20,3/4

252

Results Descriptive data. Descriptive statistics, alpha coefficients and correlation coefficients among the dependent variables are presented in Table I. Significant positive correlations are found among the affective responses under study. Interestingly enough, the correlation between result satisfaction and solution commitment is the strongest (0.49) one. These two variables are more highly correlated to each other than to satisfaction with the process. The latter variable includes satisfaction with functions fulfilled in the process other than just effective performance. Significant negative correlations have been found between affective responses and perception of time pressure. These relationships show that perception of time pressure plays an important role in individual responses, especially on result satisfaction (2 0.57, significant at the 0.01 level). Hypotheses testing. The results of the ANOVA for process satisfaction, outcome satisfaction and solution commitment are presented in Table II. According to H1, a direct effect of communication media was expected for every affective response. The results revealed a significant main effect for satisfaction with the results achieved by the group but not for satisfaction with the process or commitment to the solution. Inter-group differences in the case of results satisfaction are established using the Scheffe procedure. The Scheffe procedure shows that face-to-face groups have lower result satisfaction (M ¼ 3:6) than video-conference groups (M ¼ 4:35). Thus, the hypothesis is confirmed only for results satisfaction. In H2, we expected that time pressure would have a direct negative effect on affective responses. Results show that this is the case for satisfaction with the results and commitment to the solution achieved. Time’s main effect is non-significant for process satisfaction. These results show that individuals in groups working under time pressure experience lower satisfaction with the results and lower commitment to the solution (M ¼ 3:51 for result satisfaction, M ¼ 3:51 for solution commitment) than individuals working in groups without time pressure (M ¼ 4:2 for result satisfaction, M ¼ 3:85 for solution commitment). Thus, the hypothesis is confirmed for these two affective responses. Variables

Table I. Descriptive statistics, correlations and reliability coefficients (alpha)

1. 2. 3. 4.

Process satisfaction Result satisfaction Solution commitment Time pressure perception

Source Table II. ANOVA results: process satisfaction, result satisfaction and solution commitment

Mean

SD

1

2

3

4

4.29 3.87 3.68 2.28

0.50 1.08 0.91 1.13

– 0.437** 0.384** 20.361**

– 0.492** 2 0.572**

– 20.256**

–

Result Process satisfaction satisfaction df F-value p . F F-value p . F

Communication media 2 Time pressure 1 Communication media £ time pressure 2 Notes: *p , 0.05; **p , 0.005; ***p , 0.001

1.15 1.34 5.75

0.319 0.250 0.004**

9.64 17.65 11.22

0.001*** 0.001*** 0.001***

Solution commitment F-value p . F 0.92 4.53 4.108

0.401 0.036* 0.019*

According to H3, we expected that negative effects of time pressure on group members’ affective responses would be more deleterious in FTF than in mediated groups (video-conference and e-mail). Interactive effects obtained are significant for all three dependent variables. The interactive effect of communication media and time pressure on process satisfaction is presented in Figure 1. The contrast tests show that in the non-time pressure condition, the process satisfaction when groups work face-to-face (M ¼ 4:5) reaches the highest score, and it is significantly higher than in groups working with electronic mail (M ¼ 4:16). On the other hand, groups working in the FTF condition under time pressure (M ¼ 3:98) present the lowest process satisfaction. This value is significantly lower when compared to e-mail (M ¼ 4:32) and video-conference (M ¼ 4:40). No differences are detected in process satisfaction between groups working mediated by e-mail and video-conference. Figure 2 represents the interactive effects of independent variables on result satisfaction. Post hoc analysis revealed that groups working face-to-face “without time pressure” present significantly higher satisfaction with the result (M ¼ 4:45) than groups working mediated by e-mail, also without time pressure (M ¼ 3:87). However, under “the time pressure condition” groups working face-to-face (M ¼ 2:75) show significantly lower satisfaction with the group results than groups working via e-mail (M ¼ 3:35) and video-conference (M ¼ 4:43). The interactive effects of communication media and time pressure on solution commitment are shown in Figure 3. Again, groups working face-to-face under no time pressure present higher commitment to the group decision (M ¼ 4:32). Their commitment is significantly higher than that of groups working via e-mail (M ¼ 3:68) and video-conference

Communication media and time pressure 253

Figure 1. Interactive effects of communication media and time pressure on process satisfaction

JMP 20,3/4

254

Figure 2. Interactive effects of communication media and time pressure on result satisfaction

Figure 3. Interactive effects of communication media and time pressure on solution commitment

(M ¼ 3:55). However, in the “time pressure condition” differences between groups working with different media disappear. Discussion The aim of the present study is to determine the direct and combined effects of the communication media and time pressure in group work on the affective responses of team members while performing intellective tasks. The focus is on members’ satisfaction with group work and their commitment to the decision reached. Research on this issue has paid less attention to the affective responses than to group performance and effectiveness. However, knowledge about these issues is needed both for theoretical and practical reasons. As Baltes et al. (2002) pointed out: If the use of CMC leads to significant dissatisfaction among participants in group decision-making, then there will be significant resistance within the organization to its use and to the implementation of decisions reached using the technology, regardless of the quality of the decision reached (p. 160).

So, it is important to identify the antecedents that influence satisfaction with the group processes, which fulfill instrumental and socio-emotional functions. It is also important to identify the antecedents of satisfaction with the results, keeping in mind that this experience is not totally determined by the quality of performance and results. In fact, several authors found that, despite a lack of differences in performance quality, groups using new information technology were less satisfied with the process and the outcomes than FTF groups (Hollingshead et al., 1993; Straus, 1996; Thatcher and De la Cour, 2003). Satisfaction with the result may also be influenced by the members’ participation pattern. Opportunities to express one’s opinion or to participate in group decision-making improve satisfaction with and commitment to the decisions reached by the group. A number of studies have shown that FTF groups develop higher satisfaction than mediated groups (Olson and Olson, 2003; Straus, 1997). However, Baltes et al. (2002), in their meta-analysis, pointed out that it is necessary to take into consideration several moderators to better predict the influence of media on affective responses. One important moderator is task type. In fact, the theoretical model of task-media fit (McGrath and Hollingshead, 1994), based on social presence and media richness theories, acknowledges the role played by the type of task in the understanding of the complex relationship between media and group members’ affective responses. According to this model, when the requirements of the task fit with the features of the technology, the best performance and the highest satisfaction with the process and outcomes should occur. Our study has focused on an intellective task. According to the model, video-conference is the best fitted media, and it should produce the highest affective response of group members. E-mail should provide less satisfaction because the media is poor in transmitting the information requirements of the task. FTF should also provide less satisfaction than VC because it is too rich, and it does not filter social cues and information that is irrelevant or dysfunctional for the performance of the group. This over-richness is thought to hamper equal participation. The results of the present study support the predictions formulated based on the media-task fit model in the case of satisfaction with the results. In fact, satisfaction is higher in groups working

Communication media and time pressure 255

JMP 20,3/4

256

through video-conference than in those working via e-mail or FTF. Nevertheless, it is interesting to note that FTF does not produce any advantages in satisfaction with the results over the levels reached in CMC (e-mail). The model is not supported either when it predicts satisfaction with the process or commitment to the results. For both variables, no significant differences were found between media. These results are in line with those obtained by McGrath and Hollingshead (1994) and Huang (2003). Neither of these studies found any significant differences in satisfaction between media for groups working on intellective tasks. Thus, although the media-task fit model can contribute, in some cases, to improving predictions about satisfaction of group work mediated by ICT, other moderating variables need to be taken into consideration in order to better predict its influence on members’ affective responses. Time pressure has often been shown to have an important influence on work group processes and outcomes. Nevertheless, in mediated groups the results are less clear and little is known about the role of time pressure in members’ satisfaction in groups working with different media. Baltes et al. (2002) stated that time constraint played a moderating role between media and satisfaction. Nevertheless, these authors acknowledge that the scant amount of research carried out on this issue with mediated groups did not properly operationalize time pressure, and, thus, the results are rather weak and non-conclusive. The present study has carefully designed the manipulation strategy to produce group work with and without time pressure. The results we have obtained show that satisfaction with the results and commitment to the solution reached by the group were significantly lower under time pressure conditions than in the absence of time pressure. Surprisingly, we did not find significant differences in the satisfaction with the group process. So, time pressure is perceived by the members of the group as having a deleterious effect on the results achieved by the group and members’ commitment to them, but not on group processes. One possible interpretation for these results is that FTF and mediated groups, when working on intellective tasks under stressful conditions, tend to place their decisions in the hands of fewer members of the group and operate in a relatively centralized communication network with an unequal allocation of discussion time and participation (Brown and Miller, 2000). This way of coping with time pressure would affect members’ satisfaction with the results and their commitment to those results. Nevertheless, changes produced in the groups when they worked under pressure did not deteriorate satisfaction with the work process. The results obtained when effects of media and time pressure conditions are considered together contribute to understanding these rather puzzling results. This is one of the most relevant contributions of our study because, to our knowledge, this issue has not yet been rigorously researched. Our results show a significant effect of the interaction between media and time pressure on the three affective responses studied. They confirm our hypothesis and show that time pressure has strong deleterious effects on satisfaction and commitment produced by the members of groups working FTF. It is also important to note that members of groups working via video-conference slightly improve their satisfaction and commitment under time pressure. These results provide new insight into the understanding of the media-task fit model. Under no time pressure, FTF media produces the highest satisfaction with group process and outcome and higher commitment to the decision. Video-conference produces an intermediate level of these affective reactions, lower than that produced by

FTF but higher than CMC. Under no time constraints, groups profit from the “excessive” richness of the FTF channel, using it to perform functions other than those that are instrumental to focusing on the task. Social functions are also fulfilled. However, the situation is very different under time pressure. In these circumstances, the groups most satisfied with the process and the results and the most committed are those who worked mediated by video-conference. Time pressure even improves their satisfaction when compared to the groups with the same media but with no time pressure. Moreover, it is also worth noting that they are significantly more satisfied and committed than CMC and, above all, FTF groups. Thus, under conditions of time pressure, the media-task fit model seems fully operative. Poor media produce less satisfaction than the most adequate medium because they do not provide enough richness for what is required by the task. But, at the same time, a medium allowing for excessive richness when the time for work is scarce provides too much information flow, and probably induces the group to operate in a non-optimal way, reducing equality of participation (Brown and Miller, 2000). This way of coping with time pressure reduces the effectiveness of the group and satisfaction with the process and the results and hampers the commitment to the decision. Thus, the results of the present study point out that the media-task fit does not operate in the same way under time pressure as when groups work without time limits. In the latter condition, over-rich media contribute to increasing satisfaction of group members because they allow them to fulfill socio-emotional functions in addition to the instrumental ones. However, under time pressure, groups need media that fit the richness of the task. Over-rich media are dysfunctional, not only for the performance and efficacy of the group, but also for the satisfaction of its members. The present study offers several innovations with respect to the previous literature. First, it analyzes the combined effects of media and time pressure, controlling for the task type, on group members’ affective responses to their work and achievements. It contributes to a better understanding of time pressure effects on members’ satisfaction by creating a real condition of time pressure. It extends the media-task fit model by distinguishing the non-fit situation in two different conditions: one of over-richness of the media and one of poor media. The effects of these two non-fit conditions on affective responses of group members are different depending on time pressure experience. If groups work under time pressure, the over-richness of the media clearly deteriorates satisfaction, because the “broadband” of the communication media contributes to the centralization of decisions and unequal participation, hampering the communication structure. On the other hand, media that fit the richness requirements of the task filter non-relevant information and contribute to improving participation and to preventing centralization of the decision-making in a few members, thus improving satisfaction with the process and the result and, in turn, commitment. Thus, our results support and extend a conclusion formulated in a previous study, where we tested the media task fit model by analyzing conflict management in groups working via different media. In that study, we pointed out that “assumed” over-richness of media for the task requirements did not produce a hampering effect on conflict management behavior (Zornoza et al., 2002), Now we can further clarify the effects of over-richness of the media by pointing out that their effects on affective responses of the group members differ as a function of the time pressure experience.

Communication media and time pressure 257

JMP 20,3/4

258

The present study has some limitations that require consideration. One is the use of a student sample, rather than a real-life working sample. Thus, generalizability of these findings to real work settings is limited. In fact, the present research involved ad hoc, leaderless teams of peers, whereas many teams in organizations are ongoing and have established status hierarchies. Nevertheless, real world group discussions are often short in duration, so the results could be generalized to that type of group. Second, only one task type (intellective) has been considered in the study. Thus the media-task fit model has been only partially tested. More research is needed to test the model with different tasks and media. A third limitation is the duration of the effects found in our study. The present study has been performed in a work session. Therefore, questions can be raised about the stability of the affective responses and of the combined effects of media and time pressure that have been identified in the present study. To overcome some of these limitations, more research is needed in several directions. First, it is important to test the media-task fit model, with and without time pressure, by using other types of tasks considered in the model. Second, it would be interesting to integrate the results from this study with the results obtained in previous studies that took into account other variables. For instance, Gracia et al. (2000) and Gracia et al. (2002) have studied the role of time pressure, task type and communication media on group performance quality and group cohesiveness, pointing out the importance to investigate these variables in the study of work group outcomes. Third, group processes need to be explicitly considered in the theoretical models, as established by the adaptive structuration theory (Poole, 1991; Poole et al., 1991). In addition, longitudinal studies must try to determine whether some of the effects disappear over time, due to the capacity of the groups to adapt to the requirements of different media (Whalter, 1996). However, dynamic interaction processes between media tasks and groups’ behaviors display a large array of possible combinations, and theoretical development is needed to formulate models useful for clarifying all these complexities of group work. (Zornoza et al., 2002). References Argote, L., Turner, M.E. and Fichman, M. (1989), “To centralize or not to centralize? The effects of uncertainty and threat on group structure and performance”, Organizational Behavior and Human Decision Processes, Vol. 43, pp. 58-74. Baltes, B.B., Dickson, M.W., Sherman, M.P., Bauer, C.C. and LaGanke, J.S. (2002), “Computer-mediated communication and group decision making: a meta-analysis”, Organizational Behaviour and Human Decision Processes, Vol. 87 No. 1, pp. 156-79. Benson, L. III and Beach, L. (1996), “The effects of time constraints on the pre-choice screening of decision options”, Organizational Behavior and Human Decision Processes, Vol. 67, pp. 222-8. Brown, T.M. and Miller, C.E. (2000), “Communication networks in task-performing groups: effects of task complexity, time pressure and interpersonal dominance”, Small Group Research, Vol. 31 No. 2, pp. 131-57. Chu, P.C. and Spires, E.E. (2001), “Does time constraint on users negate the efficacy of decision support systems?”, Organizational Behaviour and Human Decision Processes, Vol. 85, pp. 226-49. Daft, R.L. and Lengel, R.H. (1984), “Information richness: a new approach to managerial behavior and organizational design”, Research in Organizational Behaviour, Vol. 6, pp. 191-233.

Daft, R.L. and Lengel, R.H. (1986), “A proposed integration among organizational information requirements, media richness and structural design”, Management Science, Vol. 32, pp. 554-671. De Dreu, C.K.W. (2003), “Time pressure and closing of the mind in negotiation”, Organizational Behavior and Human Decision Processes, Vol. 91, pp. 280-95. Galegher, J., Kraut, R. and Egido, C. (Eds) (1990), Intellectual Teamwork: Social and Technological Foundations of Cooperative Work, Lawrence Erlbaum, Hillsdale, NJ. Gibson, C.B. and Cohen, S.G. (Eds) (2003), Virtual Teams that Work: Creating Conditions for Effective Virtual Teams, Jossey Bass, New York, NY. Gladstein, D. and Reilly, N. (1985), “Group decision making under threat: the tycoon game”, Academy of Management Journal, Vol. 28, pp. 613-27. Gracia, F.J., Arcos, J.L. and Caballer, A. (2000), “Time pressure influence on group work by task type and communication channel”, Psicothema, Vol. 12 No. 2, pp. 241-6. Gracia, F.J., Caballer, A. and Peiro´, J.-M. (2002), “Effects of time pressure on group cohesiveness in different task types and communication media”, Psicothema, Vol. 14 No. 2, pp. 434-9. Gruenfeld, D.H., Mannix, E.A., Williams, K.Y. and Neale, M.A. (1996), “Group composition and decision making: how member familiarity and information distribution affect process and performance”, Organizational Behavior and Human Decision Processes, Vol. 67 No. 1, pp. 1-15. Hollingshead, A.B., McGrath, J.E. and O’Connor, K.M. (1993), “Group task performance and communication technology: a longitudinal study of computer mediated vs face-to-face work groups”, Small Group Research, Vol. 24, pp. 307-33. Huang, W.W. (2003), “Impact of GSS generic structures and task types on group communication process and outcome: some expected and unexpected research findings”, Behaviour and Information Technology, Vol. 22, pp. 17-29. Isenberg, D.J. (1981), “Some effects of time-pressure on vertical structure and decision-making accuracy in small groups”, Organizational Behavior and Human Performance, Vol. 27, pp. 119-34. Kelly, J.R. and McGrath, J.E. (1985), “Effects of time limits and task types on task performance and interaction in four-person groups”, Journal of Personality and Social Psychology, Vol. 49, pp. 395-407. Kraemer, K.L. and Pinsonneault, A. (1990), “Technology and groups: assessment of empirical research”, in Galegher, J., Kraut, R. and Egido, C. (Eds), Intellectual Teamwork: Social and Technological Foundations of Cooperative Work, Lawrence Erlbaum, Hillsdale, NJ, pp. 373-405. Kruglanski, A.W. and Freund, T. (1983), “The freezing and unfreezing of lay-inferences: effects of impressional primacy, ethnic stereotyping, and numerical anchoring”, Journal of Experimental Social Psychology, Vol. 19, pp. 448-68. McGrath, J.E. (1984), Groups: Interaction and Performance, Prentice-Hall, Englewood Cliffs, NJ. McGrath, J.E. (1993), “Introduction: the JEMCO workshop – description of a longitudinal study”, Small Group Research, Vol. 24, pp. 285-306. McGrath, J.E. and Hollingshead, A.B. (1994), Groups Interacting with Technology, Sage, London. McLeod, P.L. (1992), “An assessment of the experimental literature on electronic support of group-work: results of a meta-analysis”, Human-Computer Interaction, Vol. 7, pp. 257-80. Menecke, B.E., Valancich, J.S. and Wheeler, B.C. (2000), “The effects of media and task on user performance: a test of the task-media fit hypothesis”, Group Decision and Negotiations, Vol. 9, pp. 507-29.

Communication media and time pressure 259

JMP 20,3/4

260

Olaniran, B.A. (1996), “A model of group satisfaction in computer-mediated communication and face-to-face meetings”, Behaviour and Information Technology, Vol. 15, pp. 24-36. Olson, G.M. and Olson, J.S. (2003), “Human-computer interaction: psychological aspects of the human use of computing”, Annual Review of Psychology, Vol. 54, pp. 419-516. Ordon˜ez, L. and Benson, L. III (1997), “Decisions under time pressure: how time constraint affects risky decision making”, Organizational Behavior and Human Decision Processes, Vol. 71, pp. 121-40. Payne, J.W., Bettman, J.R. and Luce, M.F. (1996), “When time is money: decision behavior under opportunity-cost time pressure”, Organizational Behavior and Human Decision Processes, Vol. 66, pp. 131-52. Poole, M.S. (1991), “Procedures for managing meetings: social and technological innovation”, in Swanson, R. and Knapp, B. (Eds), Innovation Meeting Management, 3M Meeting Management Institute, Austin, TX, pp. 53-100. Poole, M.S., Holmes, M. and DeSanctis, G. (1991), “Conflict management in a computer-supported meeting environment”, Management Science, Vol. 37, pp. 926-53. Short, J.A., Williams, E. and Christie, B. (1976), The Social Psychology of Telecommunication, Wiley, London. Straus, S.G. (1996), “Getting a clue: the effects of communication medium and information distribution on participation and performance in computer-mediated and face-to-face groups”, Small Group Research, Vol. 27 No. 1, pp. 115-42. Straus, S.G. (1997), “Technology, group process, and group outcomes: testing the connections in computer-mediated and face-to-face groups”, Human-Computer Interaction, Vol. 12, pp. 227-66. Straus, S.G. and McGrath, J.E. (1994), “Does the medium matter? The interaction of task type and technology on group performance and member reaction”, Journal of Applied Psychology, Vol. 79, pp. 87-97. Svenson, O. and Maule, A.J. (Eds.) (1993), Time Pressure and Stress in Human Judgement and Decision Making, Plenum, New York, NY. Tan, C.Y., Wei, K.K., Sia, C.L. and Raman, R.S. (1999), “A partial test of the task-medium fit proposition in a group support system environment”, ACM Transactions on Human-Computer Interaction, Vol. 6, pp. 47-66. Thatcher, A. and De la Cour, A. (2003), “Small group decision making in face-to-face and computer-mediated environments: the role of personality”, Behaviour and Information Technology, Vol. 22, pp. 203-18. Whalter, J.B. (1996), “Computer-mediated communication: impersonal, interpersonal and hyperpersonal interaction”, Communication Research, Vol. 23, pp. 3-43. Zornoza, A., Ripoll, P. and Peiro´, J.-M. (2002), “Conflict management in groups that work in two different communication contexts: face-to-face and computer-mediated communication”, Small Group Research, Vol. 33 No. 5, pp. 481-508.

The Emerald Research Register for this journal is available at www.emeraldinsight.com/researchregister

The current issue and full text archive of this journal is available at www.emeraldinsight.com/0268-3946.htm

Effects of task interdependence and type of communication on performance in virtual teams Ramo´n Rico

Effects of task interdependence

261

Universidad Auto´noma de Madrid, Madrid, Spain, and

Susan G. Cohen University of Southern California, Los Angeles, California, USA Abstract Purpose – To investigate the effects of within-group task interdependence and the degree of communications synchrony on performance in virtual teams (VT). Design/methodology/approach – A 2 £ 2 factorial design of 240 participants in Spain, randomly assigned to 80 three-person teams, was used. Teams worked virtually (not meeting face-to-face), performing a merit-rating task in a laboratory setting. Findings – The analyses revealed an interaction effect between task interdependence and synchrony of communication. High values of VT performance were found both under conditions of “low task interdependence” and “asynchrony of communication” and under conditions of “high task interdependence” and “synchrony of communication”. The results show that superior VT performance is contingent on the match between the nature of the task and the choice of communications modality. Research limitations/implications – First, additional research will be needed to confirm and extend the findings in actual working environments. Second, a closer look is necessary at the different mediation processes employed by teams in situations where there is a better or worse fit between task characteristics and type of communication (e.g. team strategy). Practical implications – This study makes it clear that it is necessary to optimize and nurture one’s investments in communication technologies, and calls for further consideration of the requirements for the design of technological solutions in accordance with task interdependence. Originality/value – This work complements past research that has focused mainly on virtual teams using asynchronous technologies or comparing them with face-to-face teams. Keywords Team working, Team performance, Spain Paper type Research paper

Work is changing. Indeed, one might say it is evolving away from what we called the post-industrial era to today’s knowledge-based society. The effects of these changes are being felt at different levels. People’s relationship with work is gradually changing from the need to use tools to a need to respond to the unexpected. Firms are faced with demanding competitive challenges due to their immersion in global markets, the increasing complexity of the environments in which they operate, and the pace of organizational change (Mohrman et al., 1998; Cascio, 2000; Duarte and Tennant, 2001; Cohen and Gibson, 2003). The accelerating development of new information technologies may be a cause or consequence, and at the very least accompanies these organizational changes. The use of new information technologies generate opportunities and promising solutions to the need for flexibility in organizations.

Journal of Managerial Psychology Vol. 20 No. 3/4, 2005 pp. 261-274 q Emerald Group Publishing Limited 0268-3946 DOI 10.1108/02683940510589046

JMP 20,3/4

In this context, virtual teams have become a source of competitive advantage, due to the spread and growing importance of teams in modern society (Cohen and Bailey, 1997) and their key role in the design of our organizations (Mohrman et al., 1995). Virtual teams are above all teams or working groups. However, they interact via electronic communication systems. The members of these teams do not tend to meet in the conventional sense; they are scattered (Bell and Kozlowski, 2002) and membership itself is changeable, adapting to the shifting requirements of the tasks and projects undertaken (Townsend et al., 1996). The increasing popularity of virtual teams highlights two related issues. First, the use of these work units in organizations is far ahead of research on the subject, and second, as suggested by Montoya-Weiss et al. (2001) we may enrich our knowledge of how virtual teams (VTs) work using what we already know about non-virtual teams. However, we need to remain aware of the opportunities and limitations that virtuality imposes on working teams. Numerous scholars have contributed to the growing research on the effectiveness of VTS. Key findings include: . how team building process are improved by the structuration of activities with a goal-setting structure (Huang et al., 2003) or temporal coordination mechanisms (Montoya-Weiss et al., 2001); . how virtual teams are requiring a kind of “distal” leadership (Bell and Kozlowski, 2002) that requires from managers delegation and facilitation skills; and . how the type of task that the team is carrying out influences virtual team performance (Straus and McGrath, 1994; Hollingshead and McGrath, 1995). .

262

Many of them point out, however that we need to discover more about the interaction between communications technologies and the kind of task carried out by a team (Bell and Kozlowski, 2002; Maznewski and Chudoba, 2000; Maznewski and Atanassiou, 2003; Montoya-Weiss et al., 2001). Pairing up the study of these concepts might seem a mere fancy were it not for the consistent relationship between the type of task and the processes used and results obtained by teams (Goodman, 1986; McGrath, 1984; Guzzo and Shea, 1992; Levine and Moreland, 1990; Cohen and Bailey, 1997) and the necessary use of communications technologies for virtual teams to interact at all. These technologies indeed permit the dispersion of teams in space and time, while remaining a key element of their processes (Bell and Kozlowski, 2002; Olson and Olson, 2003). In this paper, we report on an experiment to test how the fit between the characteristics of the task, in particular task interdependence, and communication synchrony influences VT performance. Task interdependence Teamwork involves interaction with other team members, with whom the individual maintains a given level of interdependence. Work must, therefore, be interactive to complete a set of tasks (Stewart and Barrick, 2000) and share information, knowledge or materials in order to achieve the desired outcomes (Susman, 1976; Cummings, 1978). Interdependence in the task thus implies the degree to which team members will interact and depend upon each other to attain their goals (Campion et al., 1993).

Studies of task interdependence go back a long way. To follow the proposals of Wageman (1995) and Stewart and Barrick (2000), we may identify two main approaches. The first treats task interdependence as the product of technological requirements (Thompson, 1967; Van de Ven et al., 1976). The second conceptualizes interdependence as the outcome of perceived member needs to cooperate with each other (Shea and Guzzo, 1987; Van der Vegt et al., 1998). We agree with Wageman (1995) that task interdependence is a structural characteristic and at the same time, that the same task may be carried out with differing levels of interdependence. In this light, our approach in this paper concentrates on the extent to which instructions and the context of task performance are capable of affecting the degree of interdependence between the members of a team as they work. Type of communication technology in virtual teams If interdependence is the “glue” that holds conventional teams together, communication technologies serve as the bond linking the members of virtual teams. They allow team members to communicate and share data and information despite disparities in location and time-zone. In this way, they become the key channel for interaction in virtual teams (Bell and Kozlowski, 2002). E-mail, web-based repositories of shared knowledge, group calendars, newsgroups, instant messaging, chat, electronic whiteboards and videoconferencing tools are just some of the plethora of technological options and alternatives available to channel communication between organizational formations of this kind. Communication technologies are defined by the three dimensions of space, time and the level of support they provide to the group activity (Warkentin et al., 1997). Considering only the first two, however, these alternatives may be classified in terms of the synchrony of communication, which is the degree to which communication technologies allow teams to work together in the same space and time (Montoya-Weiss et al., 2001). The attention given to this asynchrony-synchrony continuum has been somewhat uneven. In the first place, this is partly because of the widespread stereotype of virtual teams as structures that allow work to take place continuously around the globe (Lipnack and Stamps, 2001). Second, most research has been aimed at measuring the effects of electronic communication has compared face-to-face teams without communication technology with those using asynchronous forms of computer-mediated communication (Bell and Kozlowski, 2002; Maznewski and Chudoba, 2000). Thus, a number of variables have been identified to take into account the limitations electronic communications impose on virtual teams (McGrath and Hollingshead, 1994) and measure how these may impact on performance (Warkentin et al., 1997). Face-to-face communication is an orderly process in which verbal and non-verbal cues offer feedback, facilitate turn taking and transmit subtle shades of meaning. Asynchronous communication, however, presents numerous difficulties, erasing many of the cues that permit regular intercourse and feedback: interruptions are frequent, pauses are long and information overload may arise (McGrath, 1991; Ocker et al., 1998). A number of studies show that communication is less efficient in virtual than in face-to-face teams (McGrath and Hollingshead, 1994; Straus, 1996). They reveal how

Effects of task interdependence

263

JMP 20,3/4

264

difficult it may be for team members to remain aware of each other’s presence (Olson and Olson, 2003) and highlight the obstacles to the development and upkeep of trust relationships (Jarvenpaa et al., 1998; Rocco et al., 2000). In short, electronic communication poses a number of problems due to the difficulty of eliciting social cues and constructing the kind of relational ties that smooth interaction (McGrath, 1990; Warkentin et al., 1997; Maznewski and Chudoba, 2000). Therefore, it follows that face-to-face teams would outperform virtual teams, and by extension, that VTS using synchronous as opposed to asynchronous communication technologies would also be more effective. A recent review (Maznewski and Chudoba, 2000), however, reveals some disagreement in this regard. While the results of some studies support the hypothesis, others reflect better performance in virtual than in face-to-face teams, and a final group appears to suggest that there is no difference between the two types of teams. Joint effects of task interdependence and type of communication technology on virtual team performance One straightforward way to resolve the confusion arising from such contradictory empirical evidence relating technological influence on virtual teams and their performance is to consider the role played by the type of task undertaken by the team (O’Connor et al., 1993; Hollingshead et al., 1993; Hollingshead and McGrath, 1995). A recent review (Hollingshead and McGrath, 1995) uses the typology proposed by McGrath (1984) to classify the possible tasks that a group may undertake into four classes: generate, choose, negotiate and execute. This work shows that virtual teams using asynchronous communication perform better than face-to-face teams in generation tasks (e.g. idea generation). In the case of intellective or negotiation tasks, however, face-to-face, and therefore synchronous, teams perform better than VTS using asynchronous communication. Furthermore, it has been noted that the effects of communication technologies on teams may be largely a consequence of the manner in which the task is structured in response to the constraints of a given technology (Watson et al., 1988; Hollingshead and McGrath, 1995). This consideration unequivocally points to interdependence as a structural feature of relations between the team members (Van der Vegt et al., 1998). In this light, we may ask what are the roles played by task interdependence and the type of communication technologies used in the final performance of virtual teams? In a recent study, Maznewski and Chudoba (2000) found that tasks requiring greater interdependence in the most effective teams were commonly associated with a higher frequency of communication between team members, a higher level and complexity of decision-making processes, and richer communication channels. Also, Turoff et al. (1993) reported that the amount of interaction the communication technologies must support relates to the amount of interdependence required by a task. As we have already mentioned, the technology used imposes a certain structure on the task in terms of the required level of interdependence, but it is no less true that the same task may be performed at different levels of interdependence (Wageman, 1995). This fact may provide further facets for the understanding of longitudinal studies (Hollingshead et al., 1993) that suggest that the differences comparing virtual teams using asynchronous technologies with face-to-face teams tend to decline over time. A plausible explanation for this finding is that the team adjusts the level of

interdependence in the task to the possibilities for interaction permitted by the technology. Based on the above, we investigated to what extent performance in a VT may depend on the individual team member’s degree of synchrony in the communication technology used. Hence, we hypothesize that:

Effects of task interdependence

H1. For low task interdependent teams, synchrony in the communication technology is negatively related to team performance.

265

H2. For high task interdependent teams, synchrony in the communication technology is positively related to team performance.

Method Participants A total of 240 participants were randomly assigned to 80 three-person teams. We recruited subjects from human resource management masters degree courses and final-year psychology students in Spain. These participants were considered ideal because the tool used to support the work of the virtual teams, Edustancew, was at the final stage of testing and not yet available in the market. Task setting The task used replicates Saavedra et al.’s (1993) adaptation of the merit rating task originally proposed by Marcic (1989). The items were written up in Spanish, maintaining the structure originally proposed. In this study, the task was managed electronically through the Edustancew website environment. As in the original study performed by Saavedra et al. (1993) each description had a correct and unique numerical solution, based on the weights assigned to the different phrases that describe different workers by two independent experts (Kappa ¼ 0:92). Design The true experiment was a 2 (task interdependence: low and high) £ 2 (communication technology synchrony: asynchrony and synchrony) factorial design. As in the work of Saavedra et al. (1993), the time allotted for the task was insufficient to complete all of the descriptions offered to the participants under the different conditions. Procedure The procedure followed the rules established by Saavedra et al. (1993), with appropriate adjustment of our variables and their application to a virtual teams work environment. The participants were placed in groups of 12 or 24 and each session included all of the experimental conditions with one or two teams per condition depending on the number of participants present. On arrival, the participants were seated in a room with individual work stations from which they were able to access the intranet hosting the Edustancew tool. When the participants opened the session in the system with the usernames and passwords assigned by the researchers, they entered a work area containing the necessary resources to carry out the task. As shown in Figure 1, the resources of the team consisted of different links. The first four allowed the team to access:

JMP 20,3/4

266

Figure 1. EduStance environment for virtual team work

(1) (2) (3) (4)

The instructions adjusted to the relevant experimental condition. Practice employee descriptions. The 42 employee descriptions for the definitive task. The tables of weightings to rate the employees and calculate merit increases.

Finally, a link was provided to open the forum or chat room depending on the experimental condition assigned. When the participants had entered the work environment, the experimenters asked them to access the instructions, which were read out loud with the participants. These instructions were translated and adapted to Spanish from Saavedra et al. (1993, pp. 65-66): For many employees, their annual pay raise is the most important feedback that they receive from their organization. In the following group exercise you will rate employees . . . calculate a total weighted score for the employee. Determine a recommended merit increase using the guidelines of the form.

The experimenters performed an example together with the participants in order to demonstrate how to determine the rating, how to weight it, how to calculate the total weighted score and how to recommend a merit increase. The existence of a single correct solution for each employee that would be used to determine the team’s performance was stressed.

The participants were requested to perform the task individually for five minutes in order to be sure that the task was well understood. We asked for questions and then we enacted the experimental manipulations. For this purpose, the participants were asked to exit the online work environment and were distributed in three different rooms. They were given a new password and username providing direct access to the experimental conditions. The instructions for participants working under the condition of low interdependence explained that they were working as a team, but that each member should work individually to complete each stage to determine the salary increase for the employees. In the high interdependence situation, the participants were also told that they should work as a team, but in this case each team was free to decide how to distribute responsibilities for the performance of the task. In the asynchronous communication technology conditions, the participants were provided with a web-based work environment consisting of a kind of bulletin board system. The system permits team members to communicate by “posting” messages in a hierarchical way, providing a clear structure for the messages. In the synchronous communication technology conditions, participants were provided with a chat application. This application allows team members to establish real time communication. At the same time, it provided information on the team members connected and indicated which member was online at any given moment. After accessing the new work area, the participants were asked to open a link providing three practice examples. They were given ten minutes to complete this task. Following Saavedra et al. (1993) we chose to use a practice team trial in addition to the individual one as an opportunity for the group as a whole to rehearse their task. At the end of this period, a further five minutes were allotted to give feedback to the participants on their efforts and resolve any questions that arise. The participants were then asked to open all of the necessary windows to carry out the task and to access the link containing the descriptions of employees. The experimenters reinforced the manipulations and started the 30-minute period for the task. Microsoft Word Pad documents with rating sheets were collected on a diskette given to a team member, designated by his/her own team, to record the results. While rating documents were scored, participants completed a questionnaire that included manipulation checks. Finally the experimenters debriefed the teams and thanked them for taking part. Measures In order to check the experimental manipulation, we prepared a single questionnaire comprising randomly ordered self-report Likert-type items ranging from (1) strongly agree to (5) strongly disagree. Where possible, the items for each scale were averaged out in order to obtain a single measure for each participant. Task interdependence as perceived by team members was measured following the five-item scale used by Van der Vegt et al. (2001). “I depend on my colleagues for the completion of my work” is an example of the items used. The coefficient alpha for the five-item scale was 0.72. Communication synchrony was assessed using a single item “which of the following best characterizes the type of communication you could establish with other team members during task resolution?” where 1 ¼ communication between the team members took place with excessive pauses, and 2 ¼ communication between the team

Effects of task interdependence

267

JMP 20,3/4

members took place smoothly at the same time. This item was developed following the definition of synchrony provided by Montoya-Weiss et al. (2001). Team performance was measured as the total evaluations completed correctly by a team. As we said previously each description had a correct and unique numerical solution, based on the weights assigned to the different phrases that describe different workers by two independent experts.

268 Results Manipulation checks Synchrony manipulation was assessed by a 2 (response category) £ 2 (communication conditions) chi-square analysis, which demonstrated that the manipulation worked, x2 ð1; 240Þ ¼ 232:13, p , 0:00. Only two participants were misclassified. The task interdependence manipulation was assessed by using the mean scores of each participant on the task interdependence scale. Using the median point of the scale we divided the participants into two groups of high (M ¼ 4:12; SD ¼ 0:45) and low perceived task interdependence (M ¼ 1:47; SD ¼ 0:33). A 2 (recoded category)£2 (task interdependence conditions) chi-square analysis showed that the manipulation worked, x2 ð1; 240Þ ¼ 224:5, p , 0:00. Only four participants were misclassified. Test of hypothesis Table I provides mean scores, standard deviations and cells for team performance. The results of the ANOVA shows no main effect of task interdependence, Fð1; 240Þ ¼ 2:16, p ¼ 0:14. On the other hand, a main effect of communication synchrony was found, Fð1; 240Þ ¼ 14:46, p , 0:001. This was also the case with interaction between the two, Fð1; 240Þ ¼ 183:26, p , 0:001. Thus, the differences found in the performance of the teams did not seem to depend on the degree of interdependence of the team in the performance of the task. On the other hand, higher or lower synchrony in the communication channel available to the team to perform its tasks did indeed play a significant role in team performance. Taken as a whole, the teams that used synchronous communication in the task performed significantly better (M ¼ 13:07, SD ¼ 3:83) than those using asynchronous communication (M ¼ 11:2, SD ¼ 4:23), tð77Þ ¼ 22:07; p ¼ 0:041. Finally, the analysis of interaction revealed that this enhanced or detrimental impact on performance had an inverse and significant effect on the various conditions (see Figure 2). The teams using asynchronous communication showed significantly lower performance when we compare them operating under conditions of low interdependence (M ¼ 14:9, SD ¼ 2:24) with those in conditions of high

Variable team performance Table I. Mean scores, standard deviations for team performance

M SD No. groups/cell

Low task interdependence Asynchronous Synchronous communication communication technology technology 10.1 2.12 20

14.9 2.24 20

High task interdependence Asynchronous Synchronous communication communication technology technology 16.05 2.64 20

7.5 1.7 20

Effects of task interdependence

269

Figure 2. Team performance as a function of synchrony degree of technological communication and task interdependence

interdependence (M ¼ 7:5, SD ¼ 1:70); tð35Þ ¼ 11:74; p , 0:00. Conversely, teams using synchronous communication significantly perform better in a situation of high interdependence (M ¼ 16:05, SD ¼ 2:64) compared with those teams of low interdependence conditions (M ¼ 10:1, SD ¼ 2:12); tð36Þ ¼ 27:84; p , 0:00. As a final point, we may note that the difference found in the comparison of those conditions in which the teams performed best – i.e. low task interdependence and asynchronous communication (M ¼ 14:9, SD ¼ 2:24) against high task interdependence and synchronous communication (M ¼ 16:05, SD ¼ 2:64) – was not significant, tð37Þ ¼ 21:48; p ¼ 0:14. The difference between the worst performing teams was, however, significant. Thus, those teams operating under high task interdependence and asynchronous communication conditions (M ¼ 7:05, SD ¼ 1:70) perform significantly worse than those operating in conditions of low task interdependence and synchronous communication (M ¼ 10:1, SD ¼ 2:12); tð36Þ ¼ 24:27; p , 0:00. Discussion The empirical evidence obtained from the study reveals how the performance of VTs depends on the fit between the level of interdependence required by the task undertaken by a team and the degree of synchrony provided by available communication technologies. Our findings allow us to maintain our hypotheses regarding the positive relationship between the performance of VTs and the use of synchronous communication technology in teams carrying out low interdependence tasks. The finding that task interdependence and communication synchrony interact in their effects on the performance of VTs may provide a coherent pattern integrating the findings of studies that have compared VTs and face-to-face teams performing different types of task using a variety of communications media (Farmer and Hyatt,

JMP 20,3/4

270

1994; Hollingshead et al., 1993; Hollingshead and McGrath, 1995; O’Connor et al., 1993; Ocker et al., 1998; Straus and McGrath, 1994; Valacich et al., 1994). In the terms of McGrath’s task circumplex, the assignment we set for our teams can be defined as intellective. Now, it might be thought in light of earlier research (Hollingshead et al., 1993) that teams using synchronous communication tools would be better at such tasks. However, our results show that the high levels of performance achieved by virtual teams carrying out the same kind of task, in conditions of high interdependence and synchronous communication systems and low interdependence and asynchronous communication systems do not differ significantly. These results therefore refine the affirmation made by Straus and McGrath (1994) that the effectiveness of a VT depends to a great extent on the match between the task demands and the communication technology used by the team, because it takes into account the fit between the perception of these structural characteristics by the team members and the capacity of the available communication tools to support the required level of interdependence. This reflection would provide laboratory support for the findings of field studies (Maznewski and Chudoba, 2000; Ripoelle et al., 2003), where it has been observed that the actual performance of more complex tasks at higher levels of interdependence in different virtual teams is associated with the use of synchronous communication tools permitting more frequent communication and intricate decision-making processes. At the same time, the interaction effect found maybe should help us understand why differences in the performance of a given task by different teams using synchronous and asynchronous communication systems tend to decline over time (Hollingshead et al., 1993; O’Connor et al., 1993). Our explanation, in spite we did not take a longitudinal approach in the present study, is that the teams could change their modes of interaction, gearing them more closely the available communication support. For our part, we conditioned the performance of the task at different levels of interdependence via the instructions given. When these levels were supported by the degree of synchrony provided by the communication tools available to the teams, their performance was better. The results of this study also show that teams using synchronous communications as a whole performed better than those working with asynchronous communication tools. Despite the apparent support this evidence provides for the work of Ocker et al. (1998) and Warkentin et al. (1997), among others, the fact that the VTs operating in synchronous conditions and at low levels of interdependence performed worse appears to be more consistent with the conceptual framework of the media richness theory (Straub and Karahanna, 1998). It is, in fact, highly likely that the performance shortfall shown by these teams is closely related with the unnecessary “distraction” caused by a medium that is over rich in possibilities for the transmission of communications that are superfluous to effective task performance (Hollingshead et al., 1993). Some limitations of this study should be noted. In the first place, though every experiment is designed with a view to obtaining evidence of causal relationships, the extrapolation of laboratory findings to real contexts always involves an additional effort. Thus, subsequent research will be needed to confirm our findings in actual working environments. The need for a closer look at the different mediation processes employed by teams in situations where there is a better or worse fit between task characteristics and type of communication (e.g. team strategy, the capacity to handle

conflicts appropriately, or the manner in which knowledge is generated and harnessed to achieve team objectives) represents a further limitation. We believe it is necessary to reexamine the robustness of the relationships we have found in light of such intermediating elements. Taking these limitations into account, the data obtained indeed have implications for practice, theory and research. From a theoretical perspective, we have sought to understand the effects of combinations between the levels of interdependence required by a task undertaken by a virtual team and the communications media used. In this regard, we have tried to highlight the importance of considering perceived interdependence as a basic feature of tasks. Thus, we have activated task interdependence in terms of the perception of team members, thereby providing further support from the laboratory for the opinion voiced by Wageman (1995) that the design of a task will require a given level of interdependence, but that members’ perception of task interdependence may vary as they perform their task. Finally, our findings provide further evidence to support the extension of the task-media fit described by Daft and Lengel (1986) to virtual teams. At the same time, it allows us to adapt the task-media fit more precisely to the taxonomy proposed by McGrath and Hollingshead (1994), enriching the computer systems category with synchronous communication systems. From a research standpoint, the results of this study open up an avenue for further work to throw light on the interactions found from a longitudinal perspective. This would mean taking the evolution of the team from the commencement to the completion of the task into account, adopting a similar approach to that used by Gersick (1989), or considering the transition rate in the VT in moments of complexity and simplicity arising during the performance of the task in line with the proposals of Bell and Kozlowski (2002) or the findings of Maznewski and Chudoba (2000). Our findings represent an appeal to the structure of interactions and interdependence within the team and its use of technology, and this calls for further research based on the postulates of the adaptative structuration theory proposed by DeSanctis and Poole (1994). Finally, from a practical point of view the results of this study make clear that we need to optimize and nurture our investments in communication technologies, which represent the main channel for processes within VTs. We are of course aware of the inevitable restrictions imposed by organizational contexts on the day-to-day work of teams. Such optimization would initially involve adjusting the task/communication technology fit in order to facilitate the choice of appropriate media by virtual team leaders and members. Secondly, it would call for further consideration of the requirements for the design of technological solutions in accordance with task interdependence described by Turoff et al. (1993), creating a virtual team working space equipped with tools capable of offering contingent responses to the changing demands of the task and the team members’ perception of interdependence. Viable solutions in this area might include “advisory systems” providing on-line support for the normal flow of the team’s work, perhaps through one or more of the following: . recommendation systems for the type of technology best suited for the phase of the task at which the team finds itself at any given time;

Effects of task interdependence

271

JMP 20,3/4

.

.

272

procedures designed to complement weak points in existing systems, for example by establishing additional task structuration elements in the manner of the explicit temporary coordination proposed by Montoya-Weiss et al. (2001); and elements providing a response to the needs awareness of team leaders and members in line with the analysis of Olson and Olson (2003).

If we agree with Bell and Kozlowski (2002) when they say that “virtual teams are here, and they are here to stay” (p. 45), then the extension of the research proposed here and the application of findings will be of incalculable value to organizations seeking to harness the full potential of virtual team-working in the pursuit of their objectives. References Bell, B.S. and Kozlowski, S.W.J. (2002), “A typology of virtual teams: implications for effective leadership”, Group and Organization Management, Vol. 27 No. 1, pp. 14-49. Campion, M.A., Medsker, G.J. and Higgs, A.C. (1993), “Relations between work group characteristics and effectiveness: implications for designing effective work groups”, Personnel Psychology, Vol. 46, pp. 823-50. Cascio, W. (2000), “Managing a virtual workplace”, Academy of Management Executive, Vol. 14 No. 3, pp. 81-90. Cohen, S.G. and Bailey, D.E. (1997), “What makes teams work: group effectiveness research from the shopfloor to the executive suite”, Journal of Management, Vol. 23 No. 3, pp. 239-90. Cohen, S.G. and Gibson, C.B. (2003), “In the beginning: introduction and framework”, in Gibson, C.B. and Cohen, S.G. (Eds), Virtual Teams that Work: Creating Conditions for Virtual Team Effectiveness, John Wiley & Sons, New York, NY, pp. 1-13. Cummings, T.G. (1978), “Self-regulation work groups: a socio-technical synthesis”, Academy of Management Review, Vol. 3, pp. 625-34. Daft, R. and Lengel, R. (1986), “Information richness: a new approach to managerial behavior and organizational design”, in Staw, B. and Cummings, L. (Eds), Research in Organizational Behavior, JAI, Greenwich, CT, pp. 191-223. DeSanctis, G. and Poole, M.S. (1994), “Capturing the complexity in advanced technology use: adaptive structuration theory”, Organization Science, Vol. 5 No. 2, pp. 121-47. Duarte, D.L. and Tennant, S. (2001), Mastering Virtual Teams: Strategies, Tools and Techniques that Succeed, Jossey-Bass, San Francisco, CA. Farmer, S.M. and Hyatt, C.W. (1994), “Effects of task language demands and task complexity on computer-mediated groups”, Small Group Research, Vol. 25 No. 3, pp. 331-66. Gersick, G.J.G. (1989), “Marking time: predictable transitions in task groups”, Academic Management Journal, Vol. 32 No. 2, pp. 274-309. Goodman, P.S. (1986), “Impact of task and technology on group performance”, in Goodman, P.S. (Ed.), Designing Effective Work Groups, Jossey-Bass, San Francisco, CA, pp. 120-67. Guzzo, R.A. and Shea, G.P. (1992), “Group performance and intergroup relations in organizations”, in Dunnette, M.D. and Hough, L.H. (Eds), Handbook of Industrial and Organizational Psychology, Psychological Press, Palo Alto, CA, pp. 269-313. Hollingshead, A.B. and McGrath, J.E. (1995), “Computer-assisted groups: a critical review of the empirical research”, in Guzzo, R.A. and Salas, E. (Eds), Team Effectiveness and Decision Making in Organizations, Jossey-Bass, San Francisco, CA, pp. 46-78.

Hollingshead, A.B., McGrath, J.E. and O’Connor, K. (1993), “Group task performance and communication technology: a longitudinal study of computer-mediated versus face-to-face work groups”, Small Group Research, Vol. 24 No. 3, pp. 307-33. Huang, W.W., Wei, K.K., Watson, R.T. and Tan, B.C. (2003), “Supporting virtual team-building with a GSS: an empirical investigation”, Decision Support Systems, Vol. 34 No. 4, pp. 359-67. Jarvenpaa, S.L., Knoll, K. and Leidner, D.E. (1998), “Is anybody out there? Antecedents of trust in global virtual teams”, Journal of Management Information Systems, Vol. 14 No. 4, pp. 29-64. Levine, J.M. and Moreland, R.L. (1990), “Progress in small group research”, Annual Review of Psychology, Vol. 41, pp. 585-634. Lipnack, J. and Stamps, J. (2001), Virtual Teams: Reaching across Space, Time and Organizations with Technology, (2nd ed.), John Wiley, New York, NY. McGrath, J.E. (1984), Groups: Interaction and Performance, Prentice-Hall, Englewood Cliffs, NJ. McGrath, J.E. (1990), “Time matters in groups”, in Galegher, J., Kraut, R.E. and Egido, C. (Eds), Intellectual Teamwork: Social and Technological Foundations of Cooperative Work, Lawrence Erlbaum, Hillsdale, NJ, pp. 23-62. McGrath, J.E. (1991), “Time, interaction, and performance (TIP): a theory of groups”, Small Group Research, Vol. 22, pp. 147-74. McGrath, J.E. and Hollingshead, A.B. (1994), Groups Interacting with Technology: Systems, Ideas, Evidence and an Agenda, Sage, Newbury Park, CA. Marcic, D. (1989), Organizational Behavior: Experiences and Cases, West Publishing, St Paul, MN, pp. 65-7. Maznewski, M.L. and Atanassiou, N. (2003), “Designing the knowledge-management infrastructure for virtual teams: building and using social networks and social capital”, in Gibson, C.B. and Cohen, S.G. (Eds), Virtual Teams that Work: Creating Conditions for Virtual Team Effectiveness, John Wiley & Sons, New York, NY, pp. 196-213. Maznewski, M.L. and Chudoba, K.M. (2000), “Bridging space over time: global virtual team dynamics and effectiveness”, Organization Science, Vol. 11 No. 5, pp. 473-92. Mohrman, S., Cohen, S. and Mohrman, A. (1995), Designing Team-based Organizations, Jossey-Bass, San Francisco, CA. Mohrman, S., Galbraith, J. and Lawler, E. (1998), Tomorrow’s Organization: Crafting Winning Capabilities in a Dynamic World, Jossey-Bass, San Francisco, CA. Montoya-Weiss, M.M., Massey, A.P. and Song, M. (2001), “Getting it together: temporal coordination and conflict management in global virtual teams”, Academy of Management Journal, Vol. 44 No. 6, pp. 1251-62. Ocker, R., Fjermestad, J., Hiltz, S.R. and Johnson, K. (1998), “Effects of four modes of group communication on the outcomes of software requirements determination”, Journal of Management Information Systems, Vol. 15 No. 1, pp. 99-118. O’Connor, K.M., Gruenfeld, D.H. and McGrath, J.E. (1993), “The experience and effects of conflict in continuing work groups”, Small Group Research, Vol. 24 No. 3, pp. 362-82. Olson, G.M. and Olson, J.S. (2003), “Human-computer interaction: psychological aspects of human use of computing”, Annual Review of Psychology, Vol. 54, pp. 491-516. Ripoelle, K., Gluesing, J.C., Alcordo, T.C., Baba, M.L., Britt, D., McKether, W., Monplaisir, L., Ratner, H.H. and Wagner, K.H. (2003), “Context, task and the evolution of technology use in global virtual teams”, in Gibson, C.B. and Cohen, S.G. (Eds), Virtual Teams that Work: Creating Conditions for Virtual Team Effectiveness, John Wiley & Sons, New York, NY.

Effects of task interdependence

273

JMP 20,3/4

274

Rocco, E., Finholt, T., Hofer, E.C. and Herbsleb, J. (2000), “Designing as if trust mattered”, CREW Technical Report, University of Michigan, Ann Arbor, MI. Saavedra, R., Earley, P.C. and Van Dyne, L. (1993), “Complex interdependence in task-performing groups”, Journal of Applied Psychology, Vol. 78 No. 1, pp. 61-72. Shea, G.P. and Guzzo, R.A. (1987), “Group effectiveness: what really matters?”, Sloan Management Review, Vol. 28 No. 3, pp. 25-31. Stewart, G.L. and Barrick, M.R. (2000), “Team structure and performance: assessing the mediating role of intra-team process and the moderating role of task type”, Academy of Management Journal, Vol. 43 No. 2, pp. 135-48. Straub, D. and Karahanna, E. (1998), “Knowledge worker communications and recipient availability: toward a task closure explanation of media choice”, Organization Science, Vol. 9 No. 2, pp. 160-75. Straus, S.G. (1996), “Getting a clue: the effects of communication media and information distribution on participation and performance in computer-mediated and face-to-face groups”, Small Group Research, Vol. 27 No. 1, pp. 115-42. Straus, S.G. and McGrath, J.E. (1994), “Does the medium matter? The interaction of task type and technology on group performance and member reactions”, Journal of Applied Psychology, Vol. 79, pp. 87-97. Susman, G. (1976), Autonomy at Work: A Sociotechnical Analysis of Participative Management, Praeger, New York, NY. Thompson, J.D. (1967), Organizations in Action, McGraw-Hill, New York, NY. Townsend, A.M., De Marie, S.M. and Hendrickson, A.R. (1996), “Are you ready for virtual teams?”, HR Magazine, Vol. 41 No. 9, pp. 122-6. Turoff, M., Hiltz, S.R., Bahgat, A.N.F. and Rana, A.R. (1993), “Distributed group support systems”, MIS Quarterly, Vol. 17 No. 4, pp. 399-417. Valacich, J.S., Dennis, A.R. and Connolly, T. (1994), “Idea generation in computer-based groups: a new ending to an old story”, Organizational Behavior and Human Decision Processes, Vol. 57 No. 3, pp. 448-67. Van de Ven, A.H., Delbecq, A.L. and Koenig, R. (1976), “Determinants of coordination modes within organizations”, American Sociological Review, Vol. 41, pp. 322-8. Van der Vegt, G.S., Emans, B.J. and Van de Vliert, E. (1998), “Motivating effects of task and outcome interdependence in work teams”, Group and Organization Management, Vol. 23, pp. 12-144. Van der Vegt, G.S., Emans, B.J. and Van de Vliert, E. (2001), “Patterns of interdependence in work teams: a two-level investigation of the relations with job and team satisfaction”, Personnel Psychology, Vol. 54, pp. 51-69. Wageman, R. (1995), “Interdependence and group effectiveness”, Administrative Science Quarterly, Vol. 40, pp. 145-80. Warkentin, M., Sayeed, L. and Hightower, R. (1997), “Virtual teams versus face-to-face teams: an exploratory study of a web-based conference system”, Decision Sciences, Vol. 28 No. 4, pp. 975-96. Watson, R.T., DeSanctis, G. and Poole, M.S. (1988), “Using a GDSS to facilitate group consensus: some intended and unintended consequences”, MIS Quarterly, Vol. 12 No. 3, pp. 462-79.

The Emerald Research Register for this journal is available at www.emeraldinsight.com/researchregister

The current issue and full text archive of this journal is available at www.emeraldinsight.com/0268-3946.htm

Idea generation through computer-mediated communication The effects of anonymity

Computermediated communication 275

Joa˜o Pissarra and Jorge C. Jesuino ISCTE – Instituto Superior de Cieˆncias do Trabalho e Empresa, Lisboa, Portugal Abstract Purpose – Brainstorming is a well-known group process for generating new ideas and stimulating creativity. Important as well as robust findings have been achieved in determining which factors contribute most to facilitating or hindering the group’s ideas productivity. Research aimed at comparing face-to-face (FTF) with computer-mediated communication (CMC) led to the conclusion that this latter shared with the nominal group technique the advantages of avoiding either the blocking effect or the identification of the source. More recently, attention has turned to the possible effects of group support system (GSS) in the mediating cognitive processes of generating new ideas. The present study aims to examine the effects of the type of tool and of the anonymity condition on the quality, quantity and diversity of the generated ideas, as well as on group members’ satisfaction. Design/methodology/approach – Uses a 2 £ 2 factorial design combining two different GSS tools (topic commenter vs EBS) with anonymity versus non-anonymity. Findings – It was found that anonymity generated more satisfaction among the group members. A marginal effect on satisfaction was also found to be related with the type of tools. Contrary to expectations, the EBS tool was not found to generate greater diversity of ideas. An interesting finding not anticipated was the impact of technology on the flow of ideas and on the emergence of new conceptual categories, probably due to alternative strategies of task structuring. Research limitations/implications – The use of students as subjects, and the running of the experimental work in a scholarly context, could have contributed to the elimination of fears and to freeing the participants from any inhibition in the anonymity conditions. Within an organisational context with higher social stratification, such anonymous procedures could have significant outcomes. Future research will have to examine whether this effect is relevant to other types of topics and other populations. Another aspect that it is important to re-examine is the effect of anonymity on the emergence of minority ideas, which could stimulate innovation. Practical implications – The type and characteristics of tools were shown to be a decisive factor in the participants’ satisfaction, in the communication process and in the idea generation and clustering processes. Although tenuous, this set of data could mean that the characteristics of the tools interfere with the cognitive mechanisms present in the brainstorming technique. Originality/value – Examines the effect of the technology and anonymity in ideas generation within a group context on the satisfaction of the participants. Keywords Communication technologies, Group decision support systems Paper type Research paper

Introduction The generation of new ideas is a cultural demand, vital to business, to science, and to the progress of society. New and creative ideas are essential to solve problems,

Journal of Managerial Psychology Vol. 20 No. 3/4, 2005 pp. 275-291 q Emerald Group Publishing Limited 0268-3946 DOI 10.1108/02683940510589055

JMP 20,3/4

276

economise work hours, spare the efforts of many people, and meet their social and cultural needs. The study of creativity as well as the techniques and processes that facilitate and stimulate its emergence, has grown exponentially (Campbell, 1968; VanGundy, 1992; Smith, 1998). The techniques of creativity stimulation and creative problem solving are an empirical approach that has been the object of research in various domains of psychology (Osborn, 1957; Taylor et al., 1958; McGrath, 1984). They consist mainly of instructions, and manipulations, capable of arousing the creative potential of individuals namely when working with others, either face-to-face (FTF) or mediated by computers (CMC) (for a taxonomy of these techniques see VanGundy, 1988, 1992). Among the researchers interested in developing technologies of group support and, particularly, the brainstorming technique, there seems to be a distinction and a tension between the European and the American perspectives. The former is more interpretative and the latter clearly positivist (Shaw et al., 2002). This tension has set the stage for a number of debates and contributions (Eden, 1995; Jesuino, 2002; Nagasundaram and Dennis, 1993; Valacich et al., 1994a). Eden (1995) suggested that this tension could be articulated in terms of a “narrow-band” perspective, characterised by producing experimental evidence based on laboratorial studies, or alternatively, in terms of a “wide-band” perspective, based on a more interpretativist paradigm. At present, this geo-cultural distinction has become less sharp, with both perspectives tending to converge or at least to complement each other, both contributing to explain how new ideas’ generation might improve the group decision-making. Brainstorming holds a distinct position within the above techniques, giving rise to new procedures and rules aimed at increasing productivity and creativity in groups. Brainstorming raised high expectations but its history is full of enigmas (Taylor et al., 1958; Diehl and Stroebe, 1987, 1991; Stroebe and Diehl, 1994). According to Osborn (1957), who first introduced the technique, members of a group were invited to generate ideas exempted of criticism or evaluation, which would supposedly stimulate the creativity of individuals both directly and/or mediated by other members’ ideas. In this latter case the intellectual synergy and chain reaction would have a larger positive effect in the production of ideas than when compared with the condition where neither any interaction nor any exposure would take place (Osborn, 1957). From the outset, the validity of this technique raised many doubts. Taylor et al. (1958), observed that the nominal groups were more productive, in quantity and quality, than the interactive groups. Half a century of empirical research (Mullen et al., 1991; Stroebe and Diehl, 1994) showed that in a verbal brainstorming context the nominal groups presented advantages over the face-to-face groups in terms of the number, quality and originality of the generated ideas, although in this latter parameter the results were not always consistent (Diehl and Stroebe, 1987; Pissarra, 1994). As a possible explanation Lamm and Trommsdorff (1973), suggested that the lower results of the FTF groups were due to factors like production blocking (time span between the emergence of the idea and its expression in the interactive groups), the social inhibition (negative evaluation of ideas by the author), the conformity and fear to

show divergent ideas to the others and to the cognitive interference produced by the exposition to other’s ideas. In a series of well-known studies, Diehl and Stroebe (1987, 1991) tested a number of hypotheses regarding the advantages of nominal groups in comparison to FTF, in terms of productivity, originality and quality of the generated ideas. Underlying the experiments was the assumption that in the brainstorming setting, interaction processes negatively affect the efficacy of the FTF groups. Three processes were analysed: previous evaluation/fear, social loafing (free riding) and blocking. They concluded that blocking was the main cause of the decrease in efficacy of the FTF brainstorming. They even come to sustain, that the face-to-face groups are so inefficient that there is no justification for their use in the production of ideas (Stroebe and Diehl, 1994). A similar conclusion had already been suggested by Mullen et al. (1991). The “matching effect” is another explanation for production losses within interactive groups (Paulus and Dzindolet, 1993). Brown et al. (1998) also suggested that the performance of interactive groups is poorer because the members tend to adjust their productivity to the least productive member of the group. More recently, the stimulation and synergy effect has attracted further attention as an intragroup process with consequences in the number and quality of the generated ideas (Dugosh et al., 2000; Nagasundaram and Dennis, 1993; Paulus and Yang, 2000; Taggar, 2001). The exposition to others’ ideas can arouse cognitive stimulation as well as cognitive interference, thus leading to gains and/or losses in the process of generating ideas in group (Nijstad et al., 2002). In the last decade, the research scene in this domain changed with the use of a large range of tools and systems introduced by information technologies and communication to support the teamwork – group support systems (GSS) (Jessup and Valacich, 1993). With GSS the members communicate and exchange written messages, instead of interacting verbally. These systems are a combination of electronic communication tools, designed to support the decision-making processes through analysis of alternatives, negotiating and voting (DeSanctis and Gallupe, 1987; Ellis et al., 1991). The GSS are effective in reducing loss processes – production blocking, free riding, sucker effect, evaluation apprehension – (Reinig and Shin, 2002), and also positively affect the cohesion and satisfaction of members in group meetings. Electronic brainstorming (EBS) has been considered as a better technique than verbal brainstorming or nominal groups (Dennis and Valacich, 1994). As a matter of fact, the technique allows one to combine the advantages of working alone, thus reducing blocking, but without suppressing the advantage of benefiting from the inputs of the other group members. The novelty and fascination of technology, also seems to compensate the negative effect of cognitive inertia that is the informative overload in the productivity of groups. The physical dispersion of members and the different combinations of synchronic and asynchronic interaction, in space and time, between members are also possibilities that this technology offers to the group support (Valacich et al., 1994b). Ziegler et al. (2000) in a critical revision of the research on EBS maintain that it is not possible to support the gains referred in the electronic to groups because the data is fragmented and very difficult to compare. Similar remarks were advanced by McGrath and Hollingshead (1993).

Computermediated communication 277

JMP 20,3/4

278

However, the comparison between EBS and FTF brainstorming has shown a clear advantage for the groups generating ideas in a CMC context (Valacich et al., 1994a) and particularly in larger groups (Gallupe et al., 1992). However scant attention was given to the differences between the systems and tools used in the different studies (McGrath and Hollingshead, 1994; Ziegler et al., 2000). The comparison between EBS and nominal groups has led to mixed results. Some studies suggest that EBS could yield better results than nominal groups, particularly in groups with more than nine members (Dennis et al., 1990; Valacich et al., 1992a; Dennis and Valacich, 1993; Valacich et al., 1994a). Other studies did not find any differences (Gallupe et al., 1991), or rather weak ones (Gallupe et al., 1992; Roy et al., 1996), whereas some found evidence of the advantage for nominal groups (Pinsonneault and Barki, 1999a, b). The weak consensus surrounding this topic points to the need for further research. Group brainstorming is conditioned by social factors and cognitive factors, but EBS is also influenced by the specific characteristics of the tools and systems of group support (McGrath and Hollingshead, 1994). In addition, the CMC groups performing tasks of ideas generation, are not groups in the traditional sense (Nagasundaram and Dennis, 1993), but rather a set of individuals that interact with others’ ideas, on a screen and according to a structure and organization imposed by the tool. The comparison between specific features of technology and its effects on the number, sequence, organisation, and idea categorisation, as well as on the participant’s satisfaction, remain topics on the research agenda of group dynamics within the CMC context. In the present article, we will focus on the effects on performance and group dynamics of two characteristic features of this tool: anonymity and the number of sheets of discussion available to the members of the group for presenting their contributions. The research in this domain has given priority to the quantity of generated ideas, focusing in the productivity as the main purpose of the technology of group support. But the quality of ideas, as more related with creativity (Barki and Pinsonneault, 2001) is the major challenge for brainstorming, even when known that the two measures are related (Lamm and Trommsdorff, 1973; Stroebe and Diehl, 1994). In the present study, the aim is to assess the quantity, quality and diversity of the generated contributions.

Anonymity and identification in idea generation The form of communication in GSS is typically anonymity. In a revision of regarding the use of GSSs, Fjermestad and Hiltz (1999) found that over 85 per cent of the tasks that involved anonymity were about idea generation. The same authors mention that anonymity presents benefits in situations of divergent thought. Anonymity in GSS research defined as the inability of group members to identify the origin of messages they receive, and the destination of messages they send. Thus, anonymity emerges as a striking feature of GSS (Jessup et al., 1990; Jessup and Tansik, 1991; Nunamaker et al., 1993; Valacich et al., 1992a, b), although the direction and intensity of its effects remains an open question. Anonymity can stimulate interaction because it offers fewer threats in terms of communicational environment, reduces inhibition and encourages the participation of the group’s members in the presentation of non-conventional and/or less common ideas.

Anonymity can also facilitate the presentation of uninhibited comments as well as the use of a cruder language style (Siegel et al., 1986). On the other hand, it can lead to cognitive inertia (Connolly et al., 1990; Kerr and Bruun, 1981, 1983). But idea generation, as remarked by Diehl and Stroebe (1987), is a task that requires little effort, and is usually considered interesting and motivating by the participants. This could eventually reduce the free-riding negative effect of anonymity in the group’s performance. The size and closeness of the group’s members could also influence the anonymity effect by leading the members of the larger and physically disperse groups (Jessup and Tansik, 1991) to feel more anonymous and to develop mechanisms facilitating the emergence of the free-riding effect and cognitive inertia. The members of natural groups (with history) are able to more easily identify the contributions of the members than the groups where members hardly know each other (Jessup et al., 1990). This could lead the members of the natural groups to be afraid of presenting their ideas. Anonymity could have both positive and negative effects on group interaction and on the performance of groups (McLeod, 2000). Its effect will always be contingent on the type of group, the task and the situation. The findings of the empirical research do not always confirm the advantages of the anonymous condition. For example, in accordance with Sosik et al. (1998), anonymity had no effect, nor did it stimulate the emergence of new and creative ideas. This could be due to the fact that people in an experimental situation, on the one hand do not fear presenting their contributions, but, on the other, they more easily renounce their effort and interest in generating new ideas. However, several studies draw empirical evidence showing that anonymity reduces the social barriers to participation, such as inhibition and fear (Jessup et al., 1990; Jessup and Tansik, 1991; Cooper et al., 1998; McLeod, 2000). Various types of anonymity affect the CMC (Barreto and Ellemers, 2002). One lively debate about anonymity in modern societies and in CMC and the internet, focuses on the extent to which anonymity might lead to collective disruptive behaviours, as well as to a break in norms and social values. This research led to the conclusion that the anonymity decreases the adherence to the group’s norms (Postmes et al., 2001), which is consistent with the pessimistic vision of the anonymity effects (Kiesler et al., 1984; Siegel et al., 1986) in a CMC context. More recently, these European researchers attempted to examine the effects of anonymity on the members of a group, by combining the theories of social identity and social influence (Baumeister, 1982; Spears and Lea, 1994; Reicher et al., 1995). The social identity model of deindividuation effects (SIDE) model postulates two components of the anonymity effect associated to the way in which it is manipulated, distinguishing the cognitive component from the strategic component (Barreto and Ellemers, 2002). The cognitive component refers to the salience given to self-categorisation, manipulating the available information about the other members of the group. The effect of the cognitive component shows that subjects tend to follow the norm, more often, when they interact in a context of physical dispersion than when they interact in the same physical space (Spears et al., 1990). This could mean that the anonymity effect is associated to the visibility among the members of the group (Barreto and Ellemers, 2002). The strategic component is associated with desirable and identifiable behaviours of each member within the

Computermediated communication 279

JMP 20,3/4

280

groups. A positive presentation of the self to the others usually leads to acceptance, approval and recognition by others. This could be a reason why subjects avoid behaviours that others disapprove of or view as threatening (Ellemers et al., 1999). In short, more pro-group effort is expected in cases where there is no information about the members of the group and when the participants are sure that their answers are known by the members. In this case, anonymity promotes pro-group behaviours, reducing the effects of diffusion of social responsibility and generating additional motivation favouring the group and its norms (Barreto and Ellemers, 2002). In the GSS context, Valacich et al. (1992b) distinguished two types of anonymity: process and content. In the process, anonymity participants ignore whether the other members are in fact participating. In the content anonymity prevents others from knowing who contributes with a particular messages or idea. Both of these forms of anonymity focus on technical aspect of anonymity. In the group situations that occur simultaneously and in the same place, it is difficult to ensure the anonymity of process by technology, unlike the anonymity of content that can occur either at the local level (group or team) or at the global level (organisation). This technical dimension of anonymity is not equivalent (Hayne and Rice, 1997) to social anonymity. However, content anonymity emerges as a critical aspect in idea generation and in the processes of small group communication (McLeod, 2000). Content anonymity occurs when any meaningful identifying information about group’s members is removed from any material or information exchanged. Social anonymity (Hayne and Rice, 1997) happens when participants perceive others to be de-individuated or not identifiable, because they do not perceive any cues or contexts for make any attributions of theirs identities in relating to the messages. In GSS technical anonymity is an important feature, but not sufficient condition for promoting social anonymity. By divorcing ideas from their authors, content anonymity is expected to promote the generation of more idea and improve the quality of ideas, and group performance.

Technology and performance in groups Research in the domain of group support systems has been focusing on the comparison of GSS with groups not supported by GSS. But the specific effects of technology’s features in the interaction and performance of groups have not been the focus of a systematic study (McGrath and Hollingshead, 1994; Anson et al., 1996; Oravec, 1996). GSS is a collaborative technology that can be used to improve the productivity through the facilitation of communication, the structuring and orientation of the collective effort in problem solving and the establishment and alignment between individual goals, and those of the group. EBS, uses discussion sheets, aiming at decreasing the cognitive overload and stimulating the positive effect of multi-dialogues in idea generation (Nunamaker et al., 1991; Dennis and Valacich, 1993). Groups use GSSs for at least three functions: (1) As information systems. (2) For structuring the accomplishment of the task. (3) As communication systems (McGrath and Hollingshead, 1993, 1994; McGrath and Berdahl, 1998).

The losses and gains in the group’s performance vary according to the different applications of the groups, the specific features of the systems themselves, the characteristics of the users, their learning and appropriation of the technology, as well as the cultural context surrounding the application and use of this group of technologies (Watson et al., 1994). Complex cognitive activities (choice between alternatives, risk evaluation, strategic planning, decision making, negotiation processes) require coordination of information and meanings among members. All these systems structure and shape the way, the interactions and the sequences in which the groups accomplish their assigned tasks. These different structuring patterns could improve the speed and quality of performance but they could also raise new problems for group work due to information overload (McGrath and Berdahl, 1998). Group memory is one of the organising factors of information processing. A positive relation has been found between the collective memory available through the GSS and the creativity of generated ideas within the group (Satzinger et al., 1999). The access availability to a large number and diversity of ideas is expected to encourage the construction of associative patterns. GSS systems offer many tools, which contain potential structuring features (Orlikowski, 2000) that guide the group’s interaction and condition its dynamics, cohesion, satisfaction and performance (Reinig and Shin, 2002). The analysis of the effects of these specific features of technology has not however attracted the attention of researchers. The present study The present study examines the effects of the type of tool and of the anonymity condition on the quality, quantity and diversity of the generated ideas, as well as on group members’ satisfaction. In accordance with the review dismissed above, two main hypotheses were posited. H1. Groups that perform the brainstorming sessions in an anonymous as compared with group in a non-anonymity condition will generate: (a) more satisfaction of the group members; (b) greater number of unique ideas; and (c) greater number of good ideas. Anonymity is supposed to facilitate the expression of minority ideas, by reducing social inhibition and the fear of evaluation, contributing to equalise the status of the members, as well as, to expanding the span and diversity of ideas. In the procedure adopted the group members were in visual contact, which in accordance with the SIDE model findings could reinforce pro-group behaviour and give additional motivation for their participation and interest in the execution of the task. In fact, the lack of identification clues removes fears, annulling the potential negative effects on sharing ideas with others, favouring positive states of satisfaction and disinhibition in the brainstorming sessions. H2. The groups that perform sessions within an EBS setting differ according to the GSS used. More specifically the EBS tool as compared with the topic commenter (TC) tool will generate: (a) more conceptual categories of ideas (diversity); and (b) more ideas within categories.

Computermediated communication 281

JMP 20,3/4

282

The creators of these tools insistently remark that the possibility offered to the subjects within an EBS context of opening several simultaneous dialogues, reduces the cognitive overload and stimulates divergent thinking, which might contribute to the production of more sophisticated ideas (Nunamaker et al., 1997; Reinig and Shin, 2002). Others’ ideas are external stimuli that might trigger cognitive stimulation. It is thus expected that the discussion sheets of EBS, exhibit more ideas within each category. This semantic similarity and temporal proximity in words lists was often found in verbal communication contexts (Bousfield, 1953; Gruenewald and Lockhead, 1980). Identical findings with ideas presented to isolated individuals could be found in CMC (Nijstad et al., 2002). Method A 2 £ 2 factorial design: type of tool (topic commenter vs EBS) £ anonymity (anonymity of sources vs identification of sources). The tool – topic commenter – displays in one sole sheet all the contributions of the members of the group, whereas the EBS offers to participants all the contributions displayed in six different sheets. In the anonymity situation, the contributions never refer to the source. In the identification condition, all the contributions are instead followed by the christian name of the member of the group that produced them. Participants A total of 150 volunteer college undergraduates have participated in this study – 60 male and 90 female with an average age of 20 years old (SD ¼ 2.5) for male and of 21.08 (SD ¼ 2.024) for female. The mean of the sample was 20.653 years old (SD ¼ 2.25). The subjects were invited to participate in the experiment in a classroom setting. The brainstorming sessions were arranged and the subjects were randomly distributed in groups of five members in the experimental conditions. A total of 30 groups were formed. Instruments GroupSystems, version 2.01 of Ventana Corporation. A ten items, seven point (1 – strongly unsatisfied, 7 – strongly satisfied) scale of satisfaction. The selected items were identical to the ones used in other studies, evaluates the degree of satisfaction face to the participation, to the used interface, the productivity, the communication process (Connolly et al., 1990; Dennis et al., 1990, Dennis and Valacich, 1993; Gallupe et al., 1991; Valacich et al., 1994a, b). The alpha-cronbach found was 0.88. The post-experimental questionnaire was composed of additional items in order to check the manipulation of the anonymity, as well as other dimensions of the experiment. Procedure According to the traditional rules of brainstorming technique, verbal and written instructions were distributed to the subjects. A training session of the technique and how to operate of the technological device was carried out. The basic procedures were similar to those described by Bouchard (1972) and Lamm and Trommsdorff (1973). The proposed task was the “problem of tourism” (ideas to increase the number of foreign tourists in Portugal in the year 2002). Similar tasks were used in other studies

(Dunnette et al., 1963; Street, 1974; Dennis and Valacich, 1993; Paulus and Dzindolet, 1993; Barki and Pinsonneault, 2001). The participants were welcomed at the university hall and accompanied to the room where the equipment was installed. The instructions were delivered in a written text. Afterwards the researcher read them aloud and clarified the questions raised by the subjects. Then the participants got familiar with the technology. Finally, the task was presented and, in the following 20 minutes, the group members generated ideas accessible to all on the screen. At the end of the session, the post-experimental questionnaire was individually applied. The outputs of the sessions were then subject to the appreciation of three blind experts: a woman with a degree in tourism, a man with a degree in history and with experience in programmes of rural tourism, and another woman with a degree in anthropology. Independent variables The effects of technologies were manipulated by comparing the tool designated as topic commenter (TP) with the tool designated as electronic brainstorming (EBS). The TP, presents many similarities with the common “chats” on the internet, allowing the members of the group full access to all the public ideas, both sequentially and simultaneously, using for that purpose only one discussion sheet. The EBS allowed the groups to work on six discussion sheets. The members disclosing their ideas to the public, changed automatically to the free-discussion sheet, where they could find the contributions of other participants. In this working context, the participants never had access to the integral memory of the group nor to the temporal sequence of the contributions. According to the designers of the tool this device favours the creation of parallel dialogues and divergent thinking, the mobilization of the subject’s attention to the ideas in the different discussion sheets, and the prevention of information overload. The author’s anonymity was manipulated, by asking the subjects to identify themselves and to write their name at the beginning of the session – identification of authorship’s environment – and each every time that they disclosed their contributions. In the situation of anonymity the ideas’ authorship was not disclosed. In order to check the manipulation, at the end of each session the participants were asked to which extent it was able to identify the author of the ideas. The manipulation of the variable was successful in all the cases. In the post-experimental relaxation period it was also noticed that many participants in the anonymity condition were anxious to know who was the author of certain ideas, and were surprised by the mistakes in the attribution of authorships. In each of the experimental conditions all the participants were in visual contact with all the members of the group. Dependent variables The number of non-repeated ideas was based on the evaluations of the 3 judges of all idea produced by the groups. Non-repeated ideas were all the ideas considered as such at least by two of the judges. Inter-rater agreement (Cohen’s k) was judge 1 £ judge 2 ¼ 0.71 ( p , 0.00), judge 1 £ judge 3 ¼ 0.59 ( p , 0.00) and judge 2 £ judge 3 ¼ 0.65 ( p , 0.00).

Computermediated communication 283

JMP 20,3/4

284

The number of good ideas was based on the appreciations made by the judges about the non-repeated ideas on a five-point scale. The mean of the quality of all the non-repeated ideas was performed. All the ideas with a score above the mean were considered as good ideas and aggregated by the group. The intra-class correlation coefficient (ICC) of the appreciation of the judges concerning the quality of ideas was quite acceptable (average measure ¼ 0.70, p , 0.00). The diversity was based on the conceptual categorisation of the generated ideas. Ten different conceptual categories were identified within the universe of unique ideas. For each group, the number of conceptual categories was calculated spanning from 1 to 10. The clustering of ideas was computed by dividing the number of non-repeated ideas by the number of changes of category. The “shifting” was determined by the frequency of change of conceptual category. Satisfaction was also measured ( / ¼ 0.88) for each item of the scale.

Results According to the experimental design a series of ANOVA’s was performed. All the groups expressed high satisfaction for participating in the ideas’ generation session. The fascination with technology, dominant among the youngsters, and the recognised interest in the task are likely factors contributing to this outcome. The perception of the utility of the task, expressed by the participants in general, might have contributed to the well-being and satisfaction expressed by the group members. A main effect of the anonymity variable was found on an item “satisfaction with all the aspects of the session”. The groups that performed the task in an anonymous environment expressed more satisfaction with all the aspects of the session (anonymous groups: M ¼ 5.7 vs identified groups: M ¼ 5.3; F(1.3) ¼ 4.10; p , 0.05). This result are in line with our (H1a), supporting the argument, according to which anonymity elicits a positive atmosphere in groups performing ideas’ generation tasks. In the present case, the groups that performed the task in an anonymous environment expressed, in general, more satisfaction than the identified groups. The groups’ satisfaction also shows a marginal effect of the technology in the item relative to all the aspects of the session (topic commenter: M ¼ 5.68 vs EBS: M ¼ 5.36; F(1.3) ¼ 3.54; p , 0.07) and about the communication process (topic commenter: M ¼ 5.76 vs EBS: M ¼ 5.32; F(1.3) ¼ 3.05; p , 0.09). We also found a marginal interaction effect of technology £ anonymity (F(1.3) ¼ 3.86; p , 0.06), indicating how satisfaction is a result of technological and social factors such as the identification, or lack thereof, of the authors of the ideas (see Table I). In terms of ideas, the groups produced 1,385 ideas and the judges classified 1,146 as non-repeated ideas that is 82 per cent of the group session’s product. In terms of the amount of generated ideas no significant differences were found between the experimental conditions. The anonymous groups generated more ideas (M ¼ 40.13 than identified groups M ¼ 37.17; F(1.3) ¼ 0.192, ns) but this difference does not support our (H1b). In other words, in this parameter the expected positive effect of the anonymity condition was not found, thus contradicting the results of former studies. However, the most productive situation was found in the groups that performed the session in EBS context and anonymity (M ¼ 42.5).

Productivity of the group

Topic commenter Anonymity Identified n¼8 n ¼ 15 n¼7 Mean Mean Mean

Anonymity n¼7 Mean

EBS Identified n¼8 Mean

n ¼ 15 Mean

Number of non-repeated ideas SD

37.43 10.72

35.88 23.943

36.6 18.34

42.5 12.27

38.29 21.10

40.53 16.45

Number of good ideas SD

28.71 9.72

20.88 8.32

24.53 9.5

31.63 14.43

22.86 9.6

27.53 12.82

Diversity SD

7.71 1.61

6.13 1.35

6.8 1.6

7.63 1.59

6.71 1.6

7.2 1.6

Clustering SD

1.33 0.09

1.40 0.17

1.37 0.14

1.46 0.17

1.48 0.13

1.47 0.15

28.14 8.43

26.5 27.26

27.26 13.4

31.37 11.45

27.14 11.88

29.4 11.43

Shifting SD

In terms of the quality of generated ideas, measured by the number of good ideas, we noticed a main effect of anonymity (anonymous groups: M ¼ 30.27 vs identified group: M ¼ 21.8; F(1.3) ¼ 4.364; p , 0.04). These results agree with (H1c) and reveal that anonymity had a positive effect in the number of good ideas generated by groups. Regarding the number of conceptual categories – diversity – we did not found any difference between the tools (EBS, M ¼ 7.2 vs TC, M ¼ 6.87, F(1.3) ¼ 0.197, ns). But a main effect of the variable anonymity (anonymous groups: M ¼ 7.67 vs identified groups: M ¼ 6.4; F(1.3) ¼ 4.92; p , 0.03) was found. This finding is contrary to our (H2a). The advantages of the EBS tool, as systematically stated by its creators, were not confirmed (Nunamaker et al., 1989; Nunamaker et al., 1997). The effect of the parallel dialogues did not have, in this condition, positive effect, in terms of divergent thinking as well as in the diversity of ideas developed by the group members. On the contrary, it was found that the anonymous situation was the main catalyser for the emergence of greater diversity of ideas. This seems to be related with a disinhibition effect, reducing the fear of diverging with the group. The manipulation of the independent variables on the dimension of the clusters of ideas shows a marginal effect of the technology variable (topic commenter: M ¼ 1.37; EBS: M ¼ 1.47; F(1.3) ¼ 1,47, p , 0.07). This finding partially supports (H2b) suggesting that technology influences the way in which ideas are grouped and how they flow along the generation process. In this case, technology determined the diversity and dimension of the clusters of ideas, demonstrating their structuring effect in group activities. The post-experimental questionnaire raises some points worthy of further comment. For the item “as a group we paid attention to the presented ideas” we found a main effect of the technology (TP: M ¼ 4.6; EBS: M ¼ 4.3; F(1.3) ¼ 4.09, p , 0.05). The groups that performed the task in a TP context started to pay more attention to the ideas of others. In the same vein, we found a main effect of the technology for the item “as a group we listen to the ideas of each one of the members” (TP: M ¼ 4.53; EBS: M ¼ 4.16; F(1.3) ¼ 6,45, p , 0.01). The TP tool seems to be

Computermediated communication 285

Table I. Productivity of the group

JMP 20,3/4

286

similar to the common chats in which the participants are usually very skilled, namely in quickly reading the texts on the screen. The findings might suggest that the synchronism of the messages is more mobilising of the subjects’ attention than the time lag required by the change of screen on the EBS. One might speculate that the possibility of accessing the integral memory of the sessions could yield a more clear and positive notion of the value of the generated ideas as well as contributing to the construction of a positive group identity. Paying attention requires cognitive resources. Generating unique ideas and paying attention to others in this context might have contributed to the lower productivity of the TP tool groups. The perception of the possibly utility of the ideas, assessed by the item “the generated ideas would serve in the future to make decisions about the problem”, shows a marginal effect of the technology (TP: M ¼ 4.2; EBS: M ¼ 3.7; F(1.3) ¼ 3.36, p , 0.07). Leadership is also a relevant process in-group dynamics, insufficiently studied within the context of electronic groups. At the end of the sessions the participants were asked to what extent they could identify the group leader. For this item, a main effect of anonymity was observed (anonymous groups: M ¼ 1.97 vs identified groups: M ¼ 1.3, F(1.3) ¼ 10.06; p , 0.004). Since there were no clues about the sources, it was concluded that the participants recognised a leadership effect of the ideas rather than of specific leaders. Discussion In this study we examined the effect of the technology and anonymity in the ideas generation within a group context on the satisfaction of the participants. Our hypotheses were developed from previous findings (McGrath and Hollingshead, 1994; Jessup and Tansik, 1991; Nunamaker et al., 1989, 1993, 1997; Siegel et al., 1986). The results show a positive effect of anonymity on the satisfaction of group members, on the number of good ideas and on the semantic diversity. It was clearly found that anonymity positively affects groups’ performance in idea generation, reducing the evaluation effect and the fear of disagreeing. The pessimistic hypotheses of the anonymity effect (Jessup et al., 1990; Kiesler et al., 1984) were not confirmed. The results are more in accordance with the findings of Gallupe et al. (1994), but this could be the consequence of the way in which this variable was manipulated in the present study. The fact that all the group members had visual contact with each other, could stimulate desirable behaviours, reinforcing the acceptance by the others, inhibiting the expression of disruptive behaviours and exempting the members of the responsibility in the accomplishment of the groups’ tasks. In such a situation, and in accordance with the SIDE model, decent individuals behaved decently in group and produced more ideas than the identified groups (Reicher et al., 1995). The so-called “technical anonymity of content” (Valacich et al., 1992a, b) led to positive effects in the performance and atmosphere of small groups. The anonymity of content is a necessary, but not sufficient, condition for generating social anonymity. Thus the anonymity offered by GSSs did not confirm the pessimistic hypothesis of the negative effects of the computer-mediated communication in the productivity of groups. However, the anonymity of content offered by the technology

does not prevent the attribution processes of authorships (Hayne and Rice, 1997). This is an issue requiring further experimental evidence. Anonymity is a multidimensional concept; the technical anonymity is one of the components. In the future we need to reconceptualise as a subjective, multidimensional and continuous construct. Anonymity is a subjective and very complex factor that can interact with other situational variables in a quite unpredictable ways. We need pay more attention to these situational variables in order to better understand the effects of different kind of anonymity and is interaction with others factors. The use of students as subjects, and the running of the experimental work in a scholarly context, could have contributed to the elimination of fears and to freeing the participants from any inhibition in the anonymity conditions. Within an organisational context with higher social stratification, such anonymous procedures could have significant outcomes. Future research will have to examine whether this effect is relevant to other types of topics and other populations. Another aspect that is important to re-examine is the effect of anonymity on the emergence of minority ideas, which could stimulate innovation. The type and characteristics of tools were shown to be a decisive factor in the participants’ satisfaction, in the communication process and in the idea generation and clustering processes. Although tenuous, this set of data could mean that the characteristics of the tools interfere with the cognitive mechanisms present in the brainstorming technique. The technologic tools might exert some influence in the way of addressing as well as performing the tasks. In the ideas generation process, the task structuring is an outcome of: . cognitive mechanisms (total or partial availability of ideas, spreading across a number of discussion sheets, nature of the ideas in themselves); . social processes (author’s identification, sharing the same space versus dispersion, voting procedures and ideas’ selection); and . procedural mechanisms (instructions, simultaneity, interaction patterns among members). The specific traits of the tools combined with the technical instructions too creativity use some of the above mechanisms which the present study shows to produce a marginal effect in the generation of ideas. Idea generation in group is the product of cognitive processes (Nagasundaram and Dennis, 1993) and social processes (Diehl and Stroebe, 1987, 1991; Paulus and Dzindolet, 1993). But in this task many other factors interfere, such as the specifications of the support technology, the degree of access to the memory of the group (integral availability or spread over different discussion sheets) that cannot be underestimated. In future studies, it will be worthy to examine the way in which some of these features interfere with the mechanisms of the associative memory as well as with the cognitive stimulation of the participants. References Anson, R., Fellers, J., Kelly, G. and Bostrom, R. (1996), “Facilitating research with group support systems”, Small Group Research, Vol. 2, pp. 179-214.

Computermediated communication 287

JMP 20,3/4

Barki, H. and Pinsonneault, A. (2001), “Small group brainstorming and idea quality: is electronic brainstorming the most effective approach?”, Small Group Research, Vol. 2, pp. 158-205.

288

Baumeister, R.F. (1982), “A self-presentational view of social phenomena”, Psychological Bulletin, Vol. 91, pp. 3-26.

Barreto, M. and Ellemers, N. (2002), “The impact of anonymity and group identification on pro-group behavior in computer-mediated groups”, Small Group Behavior, Vol. 5, pp. 590-610.

Bouchard, T.J. (1972), “A comparison of two group brainstorming procedures”, Journal of Applied Psychology, Vol. 1, pp. 45-9. Bousfield, W.A. (1953), “The occurrence of clustering in the recall of randomly arranged associates”, Journal of General Psychology, Vol. 36, pp. 67-81. Brown, V., Tumeo, M., Larey, T.S. and Paulus, P.B. (1998), “Modeling cognitive interactions during group brainstorming”, Small Group Research, Vol. 29, pp. 495-526. Campbell, J.P. (1968), “Individual versus group problem solving in an industrial sample”, Journal of Applied Psychology, Vol. 3, pp. 205-10. Connolly, T., Jessup, L.M. and Valacich, J. (1990), “Effects of anonymity and evaluative tone on idea generation in computer-mediated groups”, Management Science, Vol. 6, pp. 689-703. Cooper, W.H., Gallupe, R.B., Pollard, S. and Cadsby, J. (1998), “Some liberating effects of anonymous electronic brainstorming”, Small Group Research, Vol. 2, pp. 147-78. DeSanctis, G. and Gallupe, R.B. (1987), “A foundation for the study of group decision support systems”, Management Science, Vol. 5, pp. 589-609. Dennis, A.R. and Valacich, J.S. (1993), “Computer brainstorming: more heads are better one”, Journal of Applied Psychology, Vol. 4, pp. 531-7. Dennis, A.R. and Valacich, J.S. (1994), “Group, sub-group, and nominal group idea generation: new rules for a new media?”, Journal of Management, Vol. 4, pp. 723-36. Dennis, A.R., Valacich, J.S. and Nunamaker, J.F. (1990), “An experimental investigation of the effects of group size in an electronic meeting environment”, Man and Cybernetics, Vol. 20, pp. 1049-57. Diehl, M. and Stroebe, W. (1987), “Productivity loss in brainstorming groups: toward the solution of riddle”, Journal of Personality and Social Psychology, Vol. 3, pp. 497-509. Diehl, M. and Stroebe, W. (1991), “Productivity loss in idea-generation groups: tracking down the blocking effect”, Journal of Personality and Social Psychology, Vol. 3, pp. 392-403. Dugosh, K.L., Paulus, P.B., Roland, E.J. and Yang, H.C. (2000), “Cognitive stimulation in brainstorming”, Journal of Personality and Social Psychology, Vol. 79, pp. 722-35. Dunnette, M.D., Campbell, J. and Jaastad, K. (1963), “The effect of group participation on brainstorming effectiveness for two industrial samples”, Journal of Applied Psychology, Vol. 1, pp. 30-7. Eden, C. (1995), “On evaluation of the performance of wide-band GDSS”, European Journal of Operational Research, Vol. 81, pp. 302-11. Ellemers, N., Barreto, M. and Spears, R. (1999), “Commitment and strategic responses to social context”, in Ellemers, N., Spears, R. and Doosje, B. (Eds), Social Identity: Context, Commitment, Content, Blackwell, Oxford, pp. 127-46. Ellis, C.A., Gibbs, S.J. and Rein, G.L. (1991), “Groupware some issues and experiences”, Communication of the ACM, Vol. 1 No. 1, pp. 39-58.

Fjermestad, J. and Hiltz, S.R. (1999), “An assessment of group support systems experimental research: methodology and results”, Journal of Management Information Systems, Vol. 15, pp. 7-149. Gallupe, R.B., Bastianutti, L.M. and Cooper, W.H. (1991), “Unblocking brainstorms”, Journal of Applied Psychology, Vol. 71, pp. 137-42. Gallupe, R.B., Cooper, W.H., Grise´, M. and Bastianutti, L.M. (1994), “Blocking electronic brainstorms”, Journal of Applied Psychology, Vol. 79, pp. 77-86. Gallupe, R.B., Dennis, A.R., Cooper, W.H., Valacich, J.S., Bastianutti, L.M. and Nunamaker, J.E. (1992), “Electronic brainstorming and group size”, Academy of Management Journal, Vol. 35, pp. 350-69. Gruenewald, P.J. and Lockhead, G.R. (1980), “The free recall of category examples”, Journal of Experimental Psychology: Human Learning and Memory, Vol. 6, pp. 225-40. Hayne, S. and Rice, R. (1997), “Attribution accuracy when using anonymity in group support systems”, International Journal of Human Computer Studies, Vol. 47, pp. 429-50. Jessup, L.M. and Tansik, D.A. (1991), “Group problem solving in an automated environment: the effects of anonymity and proximity on group process and outcome with a group decision support systems”, Decision Sciences, Vol. 2, pp. 266-79. Jessup, L.M. and Valacich, J. (Eds) (1993), Group Support Systems: New Perspectives, Macmillan, New York, NY. Jessup, L.M., Connolly, T. and Galegher, J. (1990), “The effects of anonymity on GDSS group process with an idea-generating task”, MIS Quarterly, Vol. 14, pp. 312-21. Jesuino, J.C. (2002), “European views of computer-mediated communication”, Small Group Research, Vol. 33 No. 5, pp. 475-80. Kerr, N.L. and Bruun, S.E. (1981), “Ringelmann revisited: alternative explanations for the social loafing”, Personality and Social Psychology Bulletin, Vol. 7, pp. 224-31. Kerr, N.L. and Bruun, S.E. (1983), “Dispensability of member effort and group motivation losses: free-rider effects”, Journal of Personality and Social Psychology, Vol. 1, pp. 78-94. Kiesler, S., Siegel, J. and McGuire, T.W. (1984), “Social psychological aspects of computers and communication”, American Psychologist, Vol. 39, pp. 1123-34. Lamm, H. and Trommsdorff, G. (1973), “Group versus individual performance on tasks requiring ideational proficiency (brainstorming): a review”, European Journal Social Psychology, Vol. 4, pp. 361-88. McGrath, J.E. (1984), Groups, Interaction and Performance, Prentice-Hall, London. McGrath, J.E. and Berdahl, J.L. (1998), “Groups, technology, and time: use of computers for collaborative work”, in Scott Tindale, R. (Ed.), Theory and Research on Small Groups, Plenum Press, New York, NY, pp. 205-28. McGrath, J. and Hollingshead, A.B. (1993), “Putting the ‘group’ back in group support systems: some theoretical issues about dynamic processes in groups with technological enhancements”, in Huber, G.P., Valacich, J.S. and Jessup, L.M. (Eds), A Theory of Effects of Group Support Systems on an Organization’s Nature and Decisions – Group Support Systems: New Perspectives, Macmillan, New York, NY. McGrath, J. and Hollingshead, A.B. (1994), Groups Interacting with Technology, Sage Publications, London. McLeod, P.L. (2000), “Anonymity and consensus in computer-supported group decision making”, Research on Managing Groups and Teams, Vol. 3, pp. 175-204.

Computermediated communication 289

JMP 20,3/4

290

Mullen, B., Johnson, C. and Salas, E. (1991), “Productivity loss in brainstorming groups: a meta-analytic integration”, Basic and Applied Social Psychology, Vol. 65, pp. 219-25. Nagasundaram, M. and Dennis, A.R. (1993), “When a group is not a group: the cognitive foundation of a group idea generation”, Small Group Research, Vol. 4, pp. 463-89. Nijstad, B., Stroebe, W. and Lodewijkx, F.M. (2002), “Cognitive stimulation and interference in groups: exposure effects in an idea generation task”, Journal of Experimental Social Psychology, Vol. 38, pp. 535-44. Nunamaker, J., Vogel, D. and Konsynski, B. (1989), “Interaction of task and technology to support large groups”, Decision Support Systems, Vol. 5, pp. 139-52. Nunamaker, J.F., Briggs, R.O., Mittleman, D., Vogel, D.R. and Balthazard, P. (1997), “Lessons from a dozen years of group support systems research: a discussion of lab and field findings”, Journal of Management Information Systems, Vol. 3, pp. 163-207. Nunamaker, J.F., Dennis, A.R., Valacich, J.S., Vogel, D.R. and George, J.F. (1991), “Electronic meeting systems to support group work”, Communications of the ACM, Vol. 7, pp. 40-61. Nunamaker, J.F., Dennis, A.R., Valacich, J.S., Vogel, D.R. and George, J.F. (1993), “Issues in the design, development, use, and management of group support systems”, in Huber, G.P., Valacich, J.S. and Jessup, L.M. (Eds), A Theory of Effects of Group Support Systems on an Organization’s Nature and Decisions – Group Support Systems: New Perspectives, Macmillan, New York, NY. Oravec, J.A. (1996), Virtual Individuals, Virtual Groups: Human Dimensions of Groupware and Computer Networking, Cambridge University Press, London. Orlikowski, W.J. (2000), “Using technology and constituting structures: a practice lens for studying technology in organizations”, Organizational Science, Vol. 4, pp. 404-28. Osborn, A.F. (1957), Applied Imagination, Scribner, New York, NY. Paulus, P.B. and Dzindolet, M.T. (1993), “Social influence processes in group brainstorming”, Journal of Personality and Social Psychology, Vol. 4, pp. 575-86. Paulus, P.B. and Yang, H.C. (2000), “Idea generation in groups: a basis for creativity in organizations”, Organization Behaviour and Human Decision Processes, Vol. 82, pp. 76-87. Pinsonneault, A. and Barki, H. (1999a), “The illusion of electronic brainstorming productivity: theoretical and empirical issues”, Information Systems Research, Vol. 4, pp. 378-83. Pinsonneault, A. and Barki, H. (1999b), “Electronic brainstorming: the illusion of productivity”, Information Systems Research, Vol. 2, pp. 110-52. Pissarra, J. (1994), Brainstorming Electro´nico: Grupo Interactivo versus Grupo Nominal (Electronic Brainstorming versus Nominal Group), LNEC, Lisboa. Postmes, T., Spears, R., Sakhel, K. and De Groot, D. (2001), “Social influence in computer-mediated communication: the effects of anonymity on group behavior”, Personality and Social Psychology Bulletin, Vol. 27, pp. 1242-54. Reicher, S.D., Spears, R. and Postmes, T. (1995), “A social identity model of de-individation phenomena”, European Review of Social Psychology, Vol. 6, pp. 161-98. Reinig, B. and Shin, B. (2002), “The dynamic of group support systems on group meetings”, Journal of Management Information Systems, Vol. 2, pp. 303-25. Roy, M.C., Gauvin, S. and Limayem, M. (1996), “Electronic brainstorming: the role of feedback on productivity”, Small Group Research, Vol. 2, pp. 215-47. Satzinger, J.W., Garfield, M.J. and Nagasundaram, M. (1999), “The creative process: the effects of group memory on individual generation”, Journal of Management Information Systems, Vol. 15, pp. 143-60.

Shaw, D., Eden, C. and Ackerman, F. (2002), “Evaluation group support systems: improving brainstorming research methodology”, Aston Business School Research Papers, available at: www.abs.aston.ac.uk Siegel, J., Dubrovsky, V., Kiesler, S. and McGuire, T.W. (1986), “Group processes in computer-mediated communication”, Organizational Behavior and Human Decision Processes, Vol. 37, pp. 157-87. Smith, G.F. (1998), “Idea-generation techniques: a formulary of active ingredients”, Journal of Creative Behavior, Vol. 2, pp. 107-33. Sosik, J.J., Avolio, B. and Kahai, S. (1998), “Inspiring group creativity: comparing anonymous and identified electronic brainstorming”, Small Group Research, Vol. 1, pp. 3-31. Spears, R. and Lea, M. (1994), “Panacea or panopticon? The hidden power of computer-mediated communication”, Communication Research, Vol. 21, pp. 427-59. Spears, R., Lea, M. and Lee, S. (1990), “De-individuation and group polarisation in computer-mediated communication”, Communication Research, Vol. 21, pp. 427-59. Street, W.R. (1974), “Brainstorming by individuals, coacting and interacting groups”, Journal of Applied Psychology, Vol. 3, pp. 433-6. Stroebe, W. and Diehl, M. (1994), “Why groups are less effective than their members: on productivity losses in idea-generating groups”, European Review of Social Psychology, Vol. 5, pp. 271-303. Taggar, S. (2001), “Group composition, creative synergy, and group performance”, Journal of Creative Behavior, Vol. 4, pp. 261-86. Taylor, D.W., Berry, P.C. and Block, C.H. (1958), “Does group participation when brainstorming facilitate or inhibit creative thinking?”, Administrative Science Quarterly, Vol. 3, pp. 23-47. Valacich, J., Dennis, A.R. and Connolly, T. (1994a), “Idea generation in computer-based groups: a new ending to an old story”, Organizational Behavior and Human Decision Process, Vol. 57, pp. 448-67. Valacich, J., Dennis, A.R. and Nunamaker, J.F. (1992a), “Group size and anonymity effects on computer-mediated idea generation”, Small Group Research, Vol. 1, pp. 49-73. Valacich, J., George, J., Nunamaker, J.F. and Vogel, D. (1994b), “Physical proximity effects on computer-mediated group idea generation”, Small Group Research, Vol. 1, pp. 83-104. Valacich, J., Jessup, L., Dennis, A. and Nunamaker, J. (1992b), “A conceptual framework of anonymity in group support systems”, Proceedings of the 25th Hawaii International Conference on Systems Sciences, Kauai, HI, pp. 101-12. VanGundy, A.B. (1988), Techniques of Structured Problem Solving, Van Nostrand Reinhold, New York, NY. VanGundy, A.B. (1992), Idea Power: Techniques and Resources to Unleash the Creativity in Your Organizations, AMACOM, New York, NY. Watson, R.T, Hua Ho, T. and Raman, K.S. (1994), “Culture: a fourth dimension of group support systems”, Communications of ACM, Vol. 10, pp. 45-55. Ziegler, R., Diehl, M. and Zilstra, G. (2000), “Idea production in nominal and virtual groups: does computer-mediated communication improve group brainstorming?”, Group Processes and Intergroup Relation, Vol. 3, pp. 141-58.

Computermediated communication 291

The Emerald Research Register for this journal is available at www.emeraldinsight.com/researchregister

JMP 20,3/4

The current issue and full text archive of this journal is available at www.emeraldinsight.com/0268-3946.htm

Organisational support and group efficacy A longitudinal study of main and buffer effects

292

Amparo Osca Universidad Nacional de Educacio´n a Distancia (UNED), Madrid, Spain

Begon˜a Urien Universidad Pu´blica de Navarra, Pamplona, Spain

Genoveva Gonza´lez-Camino and M. Dolores Martı´nez-Pe´rez Universidad Nacional de Educacio´n a Distancia (UNED), Madrid, Spain, and

Nuria Martı´nez-Pe´rez University of Glamorgan, Pontypridd, UK Abstract Purpose – To analyse the influence of three different types of organisational support (supervisor’s and colleagues’ support, training, and acknowledgement and rewards) on the implementation of teamwork systems. Design/methodology/approach – Main and buffer effects of social support were tested using different subjective (job satisfaction and job involvement) and objective (production and total production management (TPM)) organisational criteria. In the longitudinal study, two sets of data were taken from a sample of workers from car-manufacturing factories in two consecutive years. Findings – The three dimensions of support explain 30 per cent of the variance in job satisfaction in time 1(T1) and 11 per cent in time 2 (T2). A total of 50 per cent of job involvement in T1 is due to the supervisor’s and colleagues’ support, whereas in T2 this influence is not apparent. Social support from supervisor’s and colleagues and acknowledgment and rewards explain 10 per cent of the variance of the objective measures (production and TPM). Research limitations/implications – The objective measures used are similar for the production line as a whole and therefore they do not consider the possible differences in performance between different production line groups. Practical implications – The importance of social support in organisational settings, in particular in the implementation of working teams. It also emphasises the need to distinguish the different stages of support from colleagues, supervisors or the organisation. Originality/value – This study indicates that support is an easy and inexpensive means not only to improve the social environment of an employee but also to make an important contribution towards productivity targets. Keywords Role conflict, Role ambiguity, Job satisfaction, Performance management Paper type Research paper

Journal of Managerial Psychology Vol. 20 No. 3/4, 2005 pp. 292-311 q Emerald Group Publishing Limited 0268-3946 DOI 10.1108/02683940510589064

Introduction According to the reciprocity norm of the social exchange theory (Gouldner, 1960), employees perceiving support from the organisation value their organisation at a greater scale and tend to actively collaborate to achieve the company’s goals (e.g. Eisenberger et al., 1986; Rousseau, 1989; Wayne et al., 1997). Thus organisational

support generates further positive work attitudes as consistently mentioned in the literature. One of the first studies on the influence of social support on work was run by LaRocco et al. (1980) who pointed out that employees that feel supported show better psychological well-being, higher job satisfaction and better performance. Furthermore, it has been found unanimously, that support is positively related to job satisfaction (Beehr et al., 1990; Thomas and Ganster, 1995; Nye and Witt, 1993). Corroborating this information Roadhes and Eisenberger (2002), in a meta-analysis of 70 studies, observed that the relationship between support and job satisfaction is in all cases positive with a magnitude between “moderate” to “strong”. The latest literature review distinguishes between several types of organisational support. Frone et al. (1997) found that colleague and supervisor support have a positive influence on job satisfaction. Bennet et al. (2001) have observed that the supervisory support and organisational acknowledgement are related to job satisfaction. Baruch-Feldman et al. (2002) in a study with traffic police also found that supervisor support is related to job satisfaction. Moreover, there is wide evidence of social support related to organisational commitment (Eisenberger et al., 1990; Guzzo et al., 1994; Wayne et al., 1997). Frone et al. (1997) found that supervisory support is positively related to temporal work commitment. Schaubroeck and Fink (1998) also obtained positive relationships between support and organisational commitment. Eisenberger et al. (2001) using a structural equation causal model, concluded that support has a direct and positive influence on affective commitment with the organisation. In a longitudinal research, Stinglhamber and Vanderbergue (2003) found that organisational support is related to organisational commitment and that supervisory support and organisational support have different links as well as highlighting the importance of distinguishing between both dimensions. The meta-analysis of Roadhes and Eisenberger (2002) established evidence of the connection support-commitment: the relationship between organisational support and affective commitment with the organisation is strong, whereas the relationship between organisational support and job involvement is just moderated. However, even though most literature on organisational support is focused on relationships involving satisfaction and commitment (Randall et al., 1999), there is also evidence of positive relationships between organisational support and performance (Armeli et al., 1998; Eisenberger et al., 1986, 1990). Organ (1988) points out that employees perceiving themselves as correctly treated by the organisation will respond with extra-effort, which will affect their job performances as well as the extra-role behaviour. Summarising, the meta-analysis made by Roadhes and Eisenberger (2002) shows that the relationships between organisational support and extra-role performance are positive and of medium magnitude, whereas the relations with other organisational outcomes are weaker. Most studies asses work performance via subjective methods. As an example, Eisenberger et al. (2001) using a structural equation causal model, found that social support affects indirectly and negatively two subjective measurements of performance. However, Vecchio (1998) points out that the use of subjective measures can often develop into difficulties such as not showing any relationship with objective measures and stronger dependency relative to the quality of the relationship between the observer and the person under assessment.

Organisational support and group efficacy 293

JMP 20,3/4

294

Organisational support has also been related to objective indicators of performance. Some examples are: . the number of traffic fines and detentions made by the police (Armeli et al., 1998; Eisenberger et al., 1990); . the number of creative suggestions of manual employees (Eisenberger et al., 1990); . the absenteeism (Eisenberger et al., 1986, 1990); and . the volunteer turnover (Roadhes et al., 2001). Baruch-Feldman et al. (2002), evaluating productivity through the number of traffic fines obtained that supervisory support was related to productivity. Other authors, for example Frese (1999) in his study with metal industry workers, do not find significant relationships between organisational support and the objective measures of organisational performance. Direct or moderating effect of social support A large amount of knowledge regarding social support comes from occupational stress literature. In this field, several models are proposed to explain the influence of social support on psychological well-being. One of the most well known models is that of Cohen and Wills (1985), where the main and moderating (buffer) effects of social support are discussed. The main effect hypothesis indicates that support is always positive and therefore will always have a beneficial effect on health. On the other hand, the moderating effect hypothesis upholds that support affects health only under stress situations. As observed previously, with the exception of the demand-control-support model studies (Karasek and Theorell, 1990), the greater part of studies in the organisational area focus on the direct effect of support, while there is considerably less research concerning the moderating effect. Frese (1999) in a longitudinal study with 90 metal factory workers found a moderating effect of the social support on the relationship between some occupational stressors and psychological ill-health measures as annoyance and anxiety. However, in other publications the moderating effect of support was not found. Schaubroeck and Fink (1998) considering different sources of support and different stressors, found direct effects of support on job satisfaction but observed that support did not reduce the effects of high demands and low control. A study in a Japanese bank (Iwata and Suzuki, 1997) examined the potential moderating effects of social support. Among 12 interaction terms between stressors and support variables, only one attained statistical significance, again illustrating a lack of buffering effects for social support on stressor-strain relationships. However, the meta-analysis of Viswesvaran et al. (1999), where a great amount of research was gathered on both hypotheses (main effect and moderating effect), showed that both explanations were supported and could be applied depending on the situational specificity. To conclude, further investigation on this area would be needed. Social support in the group efficiency models The investigation on group efficiency also identifies the influence of organisational support. Hence one of the features of the so-called high performance teams, is that its members support and motivate each other both personally and professionally.

One way of supporting the teams is improving the group efficiency with training. The main models of group performance sustain this idea (Campion et al., 1993; Gladstein, 1984; Hackman, 1987; Sundstrom et al., 1990; Tannenbaum et al., 1996). With the implementation of working teams, employees need specific training in different areas (technical and professional updating, interpersonal and team working skills, etc.). In this situation it has been found that those employees getting more support show higher job satisfaction (Campion et al., 1993; Hackman, 1987; Teague et al., 1995). Training also improves group performance by means of providing the skills needed to work as a part of a team (Sundstrom et al., 1990). Another form of support is the appreciation and rewards offered to employees. Employee acknowledgement and rewarding on account of their group contribution is one of the most effective ways that organisations have to promote team working (Hackman, 1987; Larson and LaFasto, 1989). Hyatt and Ruddy (1997) found that the teams obtaining the necessary support of superiors and organisation tended to be more effective. Team size could affect organisational support. The number of members can affect the quality of support, so the individuals of large teams may not receive the advisable social support from their supervisor (Dubinsky et al., 1995; Green et al., 1996). Larger units may attenuate the amount of social support a supervisor can give to any single agent. It is also presumed that support at the beginning of the team establishment process is of great importance because team members may experience role conflict and uncertainty due to their new roles and duties (new tasks, work distribution, higher coordination requirements, etc.). Burke and Greenglass (2001) analysed social support in a hospital during a staff cutback period and found that nurses perceiving more support had higher satisfaction and less somatic and psychological symptoms. This paper aims to analyse the role of social support when an important organisational changing process such as teamwork implementation programs are taking place. Following Eisenberger et al. (1997) differentiation of the three ways of support (justice, supervisor support, rewards and working conditions) a scale adapted to the specific situation of the sample was designed. This scale integrated three dimensions: (1) Supervisory and colleagues support. (2) Training and acknowledgement. (3) Rewards on the new working system. The influence of these types of support on job satisfaction was also analysed together with job involvement and performance through two objective indexes that the organisation provided. The hypotheses of Cohen and Wills (1985) on main and moderating effects of support were also tested by analysing their relationship with role stressors (role conflict and role ambiguity). The implementation of work teams (the new tasks, the shared responsibilities, the need of coordination, etc.) could produce role ambiguity and conflict because of the new demands and duties involved in the new working structures. Finally, in order to evaluate the evolution of the situation, the effects of these relationships are also analysed in two instances with a temporal lapse of a year in between.

Organisational support and group efficacy 295

JMP 20,3/4

296

Hypotheses Analyses of effects at time 1 (T1) H1. The three dimensions of organisational support considered in the first data collection (T1) will be directly and positively related to job satisfaction, job involvement and performance in T1. Therefore in this first hypothesis the main effect of support is tested. H2. The three dimensions of organisational support in T1 will moderate the relationship between role conflict and ambiguity in T1 and the resulting variables: job satisfaction, job involvement and performance in T1. Therefore in this second hypothesis the support’s buffer effect is tested. Analyses of effects at time 2 (T2) H3. The three dimensions of organisational support in T1 will predict job satisfaction, commitment with the new working system and performance in T2. H4. The three dimensions of organisational support will moderate the relationship between role conflict and role ambiguity in T1, the change in job satisfaction from T1 to T2, commitment with the new working system and performance. Method Procedure This study is part of a larger investigation carried out on work groups’ implementation in the Spanish site of a very well known multinational firm within the automobile industry. The sample used was collected in the machining area. Two different engine plants were located in the same site: a traditional engine plant producing steel engines and a new plant producing aluminium engines for newer car models. In order to achieve a more efficient plant, work groups were introduced in the new part of the plant as well as other organisational and technological changes. The groups were autonomous work groups (Banker et al., 1996) formed by multi-skilled workers responsible for the quality, production, maintenance, problem solving, reporting and continuous improvement tasks. A longitudinal design was used to collect data over two periods of time. The first data collection was made after subjects worked in a group basis for at least six months. The new organisational system was introduced in February. The first collection of data was made in December with the following set collected a year later. The first data collection was performed from 15 work groups from five different production lines working three shifts. Every production line performed component assemblies. Sales expectations were not met that particular year so one of the shifts was removed at the second data collection. Therefore, the second data set was taken from ten working groups having two working shifts each. An individual level of analysis was used for this study. Sample The sample consisted of 130 operators at time 1 (T1) and 107 at time 2 (T2). Although the machining area had been designed to work three shifts, one of them was sent to the old plant due to failure in meeting sales targets.

Regarding socio-demographical aspects, the operators of both samples were men only. This situation was usual at the time when the samples were gathered. To guarantee the anonymity of the participants, the age and level of studies were collected using interval scales. Concerning the age, five alternatives where offered which covered the range between 18 and more than 65 years old. The most numerous age group (48.52 per cent) was between 27 and 36 years old, 21.48 per cent was between 18 and 26 years old, followed by 37 to 46 years old (16.29 per cent) and finally 1.49 per cent of the sample was 57 to 65 years old. As regards to the level of studies, the largest group was represented by members with a degree in FP-II (equivalent to a higher professional education), representing a 60.9 per cent of the sample, followed by a 16.15 per cent with FP-I (equivalent to further education), 7.69 per cent with secondary education, 7.69 per cent with primary education and finally 2.3 per cent without any specific studies. Measures In all variables, except for productivity measures, participants responded on a five-point scale which ranged from 1 “not at all” to 5 “very much”. Organisational support. Following Eisenberger et al. (1986) support classification, a scale was developed assessing three dimensions of support: (1) Supervisor and colleagues support: this subscale consisted of six items with reliability a ¼ 0:72. An example of the items measuring this type of support is: “how much has manager verbal and non verbal praise increased in the new working system?” (2) Training in the new working system: this subscale consists of three items with reliability a ¼ 0:84. An example of the items that made up this scale is “how much information have you received about the new working system?” (3) Acknowledgement and rewards in the new working system: the scale has two items with reliability a ¼ 0:81. The scale inquires about material rewards and promotions. An example of the items used: “has your salary been increased in any way (fix or variable) due to the organisational change?” Role stressors. Role ambiguity (a ¼ 0:80) and role conflict (a ¼ 0:75) are assessed by Rizzo et al.’s (1970) scale adapted by Peiro´ et al. (1986). Two items of the conflict scale were eliminated since they were considered not relevant for the study. Job satisfaction. This variable was measured by applying “The job diagnosis survey” (Hackman and Oldham, 1976). After conducted a reliability analysis, the scale was reduced to nine items. This scale asked for the satisfaction with group mates, supervisors, working system, rewards, etc. (T1: a ¼ 0:77; T2: a ¼ 0:79). Job involvement. This scale was designed for this study and is composed by four items. An example of the items used is “do you think that now you are achieving more professional and personal development than before?” (T1: a ¼ 0:78; T2: a ¼ 0:80). Productivity measures. These measures have been chosen depending on the area. They consist of general productivity parameters normally used to assess performance. The data was provided by the area manager. Because the company aims to reinforce

Organisational support and group efficacy 297

JMP 20,3/4

298

line integration avoiding competition between groups and shifts, productivity was measured at line level instead of at group level: . Production level: is the ratio of the number of pieces produced over the number of pieces the machine is able to process at the bottle neck of the line. . Total production maintenance (TPM) operations performed in each line: as a positive productivity indicator, TPM measures predict machinery malfunctions allowing enough time to increase the number of resources made available for the employee to use. Control variables. Team size and time working within the team (team tenure) are the control variables. Sometimes team size appears to be positively related to productivity and other times negatively. Gooding and Wagner (1985) conducted a meta-analysis and found neither no relationships between size and productivity nor negativity relationship depending of the measure used. Moreover, larger units may attenuate the amount of social support a supervisor can give to any single agent. Regarding the team tenure, newly formed groups tend to have lower levels of productivity than established groups presumably because they have not developed the shared mental models that underlie an effective team work (Guzzo and Dickson, 1996). Results Preliminary analysis of the data was made through descriptive statistics and correlation analysis. Table I shows correlations between measurements made at T1. These relations follow the hypothesis of the main effect of support on subjective measures and acknowledgement and rewards related to TPM. In Table II it can be seen that the evaluation made to acknowledge and reward in T1 is also related to all (subjective and objective) measurements of efficiency in T2. In order to verify the hypotheses of main and buffering effects, ten moderated hierarchical regression analyses were carried out. Variables were introduced by blocks into the regression equation. Previously new variables were created by multiplying the scores of parameters included in the interaction terms (role stress by organisational support). Control variables (team tenure and team size) are introduced in both times. In order to control the variance in T2, dependent variables of T1 were also considered. Centred scores were used to reduce multicollinearity during the regression analyses. Team tenure and team size were introduced first as control variables in the regression equation for analyses at T1. In the second step, role clarity and role conflict were introduced. In third place, supervisor and colleagues support, training on the new working system and perceived rewards were also introduced, and finally in forth place the interaction effects between the two role stressors and the three types of support was added. For T2 analyses, the sequence of entering was, first the corresponding variable in T1 for the dependent variable in the analysis; second, the two control variables (time and size) in T2; third, T1 role clarity and role conflict; fourth, T1 superiors and fellow workers support, training in the new working system and perceived rewards; and in the fifth step, the interaction effects between the two role stressors and the three types of support, collected in T1.

SD

3.42 2.08 3.79 3.59 67.56 10.91

2

3

4

0.077 2 0.097 2 0.356** 2 0.163** 0.025 2 0.069

5

6

7

8

9

10

20.029 0.365** 20.310** 0.504** 0.397** 20.441** 0.669** 0.362** 0.504** 20.230** 0.757** 0.332** 0.369** 0.550** 0.031 20.115 0.126 2 0.034 2 0.066 0.010 0.029 0.175 20.189* 2 0.025 0.097 0.002 2 0.619**

0.012 0.045 2 0.013 0.109 0.168** 0.438** 0.053 2 0.083 2 0.285** 20.359**

1

0.88 0.079 0.053 1.16 2 0.041 2 0.097 0.56 0.003 2 0.237** 0.81 0.080 2 0.088 2.24 0.058 2 0.196* 1.79 2 0.040 0.156

11.98 6.10 15.60 10.21 1.94 0.43 2.47 0.75 3.81 0.53

Notes: *p , 0:05; **p , 0:005

1. Team size T1 2. Team tenure T1 3. Role ambiguity T1 4. Role conflict T1 5. Super/fellows support T1 6. Acknowledgement and rewards T1 7. Training T1 8. Job satisfaction T1 9. Job involvement T1 10. Production T1 11. TPM T1

Mean

Organisational support and group efficacy 299

Table I. Means, standard deviations and correlations (time 1)

Table II. Means, standard deviations and correlations (time 2)

Notes: *p , 0:05; **p , 0:005

Team size T2 Team tenure T2 Role ambiguity T1 Role conflict T1 Super/fellows support T1 Acknowledgement T1 Training T1 Job satisfaction T2 Job involvement T2 Production T2 TPM T2

10.92 25.86 1.94 2.47 3.81 3.42 2.08 4.08 4.43 77.48 30.72

2.91 5.4 0.43 0.75 0.53 0.88 1.16 0.46 1.45 2.46 1.65

SD 2 0.162 2 0.061 0.126 2 0.036 0.023 0.044 0.050 0.263* 0.196 2 0.402**

1

20.053 20.093 20.041 0.275** 20.003 20.072 20.237* 20.044 0.214

2

0.438** 2 0.285** 0.077 2 0.097 2 0.330** 2 0.202 2 0.339** 2 0.080

3

20.359** 20.029 20.310** 20.372** 20.087 20.096 0.010

4

0.365** 0.504** 0.489** 0.238* 0.258* 0.229*

5

0.397* 0.122 0.018 0.195 20.099

6

0.081 0.165 0.150 0.072

7

0.387 0.452 0.058

8

300

1. 2. 3. 4. 5. 6. 7. 8. 9. 10. 11.

Mean

0.329 0.017

9

0.033

10

JMP 20,3/4

Results having satisfaction as criterion are shown in Table III (T1 analyses) and Table IV (T2 analyses). Direct effects were found mainly in T1 analysis since significant negative direct effects were observed with both role stressors, role ambiguity (b ¼ 20:24, p , 0:001) and role conflict (b ¼ 20:29, p , 0:001). Positive direct effects with all types of support studied were found: superiors and fellow workers support (b ¼ 0:44, p , 0:001), training on the new working system (b ¼ 0:12, p , 0:05), acknowledgement and rewards (b ¼ 20:15, p , 0:01). For T2 analyses, only T1 satisfaction (b ¼ 0:41, p , 0:001) and superiors and colleagues support (b ¼ 0:38, p , 0:01) had a significant relationship with T2 satisfaction. This model explained 56 per cent of the variance in satisfaction in T1 and 24 per cent in T2. The inclusion of the interaction terms did not increment significantly the explained variance either for T1 satisfaction nor for T2 satisfaction, even though the interaction between training and role conflict was significant in T1 analysis, and the interaction between superiors and fellow workers support and role conflict was significant in T2. Tables III and IV also show the results for the dependent variables commitment with the new system in T1 and T2 respectively. As it can be observed, in time 1 the model explains 60 per cent of the variance, where superiors and fellow workers support (b ¼ 0:77, p , 0:001) explain most of this variance (53 per cent) so the higher this support is, the higher the self-interest is expected. Another variable that also contributes significantly to a positive relationship is role conflict. The higher the role conflict the higher the self-interest. Commitment as criterion in T2 only shows a significant relationship with the control variable group size also in T2 (b ¼ 20:23, p , 0:01) explaining a 10 per cent of the variance. The smaller groups seemed to have higher commitment. Considering production and TPM as criterion variables in time 1 (Table V) and time 2 (Table VI), it can be observed that for both dependent variables in time 1, only two types of support have significant relationships: superiors and colleagues support (b ¼ 20:29, p , 0:01 for T1 production; b ¼ 0:35, p , 0:01 for T1 TPM) and acknowledgement and rewards (b ¼ 0:24, p , 0:05 for T1 production; b ¼ 20:28, p , 0:01 for T1 TPM). These models explain the 8 per cent of the variance when T1 production is the dependent variable and the 12 per cent of the variance when T1 TPM is the dependent variable. Results of production and TPM at time 2 (Table VI) show that the greater part of the variance in both models is explained as a result of the production (55 per cent) and TPM (72 per cent) respectively at T1 via increasing time 2 measures when time 1 measures rise. Taking T2 production as criterion, two other variables contribute to the model significance: size of the team in T2 (b ¼ 20:23, p , 0:01, so the higher the T2 team size the lower the T2 production) and role ambiguity in T1 (b ¼ 20:21, p , 0:01, so the higher the T1 role ambiguity the lower the T2 production). Interaction effects were not found. This model explains the 64 per cent of the variance. In addition to time 1 TPM effect when taking T2 TPM as dependent variable there has been found a negative significant relationship of T2 TPM to the size of the team (b ¼ 20:27, p , 0:001), so that the higher the team size the lower level of T2 TPM is expected. Direct effects with role stressors or with types of support were not found. Interactions effects were not found either. This model explains 80 per cent of the variance.

Organisational support and group efficacy 301

Table III. Results of regression analyses. Dependent variables: Job satisfaction and job involvement (time 1)

Notes: *p , 0:05; **p , 0:01; ***p , 0:001

Total

Step 3 Supervisor/colleagues support £ ambiguity T1 Supervisor/colleages support £ conflict T1 Training £ ambiguity T1 Training £ conflict T1 Rewards £ ambiguity T1 Rewards £ conflict T1 0.05 0.08 0.35 20.49* 20.24 0.22

0.44*** 0.12* 0.15**

20.24*** 20.29***

Step 1 Role ambiguity T1 Role conflict T1

Step 2 Supervisor/colleagues support T1 Training support T1 Rewards support T1

20.23*** 20.01

Team tenure T1 Team size T1

b

21.96***

39.17***

17.87***

6.37**

0.56

0.56

0.24

0.05

Satisfaction F R2

0.01

0.31***

0.19***

DR2

0.02 20.58 20.33 20.17 20.16 0.11

0.77*** 20.04 0.08

20.06 20.24***

0.07 20.06

b

25.88***

46.93***

4.96***

1.11

0.60

0.60

0.07

0.00

Involvement F R2

302

Steps

0.01

0.53***

0.07**

DR 2

JMP 20,3/4

0.38** 20.17 0.11

Step 4 Supervisor/colleagues support T1 Training support T1 Rewards support T1

Notes: *p , 0:05; **p , 0:01; ***p , 0:001

21.23 1.88* 1.12 20.68 0.99 20.37

20.17 20.13

Step 3 Role ambiguity T1 Role conflict T1

Step 5 Supervisor/colleagues support £ ambiguity T1 Supervisor/colleages support £ conflict T1 Training £ ambiguity T1 Training £ conflict T1 Rewards £ ambiguity T1 Rewards £ conflict T1 Total

20.04 0.08

0.41***

b

Step 2 Team tenure T2 Team size T2

Step 1 Measures T1

Steps

2.81**

4.44***

4.23**

5.66***

0.24

0.25

0.16

0.14

0.16

Satisfaction R2

16.39***

F

0.04

0.11**

0.04

0.00

0.16***

DR 2

2 0.64 0.89 0.84 2 0.15 0.19 2 0.08

0.16 0.11 0.02

2 0.14 2 0.04

2 0.18 2 0.23**

0.18

b

1.26

2.09*

2.91**

4.03**

0.02

0.09**

.02

0.02

0.02

0.10

.10

0.09

0.04

DR 2

0.02

Involvement R2

2.81

F

Organisational support and group efficacy 303

Table IV. Results of regression analyses. Dependent variables: Job satisfaction and job involvement (time 2)

Table V. Results of regression analyses. Dependent variables: production and TPM (time 1) 20.05 20.02 20.29** 0.05 0.24* 20.02 21.62* 20.51 0.72 20.10 20.17

Step Role ambiguity T1 Role conflict T1

Step Supervisor/colleagues support T1 Training support T1 Rewards support T1

Step Supervisor/colleagues support £ ambiguity T1 Supervisor/colleages support £ conflict T1 Training £ ambiguity T1 Training £ conflict T1 Rewards £ ambiguity T1 Rewards £ conflict T1

Notes: *p , 0:05; **p , 0:01; ***p , 0:001

Step

20.17 0.07

Team tenure T1 Team size T1

b

1.72*

2.00*

0.87

0.08

0.06

0.01

0.01

Production R2

1.57

F

0.07

0.09*

0.00

DR 2

0.37 0.27 0.17 2 0.84 2 0.81* 0.87*

0.35** 2 0.03 2 0.28**

2 0.06 0.07

0.13 2 0.05

b

2.09*

2.57**

0.64

0.98

F

TPM

0.12

0.10

0.01

0.00

R2

304

Steps

0.07

0.10**

0.00

DR 2

JMP 20,3/4

Notes: *p , 0:05; **p , 0:01; ***p , 0:001

Step 5

Step 4 Supervisor/colleagues support £ ambiguity T1 Supervisor/colleages support £ conflict T1 Training £ ambiguity T1 Training £ conflict T1 Rewards £ ambiguity T1 Rewards £ conflict T1 0.15 20.80 0.02 20.29 0.01 0.22

0.10 0.09 0.00

20.21** 20.01

Step 2 Role ambiguity T1 Role conflict T1

Step 3 Supervisor/colleagues support T1 Training support T1 Rewards support T1

0.09 20.23**

0.74***

b

Step 1 Team tenure T2 Team size T2

Measures T1

Steps

9.23***

16.83***

26.00***

36.47***

89.82***

0.61

0.64

0.63

0.59

0.55

Production F R2

0.01

0.01

0.04**

0.05**

0.55***

DR 2

20.06 20.41 20.09 0.10 0.30 20.11

0.05 0.06 0.02

20.10 20.05

20.00 20.27***

0.85***

b

TPM

21.66***

38.97***

61.61***

94.82***

194.98***

F

DR 2

0.73***

0.07***

0.01*

0.00

0.01

R2

0.72

0.79

0.80

0.80

0.80

Organisational support and group efficacy 305

Table VI. Results of regression analyses. Dependent variables: production and TPM (time 2)

JMP 20,3/4

306

Discussion The first objective of this paper was to analyse the organisational support given to employees in an organisational changing process such as the implementation of working teams. Due to the lack of assessment instruments in Spanish language in this area of work, and following Eisenberger et al. (1986) proposal, a tool concerning organisational support that assesses supervisor and colleagues support, training on the new working system and acknowledgement and rewards was commissioned. Although a number of group efficiency models consider the importance of social support when implementing work groups, research has overlooked this specific issue. This work aimed to consider work stress models and group efficacy models simultaneously and the role of support on them. The first hypothesis analysed the direct influence of social support on several objective and subjective organisational outcomes. According to the hypothesis of support main effect (Cohen and Wills, 1985), the results showed that the three kinds of support considered here can explain a great percentage of the variance on job satisfaction. These results are even greater than those obtained by Roadhes and Eisenberger (2002) in their meta-analysis. As in Roadhes and Eisenberger’s study, the results show that the highest correlations include satisfaction. A longitudinal analysis using the outcomes one year later (H3), showed that supervisor and colleagues support still affected job satisfaction a year later although, as expected, with less intensity. The most relevant organisational support type resulting from this investigation is the support from supervisors and colleagues (Bennet et al., 2001; Baruch-Feldman et al., 2002; Frone et al., 1997) as it is connected to both objective and subjective group performances. Acknowledgement and rewards also have an influence in both satisfaction and performance but at a lesser extent. Nevertheless, the results show training as the least important type of support as only affects satisfaction towards the position held and not, contrary to what is believed, productivity levels. About 10 per cent of the variance in performance can be explained through measures of support like the ones that Roadhes and Eisenberger (2002) found in their meta-analysis. However, the relationship between support measures and performance indicators deserves further clarification. Contrary to (H1), supervisory and colleagues support reduced performance; those perceiving higher levels of support achieved less production but increased TPM. An opposite pattern was found when analysing the influence of acknowledgement and rewards. As expected, acknowledgment and rewards positively affected performance, yet TPM was found to decrease. These results are possibly due to different correlation caused by the often informal type of bond between colleagues and supervisors, whereas the support received from acknowledgment and rewards is a more formal type of support. Colleague’s support has more influence in the quality of production while acknowledgment and rewards in the quantity of production. Even so, in order to obtain all implications, further analysis would be needed. Longitudinal analysis showed that organisational support did not predict performance one year later. H2 and H4 test the buffer-effect (direct and indirect respectively) of support in the stress situation generated by the implementation of work groups. Although some considerable interactions were found, no case was significant enough to be identified as buffer effect. This fact was also observed in the studies carried out by Schaubroeck and Fink (1998) and Iwata and Suzuki (1997).

The lack of evidence for the buffering effect of support in organisational settings has been related to a weak research design (Ganster et al., 1986, LaRocco et al., 1980). According to Terry et al. (1995) the lack of specificity on measures of social support and other reasons such as lack of variability and the slight magnitude of the samples, can account for the failure of previous studies to provide clear support for the stress-buffering model. The complexity of existing links of support to workplace outcomes depends on a number of important variables including the sources of support and the objective versus subjective nature of the results among others (Gerstner and Day, 1997; Schriesheim et al., 1999). These findings highlight the importance of social support in organisational settings (Leather et al., 1998), in particular in the implementation of working teams (Campion et al., 1993; Gladstein, 1984; Hackman, 1987; Sundstrom et al., 1990; Tannenbaum et al., 1996). It also emphasises the necessity to distinguish the different stages of support from colleagues, supervisors or the organisation as also suggested by other authors (Roadhes and Eisenberger, 2002; Stinglhamber and Vanderbergue, 2003). At this point in time, business environment companies aim to optimise their resources by reducing costs and maintaining high standards in quality of life at work. In this framework, the results of this study are of special relevance. Given that one of the main objectives of the implementation of working teams is to increase productivity and according to the present research, support to employees should be considered as a method to improve work performance as well as job satisfaction and involvement. Managers should be aware of this fact when designing intervention programs at their organisations. This study indicates that support is an easy and inexpensive means to, not only improve the social environment of an employee, but also to make an important contribution towards productivity targets. Limitations and future directions A main limitation of this study comes from the objective measures used. As mentioned above, these measures are similar for the production line as a whole and therefore they do not consider the possible differences in performance between different production line groups. Even though this indicator reduces the variability of data, the results highlight the interest of considering this kind of information. However, an advantage of this data collection method is that it avoids rivalry, conflicts between the different work shifts that constitute the production line and, in this case, is also the method in place for monitoring productivity. In the past there has been a proliferation of productivity measures (Mishra and Gupta, 1994). Even though this type of measures have some limitations it was necessary for the research to include them as a source of increasing its relevance and creditability. For example, the introduction of advanced manufacturing technology in teamwork system has seen an increase in the diversity of production systems within the sector. As a result, the operational measures that these firms employ vary quite widely. This makes a systematic comparison across organisations difficult. For some authors, the relationships analysed in this study are more complex and include organisational as well as personal aspects. Among the organisational aspects, Moideenkutty et al. (2001) points out the job characteristics and participation in decision making as a social support cause. Dispositional and moderating variables of the relation between social support and its consequences have been pointed out among

Organisational support and group efficacy 307

JMP 20,3/4

308

the personal variables. Therefore, the need for analysing personal variables as self-esteem or perceived control and their influence, not only on the use of social support but also on the acceptance and openness towards organisational changes, is stressed (Wanberg and Banas, 2000). As moderation variables of support, Armeli et al. (1998) considered the socio-emotional necessities, Eisenberger et al. (2001) the negative affectivity and the feeling of obligation, while Terry et al. (1995) analysed coping styles. What is eventually pointed out in these studies is the necessity of elaborating more complex models. In forthcoming studies, it would be important to improve the measures of support. Although the authors are aware of the limitations of this investigation tool since it does not include all the aspects analysed in the studies of social support, it is believed that it fulfils the needs for which it was originally intended. For example, given that some studies found that in organisational situations the supervisory support is more important (Dormann and Zapf, 1999) than originally thought, it would be interesting to distinguish between the supervisor and workmate support and whether it could even be an interaction between them both (Elfering et al., 2002). On the other hand, the measure of job involvement used does not include the remaining organisational aspects that comprise the organisational commitment’s wider dimension. Other aspects such as the wish to continue and normative commitment (Allen and Meyer, 1990, 1996) could form an interesting basis for further work. To conclude it is worth mentioning that this research was conducted using a single manufacture organisation where respondents were primarily employed within the production area, limiting the generability of the results to other industries and work settings. References Allen, N.J. and Meyer, J.P. (1990), “Organizational socialization tactics: a longitudinal analysis of links to newcomers’ commitment and role orientation”, Academy of Management Journal, Vol. 33 No. 4, pp. 847-58. Allen, N.J. and Meyer, J.P. (1996), “Affective, continuance, and normative commitment to the organization: an examination of construct validity”, Journal of Vocational Behavior, Vol. 49 No. 3, pp. 252-76. Armeli, S., Eisenberger, R.T., Fasolo, P. and Lynch, P. (1998), “Perceived organizational support and police performance: the moderating influence of socio-emotional needs”, Journal of Applied Psychology, Vol. 83 No. 2, pp. 88-297. Banker, R.D., Field, J., Schroeder, R.G. and Sinha, K.K. (1996), “Impact of work teams on manufacturing performance: a longitudinal study”, Academy of Management Journal, Vol. 39 No. 4, pp. 867-90. Baruch-Feldman, C., Brondolo, E., Ben-Dayan, D. and Schwartz, J. (2002), “Sources of social support and burnout, job satisfaction, and productivity”, Journal of Occupational Health Psychology, Vol. 7 No. 1, pp. 84-93. Beehr, T.A., King, L. and King, D. (1990), “Social support and occupational stress: talking to supervisors”, Journal of Vocational Behavior, Vol. 36 No. 1, pp. 61-81. Bennet, P., Lowe, R., Mattwes, V., Dourali, M. and Tattersall, A. (2001), “Stress in nurses: coping, managerial support and work demand”, Stress and Health: Journal of the International Society for the Investigation of Stress, Vol. 7 No. 1, pp. 55-63.

Burke, R.J. and Greenglass, E.R. (2001), “Hospital restructuring and nursing staff wellbeing: the role of perceived hospital and union support”, Anxiety, Stress and Coping: An International Journal, Vol. 14 No. 1, pp. 93-115. Campion, M.A., Medsker, G. and Higgs, A.C. (1993), “Relations between work group characteristics and effectiveness: implications for designing effective work groups”, Personnel Psychology, Vol. 36 No. 4, pp. 823-50. Cohen, S. and Wills, T.A. (1985), “Stress, social support, and the buffering hypothesis”, Psychological Bulletin, Vol. 98 No. 2, pp. 310-57. Dormann, C. and Zapf, D. (1999), “Social support, social stressors at work, and depressive symptoms: testing for main and moderating effects with structural equations in a three-wave longitudinal study”, Journal of Applied Psychology, Vol. 84 No. 6, pp. 874-84. Dubinsky, A.J., Yammarino, F.J. and Jolson, M.A. (1995), “An examination of linkages between personal characteristics and dimensions of transformational leadership”, Journal of Business and Psychology, Vol. 9 No. 3, pp. 315-35. Eisenberger, R., Armeli, S. and Rexwinkel, B. (2001), “Reciprocation of perceived organizational support”, Journal of Applied Psychology, Vol. 86 No. 1, pp. 42-51. Eisenberger, R., Cummings, J. and Armeli, S. (1997), “Perceived organizational support, discretionary treatment, and job satisfaction”, Journal of Applied Psychology, Vol. 82 No. 5, pp. 812-20. Eisenberger, R., Fasolo, P. and Davis-LaMastro, V. (1990), “Perceived organizational support and employee diligence, commitment, and innovation”, Journal of Applied Psychology, Vol. 75 No. 1, pp. 51-9. Eisenberger, R., Huntington, R., Hutchison, S. and Sowa, D. (1986), “Perceived organizational support”, Journal of Applied Psychology, Vol. 71 No. 3, pp. 500-7. Elfering, A., Semmer, N.K., Schade, V., Grund, S. and Boos, N. (2002), “Supportive colleague, unsupportive supervisor: the role of provider-specific constellations of social support at work in the development of low back pain”, Journal of Occupational Health Psychology, Vol. 7 No. 2, pp. 130-40. Frese, M. (1999), “Social support as a moderator of the relationship between work stressors and psychological dysfunctioning: a longitudinal study with objective measures”, Journal of Occupational Health Psychology, Vol. 4 No. 3, pp. 179-92. Frone, M., Yardley, J.K. and Markel, K.S. (1997), “Developing and testing an integrative model of the work-family interface”, Journal of Vocational Behavior, Vol. 50 No. 2, pp. 145-67. Ganster, D.C., Fusilier, M.R. and Mayes, B.T. (1986), “Role of social support in the experience of stress at work”, Journal of Applied Psychology, Vol. 71 No. 1, pp. 102-10. Gerstner, C.R. and Day, D.V. (1997), “Meta-analytic review of leader-member exchange theory: correlates and construct issues”, Journal of Applied Psychology, Vol. 82 No. 6, pp. 827-44. Gladstein, D.L. (1984), “Groups in context: a model of task group effectiveness”, Administrative Science Quarterly, Vol. 29 No. 4, pp. 499-517. Gooding, R.Z. and Wagner, J.A. (1985), “A meta-analytic review of the relationship between size and performance: the productivity and efficiency of organizations and their subunits”, Administrative Science Quarterly, Vol. 30 No. 4, pp. 462-81. Gouldner, A.W. (1960), “The norm of reciprocity: a preliminary statement”, American Sociological Review, Vol. 25, pp. 161-78. Green, S.G., Anderson, S. and Shivers, S.L. (1996), “Demographic and organizational influences on leader-member exchange and related work attitudes”, Organizational Behavior and Human Decision Processes, Vol. 66 No. 2, pp. 203-14.

Organisational support and group efficacy 309

JMP 20,3/4

310

Guzzo, R.A. and Dickson, M.W. (1996), “Teams in organizations: recent research on performance and effectiveness”, Annual Review of Psychology, Vol. 47, pp. 307-38. Guzzo, R., Noonan, K.A. and Elron, E. (1994), “Expatriate managers and the psychological contract”, Journal of Applied Psychology, Vol. 79 No. 4, pp. 617-26. Hackman, J.R. (1987), The Design of Work Teams: Handbook of Organizational Behavior, Prentice-Hall, New York, NY. Hackman, J.R. and Oldham, G.R. (1976), “Motivation through the design of work: tests of a organizational theory”, Behavior and Human Performance, Vol. 6, pp. 157-60. Hyatt, D.E. and Ruddy, T.M. (1997), “An examination of the relationship between work group characteristics and performance: once more into the breach”, Personnel Psychology, Vol. 50 No. 3, pp. 553-85. Iwata, N. and Suzuki, K. (1997), “Role stress-mental health relations in Japanese bank workers: a moderating effect of social support”, Applied Psychology: An International Review, Vol. 46 No. 2, pp. 207-18. Karasek, R.A. and Theorell, T. (1990), Healthy Work: Stress and Productivity and the Reconstructuion of Working Life, Basic Books, New York, NY. LaRocco, J.M., House, J.S. and French, J.R. (1980), “Social support, occupational stress, and health”, Journal of Health and Social Behavior, Vol. 21 No. 3, pp. 202-18. Larson, C.E. and LaFasto, F.M. (1989), Teamwork: What Must Go Right/What Can Go Wrong, Sage Publications, Thousand Oaks, CA. Leather, P., Lawrence, C., Beale, D., Cox, T. and Dickson, R. (1998), “Exposure to occupational violence and the buffering effects of intra-organizational support”, Work and Stress, Vol. 12 No. 2, pp. 161-78. Mishra, P.C. and Gupta, J. (1994), “Performance as a function of employees’ motivation and job involvement”, Psychological Studies, Vol. 39 No. 1, pp. 18-20. Moideenkutty, U., Blau, G., Kumar, R. and Nalakath, A. (2001), “Perceived organisational support as a mediator of the relationship of perceived situational factors to affective organisational commitment”, Applied Psychology: An International Review, Vol. 50 No. 4, pp. 615-34. Nye, L.G. and Witt, L.A. (1993), “Dimensionality and construct validity of the perceptions of organizational politics scale (POPS)”, Educational and Psychological Measurement, Vol. 53 No. 3, pp. 821-9. Organ, D.W. (1988), “A restatement of the satisfaction-performance hypothesis”, Journal of Management, Vol. 14 No. 4, pp. 547-57. Peiro´, J.-M., Melia´, J.L., Torres, M.A. and Zurriaga, R. (1986), “La medida de la experiencia de la ambigu¨edad en el desempen˜o de los roles: el cuestionario general de ambigu¨edad de rol en ambientes organizacionales (“The measurement of the ambiguity experience when performing a role: the general questionnaire of role ambiguity in organisational environments”), Evaluacio´n Psicolo´gica, Vol. 3, pp. 27-53. Randall, M., Cropanzano, R., Bormann, C.A. and Birjulin, A. (1999), “Organizational politics and organizational support as predictors of work attitudes, job performance, and organizational citizenship behavior”, Journal of Organizational Behavior, Vol. 20 No. 2, pp. 159-74. Rizzo, J.R., House, R.J. and Lirtzman, S.I. (1970), “Role conflict and ambiguity in complex organizations”, Administrative Science Quarterly, Vol. 15 No. 2, pp. 150-63. Roadhes, L. and Eisenberger, R. (2002), “Perceived organizational support: a review of the literature”, Journal of Applied Psychology, Vol. 87, pp. 698-714.

Roadhes, L., Eisenberger, R. and Armeli, S. (2001), “Affectively commitment to the organization: the contribution of perceived organizational support”, Journal of Applied Psychology, Vol. 86, pp. 825-36. Rousseau, D.M. (1989), “Psychological and implied contracts in organizations”, Employee Responsibilities and Rights Journal, Vol. 2 No. 2, pp. 121-39. Schaubroeck, J. and Fink, L.S. (1998), “Facilitating and inhibiting effects of job control and social support on stress outcomes and role behavior: a contingency model”, Journal of Organizational Behavior, Vol. 19 No. 2, pp. 167-95. Schriesheim, C., Gables, F., Castro, S.L. and Cogliser, C. (1999), “Leader-member exchange (LMX) research: a comprehensive review of theory, measurement, and data-analytic practices”, Leadership Quarterly, Vol. 10 No. 1, pp. 63-113. Stinglhamber, F. and Vanderbergue, C. (2003), “Organizational and supervisors as sources of support and targets of commitment: a longitudinal study”, Journal of Organizational Behavior, Vol. 24, pp. 251-70. Sundstrom, E., de Meuse, K.P. and Futrell, D. (1990), “Work teams: applications and effectiveness”, American Psychologist, Vol. 45 No. 2, pp. 120-33. Tannenbaum, S.I., Salas, E. and Cannon-Bowers, J.A. (1996), “Promoting team effectiveness”, in West, M.A. (Ed.), Handbook of Work Group Psychology, John Wiley & Sons, London, pp. 503-29. Teague, G.B., Drake, R.E. and Ackerson, T.H. (1995), “Evaluating use of continuous treatment teams for persons with mental illness and substance abuse”, Psychiatric Services, Vol. 46 No. 7, pp. 689-95. Terry, D., Rawle, R. and Callan, V.J. (1995), “The effects of social support on adjustment to stress: the mediating role of coping”, Personal Relationships, Vol. 2 No. 2, pp. 97-124. Thomas, L. and Ganster, D.C. (1995), “Impact of family-supportive work variables on work-family conflict and strain: a control perspective”, Journal of Applied Psychology, Vol. 80 No. 1, pp. 6-15. Vecchio, R. (1998), “Leader-member exchange, objective performance, employment duration, and supervisor ratings: testing for moderation and mediation”, Journal of Business and Psychology, Vol. 12 No. 3, pp. 327-41. Viswesvaran, C., Sanchez, J.I. and Fisher, J. (1999), “The role of social support in the process of work stress: a meta-analysis”, Journal of Vocational Behavior, Vol. 54 No. 2, pp. 314-34. Wanberg, C.R. and Banas, J.T. (2000), “Predictors and outcomes of openness to changes in a reorganizing workplace”, Journal of Applied Psychology, Vol. 85 No. 1, pp. 132-42. Wayne, S.J., Shore, L.M. and Liden, R.C. (1997), “Perceived organizational support and leader-member exchange: a social exchange perspective”, Academy of Management Journal, Vol. 40 No. 1, pp. 82-111.

Organisational support and group efficacy 311

The Emerald Research Register for this journal is available at www.emeraldinsight.com/researchregister

The current issue and full text archive of this journal is available at www.emeraldinsight.com/0268-3946.htm

JMP 20,3/4

Change-oriented leadership, satisfaction and performance in work groups

312

Effects of team climate and group potency Francisco Gil Universidad Complutense de Madrid, Madrid, Spain

Ramo´n Rico Universidad Auto´noma de Madrid, Madrid, Spain

Carlos M. Alcover Universidad Rey Juan Carlos, Madrid, Spain, and

´ ngel Barrasa A Universidad Complutense de Madrid, Madrid, Spain Abstract Purpose – To analyse the impact of change-oriented leaders on group outcomes. An explanatory model is proposed, in which the team climate (in particular as it relates to innovation) mediates between change-oriented leadership and group outcomes, while group potency reinforces this relationship. Design/methodology/approach – This study is designed as a correlative and cross-level research. The sample comprises 318 health-care professionals in 78 health-care teams at different public hospitals throughout Spain. Findings – Hierarchical regression analysis was used to evaluate mediating and moderating effects. Results offer considerable empirical support for the proposed model. Research limitations/implications – It would be of interest to increase the sample, differentiate it by service, and to get samples from other sectors, as well as to carry out experimental and longitudinal research. It would also be interesting to further explore the conditions that implement change-oriented leadership impact, analysing environment, external relations and so on, to examine the relationships between other variables and to study their effects on new forms of work organisation and on virtual teams. Practical implications – To make more useful change-oriented leader actions, it would be advisable to identify, modify or improve team climate, using strategies such as management by objectives, delegation and empowerment and so on. It would also be necessary to boost group potency before going ahead with change, for example, by developing the skills of team members, or by fostering the self-confidence of the team. Originality/value – This paper contributes to developing actual research about how change-oriented leaders influence team outputs. Keywords Transformational leadership, Team performance, Job satisfaction, Spain Paper type Research paper Journal of Managerial Psychology Vol. 20 No. 3/4, 2005 pp. 312-328 q Emerald Group Publishing Limited 0268-3946 DOI 10.1108/02683940510589073

This research has been founded by the Universidad Complutense of Madrid (PR3/04-12455). The authors are grateful to G. Yukl who provides Managerial Practices Survey (TRCQ-15G) questionnaire for research purposes and to L. Munduate and F.J. Medina from Sevilla University for their collaboration.

Introduction Teams form the basic functional unit of organisations (Mohrman et al., 1995). Practically all them, including healthcare institutions (Heinemann and Zeiss, 2002; Poole and Real, 2003), use teams in one way or another. The organisational outcomes therefore depend on appropriate design and the proper functioning of work units and teams (Guzzo and Dickson, 1996). Research has revealed the importance of group structures and processes for team outcomes, and numerous models and constructs have been proposed (see Salas et al., 2004). One main variable is leadership, even so, only a few models (Avolio et al., 1996; Gladstein, 1984; Kozlowski et al., 1996; Stewart and Barrick, 2000) explicitly take leadership into account as a determining factor in team outcomes. Nevertheless, leadership (or the lack of it) has been identified as a key variable for the functioning of teams and one of the main reasons for the success or failure with which team-based work systems are implemented (Katzenbach, 1997). The leadership research is based on the classical bi-factorial models, although more recently has been developed “new paradigms”, as charismatic and transformational leadership, among others. The links between different forms of leadership and team outcomes as proposed in these models are not, however, wholly consistent (Bass et al., 2003; Stewart and Barrick, 2000), which suggests a need for further research. The tri-dimensional leadership model, recently developed by Yukl (Yukl, 2004; Yukl et al., 2002), identifies three major categories, adding change dimension to the classical bi-factorial models (task and relation-oriented leadership). By proposing the incorporation of the category of change, the tri-dimensional model allows the integration of the two major traditions of management and leadership theory, which have normally stood apart, each with its own literature. Rather than seeking to establish distinctions between managers and leaders, the two can be explained jointly using the same processes and models (Yukl, 2002). The view that both leaders and managers employ a mix of leadership and management behaviours appears much closer to reality so that they must combine the necessary skills to direct day-to-day affairs effectively (a role traditionally associated with management) while at the same time anticipating and managing change (leadership main role). Change-oriented leadership As a consequence of globalisation, application of new technologies, coping with a turbulent environment, etc., organisations face with ongoing processes of transformation. They assign the responsibility of anticipating change and providing guidance to their managers, executives and leaders (Kotter, 1990) who need new roles (Shamir, 1999). The transformational and charismatic leadership theories (Bass, 1985; Conger and Kanungo, 1988; House, 1977) refer to certain kinds of change-oriented behaviour, and there is considerable evidence that such patterns are related to effective leadership, as shown in the meta-analysis carried out by Lowe et al. (1996). The tri-dimensional model proposed by Yukl (2004) is compatible with the transformational and charismatic leadership theories, although its aim is in fact to explain leadership processes at a different conceptual level of analysis. This model seeks to describe the influence of leaders on organisational processes (rather than on the motivation and perceptions of subordinates), analyse contingent (as opposed to universal) aspects of effective leadership, and highlight the importance of leadership

Change-oriented leadership

313

JMP 20,3/4

314

processes (instead of focusing on a leader figure). It also represents an effort to identify the behaviour patterns that make up each category, in such away that: each type of behaviour is observable, is potentially applicable to leaders of all kinds in organisations, is fundamentally relevant to the category in question, and is based on prior theory and research. Change management is raised in a variety of organisational theories (Tushman and Romanelli, 1985) and is supported by current research (Ekwall and Arvonen, 1991; Gil et al., 2003; Yukl, 1998; Yukl et al., 2002). This category comprises the following behaviour sets: monitoring the environment, encouraging innovative thinking, explaining need for change, envisioning change, and taking personal risks. Model of the relationship between change-oriented leadership, satisfaction and performance The aim of this research is to analyse the influence the change-oriented leader can have on the team outcomes, performance and satisfaction, mediated by some team processes. We propose (see Figure 1) that the relationship between leadership and outcomes is mediated by the team climate, and that this mediation is reinforced by group potency. Group climate Team climate have been defined as shared perceptions referring to the “proximal work group”. This is considered as the “permanent or semi-permanent team to which individuals are assigned, whom they identify with, and whom they interact with regularly in order to perform work-related tasks” (Anderson and West, 1998, p. 236). These authors developed the team climate inventory (TCI) applied to innovation and identified four factors: vision, participation, task orientation, and support for innovation. The last one is defined as “. . . the expectation, approval and practical support of attempts to introduce new and improved ways of doing things in the work environment” (West, 1990, in Anderson and West, 1998, p. 240). Of these four factors, support for innovation has been confirmed as the most consistent predictor of team innovations in external evaluations (Burningham and West, 1995). The TCI questionnaire was developed and is mainly used to evaluate the predictive dimensions of innovation, though it is also considered useful in appraisals of other group outcomes (Anderson and West, 1998). Although leadership and climate are two variables that are implicitly interlinked in research, theoretical development and empirical research are limited. In the early theories and research, leadership is proposed as an organisational factor affecting

Figure 1. Proposed model

perceptions of climate. In the classic study by Lewin et al. (1939), it was observed that varying leadership styles induced experimentally (authoritarian, democratic and laissez faire) influenced perceptions of climate and the behavioural responses of subjects. Litwin and Stringer (1968), who created organisations directed by leaders exhibiting styles (bureaucratic, cooperative and productivity-oriented), obtained similar results. Kozlowski and Doherty (1989) find that the interaction between the leader and the subordinate had a significant impact on perceptions of climate. Gonza´lez-Navarro et al. (1993) analysed leader-member relations in primary healthcare teams, confirming the positive relationship between leadership styles focusing on people or tasks and perceptions of climate in various dimensions (support, goals, innovation and rules). However, they were unable to substantiate other hypotheses related with perceptions of the influence of the coordinator. Man˜as et al. (1999), meanwhile, performed a longitudinal study in which they demonstrated concurrent effects in the leader-members relationship, but not deferred effects. Studies of the relationship between leadership and climate have been confined to exploring the links between the classical dimensions of leadership (focusing on the task and the relationship), in particular through the analysis of leader-member interactions (Kozlowski and Doherty, 1989; Man˜as et al., 1999; Gonza´lez-Navarro et al., 1993), but they do not consider change or similar leadership dimension. Furthermore, the majority of these studies use measures of organisational climate rather than specific team climate measures such as those provided by the TCI. A model has recently been proposed (West and Hirst, 2003), in which these climate variables mediate the relationship between the context (both group and organisational) and innovation, assigning a moderating role to leadership, but this model has not been definitely proved. A number of studies (Bass et al., 2003; Carron, 1982; Spink, 1998) provide empirical support for the mediation effect of group processes between leadership and outcomes. Team climate, as it provides an indicator of significant group processes, has a mediating role between leadership and outcomes (performance and satisfaction). We propose the same relationship to change-oriented leadership, as well that to the climate of innovation, the most closely linked process. H1. The relationship between change-oriented leadership and team performance is mediated by the global climate (H1a) and by the climate of innovation (H1b). H2. The relationship between change-oriented leadership and team satisfaction is mediated by the global climate (H2a) and the climate of innovation (H2b). Group potency Potency is a construct between self-efficacy and collective motivation (Alcover and Gil, 2000), has been defined as “the collective belief in a group that it can be effective” (Guzzo et al., 1993, p. 87). Teams differ from each other depending on the collective belief of their members in their potential effectiveness as such. This belief is related with current levels of effectiveness, appears to act both as cause and consequence, and is influenced by the contexts within which groups act. Group potency has been identified as a significant cognitive influence on performance (Gil and Alcover, 2002; Guzzo et al., 1993; Jordan et al., 2002; Pearce et al.,

Change-oriented leadership

315

JMP 20,3/4

316

2002). Campion et al. (1993, 1996) found that group potency was a significant predictor not only of productivity, but also of the satisfaction of team members and management assessments of its performance. Group potency was the only variable that significantly predicted all measures in both studies. The meta-analysis recently carried out by Gully et al. (2002) confirms the positive relationship between potency and performance. The relationship of a number of group variables with potency has also been explored. These include flexibility in the composition of teams (Alcover and Gil, 1999), leadership and, in particular, transformational and team leadership (Bass et al., 2003; Kahai and Sosik, 1998; Sivasubramaniam et al., 2002; Sosik et al., 1997; Sosik et al., 1998). Shamir et al. (1993) explain that charismatic leadership may boost participation in group effort and can be linked to the collective identity, increasing potency and performance. The majority of models and studies concerning the relationship between leadership and group performance treat potency as a mediating variable (Bass et al., 2003; Lester et al., 2002; Sivasubramaniam et al., 2002). Potency may also, however, be considered as a variable that moderates the relationship between leadership and other variables. For example, Foels et al. (2000) have confirmed the existence of this moderating effect between democratic leadership and satisfaction. As far as leaders who promote change are concerned, the moderating effect of potency may be understood to the extent that it may be assumed teams will react differently to the leaders’ demands depending on their group potency. Thus, where the demand for change brings uncertainty and risk, it is likely that the more self-confident teams will more readily accept this with a positive impact on group processes (climate, and especially the climate of innovation). Referring to the challenges facing actual organisations, Shamir (1999) assigns to leaders the difficult tasks of instilling a sense of psychological safety to help people cope with the anxiety inherent in uncertainty and change, and of providing the conditions of stability and continuity necessary for individual and organisational learning. Some of these conditions will be met when teams have a high level of group potency. Finally, though the influence of different group processes on potency has been examined (Lester et al., 2002), one unexplored factor is the relationship between group potency and other beliefs about the team, in particular as regards the team climate. We propose the following hypotheses in relation to the moderating effect of potency on team climate. H3. The relationship between change-oriented leadership and team climate (H3a) and climate of innovation (H3b) are moderated by potency with positive effects. Similarly, we predict a more general effect, such that potency influences the climate mediation process between leadership and outcomes variables. H4. Potency positively reinforces global team climate (H4a) and climate innovation (H4b) mediation between change-oriented leadership and team performance. H5. Potency positively reinforces global team climate (H5a) and climate innovation (H5b) mediation between change-oriented leadership and team satisfaction.

Method Sample The sample comprises 318 healthcare professionals of 78 healthcare teams at different public hospitals throughout Spain (Madrid, Barcelona, Ma´laga, La Corun˜a, Sevilla and Ca´diz). A total of 11 teams were discarded because they do not complete an appropriate number of responses from members (teams with a response rate of below 30 per cent of total members) or because they lacked at least two external measures of performance. A total of 67 teams formed the final sample. In all cases, the organisations were formally designed around work units (teams). The work teams were defined in accordance with the proximal work group concept discussed above. The size of the work teams ranged from 3 to 24 members, with an average per team of 10.6 (SD ¼ 5:33). The response rate obtained was 68.4 per cent. Females comprise 66.4 per cent of the sample. The average age of subjects was 41.4 years (SD ¼ 8:65). Measures Because this study was performed at the level of the team, having collected questionnaires at the level of the individual, it was necessary first to aggregate data in order to obtain the team construct. The intra-class correlation (ICC) index provides an indication of the extent to which the perceptions of group members are shared, it compares the inter-group with the intra-group variance (Chan, 1998; Kenny and La Voie, 1985). The higher the ICC index, the greater is the variance at the individual level attributable to the relevant team. Normally, an ICC of over 0.20 is considered to indicate that a variable may be labelled a group attribute, justifying aggregation. All of the aggregate variables are significantly higher than this threshold: change-oriented leadership (0.60), group potency (0.59), team climate (0.55), climate of innovation (0.53) and satisfaction (0.52). Change-oriented leadership. Behaviour associated with change-oriented leadership is evaluated using a recent version of questionnaire, managerial practices survey (TRCQ-15G), designed by Yukl on the basis of earlier inventories (Yukl et al., 2002). The questionnaire comprises three scales: task, relation and change-oriented leadership. Earlier studies have demonstrated that the psychometric characteristics of this questionnaire are appropriate (Yukl, 1998; Yukl et al., 2002; Gil et al., 2003). The change-oriented leadership scale contains five subscales with four items each. These subscales are monitoring the environment (a ¼ 0:90); encouraging innovative thinking (a ¼ 0:66); explaining need for change (a ¼ 0:95); envisioning change (a ¼ 0:93); and taking personal risks (a ¼ 0:96). Some items of the first two subscales are: “analyses external events and trends to identify threats and opportunities” and “asks people to look at a problem from a different perspective”. The response emphasises magnitude rather than frequency (1 ¼ not at all, 5 ¼ to a great extent, with the option of “don’t know” or “not applicable”). An aggregate measure was obtained at the team level (ICC ¼ 0.60). Team climate. We used the TCI designed by Anderson and West (1994). The inventory contains 38 items (a ¼ 0:96; ICC ¼ 0:55) with five-point Likert responses (1 ¼ disagree completely, 5 ¼ completely agree) grouped into four factors comprising objectives (11 items, a ¼ 0:93); participation (12 items, a ¼ 0:94); task orientation (7 items, a ¼ 0:84); and innovation (8 items a ¼ 0:82, ICC ¼ 0:53). Some items are “everyone’s view is listened to, even if it is in a minority” (participation) and “the team is open and responsive to change” (innovation).

Change-oriented leadership

317

JMP 20,3/4

318

Group potency was assessed using Guzzo et al. (1993) scale of 8 items (a ¼ 0:88; ICC ¼ 0:59). Some items are “this team believes it can become unusually good at producing high-quality work” and “this team feels it can solve any problem it encounters”. Responses were scored using a six-point Likert scale (1 ¼ disagree completely, 6 ¼ completely agree). Satisfaction. Team satisfaction was assessed using Gladstein’s (1984) scale of three items (a ¼ 0:85; ICC ¼ 0:52), which indicate the degree to which subjects display satisfaction with their colleagues, the manner of team working and with the team as a whole. Responses were scored using a five-point Likert scale (1 ¼ disagree completely, 5 ¼ completely agree). Team performance. Team performance was assessed via external supervisors and managers with a good knowledge of the team. Each team has been scored as a unit. A scale applied by Ancona and Caldwell (1992) was used. This comprises of five items related to team’s efficiency, quality of technical innovations, adherence to schedules, adherence to budgets and ability to resolve conflicts (a ¼ 0:83). Each dimension was scored by managers using a five-point Likert scale (1 ¼ disagree completely, 5 ¼ completely agree). Between two and three evaluations were obtained from various supervisors and managers (teams without at least two such evaluations were discarded), resulting in an inter-judge coefficient of 0.74. Finally, the control measures employed were team size (number of team members) with an average of 10.6 (SD ¼ 5.33), and team tenure (time each member form part of the team) with an average of 9.6 (SD ¼ 5.49). The ICC was 0.59. Procedure Through human resources departments of each hospital we held meetings with chiefs and managers responsible for the work units concerned to explain the research project. Team members were invited to participate voluntarily by completing an individual and anonymous questionnaire. External supervisors and managers were also asked to complete a specific questionnaire, also individually and anonymously, to score group performance. Results Descriptive statistics and correlation coefficients for all of the variables are presented in Table I. The diagonal line reflects the Cronbach a for the scales used in this study. Firstly, the team size and tenure variables are not correlated with any of the variables M

Table I. Means, standard deviations and correlations of variables at team level

1. 2. 3. 4. 5. 6. 7. 8.

DT

1

Change-oriented leadership 3.00 0.86 (0.94) Group potency 4.18 0.84 0.60** Team climate 3.34 0.61 0.63** Team climate. Innovation 3.22 0.72 0.53** Team size 10.59 5.33 0.15 Team tenure 9.58 5.49 0.07 Team effectiveness 3.77 0.69 0.46** Satisfaction 3.49 0.75 0.51**

Notes: *p , 0.05;

**

2

3

4

(0.88) 0.86** 0.83** 0.03 0.09 0.54** 0.84**

(0.96) 0.93** 0.10 0.05 0.56** 0.85**

(0.82) 0.02 0.00 0.56** 0.82**

5

6

7

8

– 0.16 – 0.26* 0.15 – 0.09 0.18 0.55** (0.85)

p , 0.01; internal consistency of the scales on the diagonal

forming part of the model tested, with the exception of the first variable with group performance. Change-oriented leadership and potency are significantly correlated ( p , 0.01 in both cases) with group performance and satisfaction respectively. The relationship between potency and satisfaction is particularly high (r ¼ 0:84). The climate measures, meanwhile, are also significantly correlated among themselves ( p , 0.01) and high with r ¼ 0:93. These climate measures are also significantly correlated ( p , 0.01) with performance and satisfaction, although these relationships are stronger over all in the latter case. In order to test H1a and H1b regarding the mediation of team climate as group process in the relationship between change-oriented leadership and team performance, we carried out a set of hierarchical regressions along the lines described by Baron and Kenny (1986) for such cases. As shown in Table II, the global climate and the climate of innovation mediate the relationship between change-oriented leadership and performance. In both cases, the results reflected in Table II reveal that the effect of change-oriented leadership on team performance diminishes when the global climate and the innovation climate are controlled. Signification results for changes in the coefficients following the Sobel (1982) test were significant for global climate (z ¼ 2:93; p , 0.01) and for the climate of innovation (z ¼ 2:72; p , 0:01). The hierarchical regression analysis used to test the mediating effect of team climate and innovation climate (H2a and H2b) on the relationship between change-oriented leadership and group satisfaction produced results to support both hypotheses (see Table II). Thus, we observed an increment of 0.48 and 0.43 respectively for R 2, both being significant at the level of 0.01 for the purposes of controlling the effect of the three perceptions of climate on the relationship between change-oriented leadership and satisfaction. Checks performed using the Sobel (1982) test were also significant in the case of the global climate (z ¼ 6:20; p, 0.01) and for the climate of innovation (z ¼ 4:83, p , 0.01). H3a and H3b respectively predicted that potency would have a moderating effect on the relationship between change-oriented leadership and the team members’ perceptions of climate. These hypotheses were tested separately for global climate and for the innovation climate subscale using hierarchical regressions. This moderation would be supported by a significant change in the square of the multiple correlation

Step

Independent variable

1

Team size Team tenure Change-oriented leadership Change-oriented leadership Team climate Team size Team tenure Change-oriented leadership Change-oriented leadership Team climate. Innovation

2 3 1 2 3

Notes: an = 67 (teams); *p , 0.05;

b

**

Testing for mediation affectinga Performance Satisfaction R2 DR 2 b R2

0.02 20.00 0.39** 0.03 0.47** 0.02 20.00 0.39** 0.11 0.41** p , 0.01

0.08

0.08

0.25 0.35

0.17** 0.09*

0.08

0.08

0.25 0.34

**

0.17 0.09**

0.07 20.00 0.50** 20.04 0.89** 0.07 20.00 0.50** 0.10 0.76**

319

DR 2

0.00

0.00

0.25 0.73

0.25** 0.48**

0.00

0.00

0.25 0.68

Change-oriented leadership

**

0.25 0.43**

Table II. Results of hierarchical regression analyses testing for mediation affecting performance and satisfaction

JMP 20,3/4

320

coefficient (R 2) where the interaction between change-oriented leadership and group potency was included. As reflected in Table III, such moderation does indeed appear for the global team climate (DR 2 ¼ 0:02; p, 0.01) and for the climate of innovation (DR 2 ¼ 0.03; p , 0.01) (see also Figures 2 and 3). In order to verify H4a and H4b regarding the combination of effects moderating group potency in the mediation tested in H1a and H1b (mediation of team climate and innovation climate between change-oriented leadership and performance), the teams were divided using the median as the cut-off point (Mdn ¼ 4:20) into high (M ¼ 4:87; SD ¼ 0:40) and low (M ¼ 3:45; DT ¼ 0:48) group potency classes. Separate hierarchical regression analyses were performed on each class. This verified the Step

Independent variable

1

Team size Team tenure R2 Change-oriented leadership R2 DR 2 Group potency R2 DR 2 Change-oriented leadership £ group potency R2 DR 2

2 3

Table III. Results of hierarchical regression analyses testing for moderation

Figure 2. Interaction effect of team potency and change-oriented leadership on team climate

4

Notes: an = 67 (teams); *p , 0.05;

**

p , 0.01

Team climatea

Team climate. Innovationa

0.01 0.01 0.01 0.61** 0.37 0.36** 0.75** 0.75 0.36** 0.16 0.78 0.02**

0.02 20.14 0.02 0.51** 0.28 0.26** 0.82** 0.71 0.43** 0.17** 0.74 0.03**

Change-oriented leadership

321

Figure 3. Interaction effect of team potency and change-oriented leadership on team innovation climate

mediation of global climate and the innovation climate between change-oriented leadership and performance in high potency but not in low potency teams (see Table IV). The results reflected in this table reveal that the effect of change-oriented leadership on team performance diminishes when the global climate and the innovation climate are controlled, but only in teams with high group potency. Signification results for changes in the coefficients following the Sobel (1982) test were significant for global climate (z ¼ 2:47; p , 0.05) and for the climate of innovation (z ¼ 1:97; p , 0.05). Teams were divided into high and low potency groups in the same way in order to test H5a and H5b regarding the existence of a combination of the moderating effect of group potency on the mediation of team and innovation climate between change-oriented leadership and satisfaction. This verified that there is no mediation of global climate and the innovation climate between change-oriented leadership and satisfaction in high potency teams. Such mediation was, however, found to exist in low potency teams (see Table IV). The results reflected in this table reveal that the effect of change-oriented leadership on team performance diminishes when the global climate and the innovation climate are controlled, but only in teams with low group potency. Signification results for changes in the coefficients following the Sobel (1982) test were significant for global climate (z ¼ 2.31; p , 0.05), but not for the climate of innovation (z ¼ 1.14; p , 0.25). Discussion Summary of results and conclusions The results provide empirical support for H1a and H1b regarding performance, and for H2a and H2b regarding satisfaction. This confirms the existence of a general mediation effect of global climate, and of the innovation climate, on the relationship

p , 0.01

**

Notes: an = 33 (teams); *p , 0.05;

2 3

1

2 3

0.09 20.05 0.26 0.21 0.12 0.09 20.05 0.26 0.21 0.21

Team size Team tenure Change-oriented leadership Change-oriented leadership Team climate Team size Team tenure Change-oriented leadership Change-oriented leadership Team climate. Innovation

1

b

Independent variable

Table IV. Results of hierarchical regression analyses testing for mediation affecting performance and satisfaction for low and high potency teams 0.20 0.24

0.08

0.20 0.23

0.08

0.12 0.01

0.08

0.12 0.02

0.08

DR

2

0.22 0.00 0.29* 2 0.31 0.78** 0.22 0.00 0.29* 2 0.13 0.64**

b

Performance

0.19 0.35

0.08

0.19 0.39

0.08

High R2

0.11* 0.18**

0.08

0.11* 0.20**

0.08

DR

2

0.20 0.07 0.33* 0.13 0.76** 0.20 0.07 0.33* 0.26* 0.61**

b

0.23 0.51

0.09

0.23 0.57

0.09

Low R2

0.14* 0.28**

0.09

0.14* 0.33**

0.09

DR 2

2 0.01 2 0.04 0.15 2 0.24 0.81** 2 0.01 2 0.04 0.15 2 0.11 0.69**

b

Satisfaction

322

Step

Low R2

Testing for mediation affectinga

0.04 0.53

0.00

0.04 0.61

0.00

High R2

0.03 0.49**

0.00

0.03 0.57**

0.00

DR 2

JMP 20,3/4

between change-oriented leadership and both team outcomes, performance and satisfaction. Empirical support is also provided for H3a and H3b regarding the moderating effect of potency on the relationship between change-oriented leadership and global climate and innovation climate. This relationship is maximised in high potency teams, but is hardly visible for low potency teams. Finally, we have found uneven empirical evidence for the last hypotheses. Thus, we found empirical support for H4a and H4b regarding performance to the extent that the mediation effect is reinforced in high potency teams but vanishes in low potency teams. Contrary to our expectations, in H5a and H5b, relatives to satisfaction, mediation is reinforced in low potency teams and vanishes in high potency teams. In general, the results of present study offer considerable empirical support for the proposed model. The only unexpected result concerns the moderating effect of potency on the mediation of climate between leadership and satisfaction, which is contrary to the performance results. This may in part be because the two measures differ (aggregated subjective evaluations of satisfaction by the individual subjects in the first case, and external performance scorings by managers in the second). In any case, it is surprising that climate might mediate between the change-oriented leaders and satisfaction only in low potency teams. This fact might be explained by considering that proposals for change made by the leader may have certain attractiveness in that they presuppose innovation and improvement. Thus, satisfaction would increase to the extent that such proposals are launched in a favourable climate, since satisfaction is strongly associated with a positive climate. This does not, however, work for high potency teams, possibly because their own self-confidence is a powerful, and perhaps sufficient, source of satisfaction, which may diminish the influence of other variables, including the proposals made by the change-oriented leader or the existence of a positive climate. This explanation would need to be tested in subsequent research. Theoretical and applied implications. The confirmation, for the most part, of the proposed model supports the findings obtained from other studies (Bass et al., 2003; Carron, 1982; Sivasubramaniam et al., 2002; Spink, 1998) into the mediating role played by group processes between leadership and team outcomes. The present study also confirms that this occurs when climate is taken as the measure of group processes, both in terms of global climate and the climate of innovation. However, the potency variable, that is normally considered as a mediating variable (Bass et al., 2003; Guzzo et al., 1993; Lester et al., 2002; Man˜as et al., 1999; Sivasubramaniam et al., 2002) together with other group processes, in the present study appears rather as a moderating variable, given the major differences observed between high and low potency teams. These findings also have important applied implications. Firstly, given that the actions and strategies implemented by the change-oriented leader are mediated and moderated by other variables, it would be advisable to identify and, where necessary, modify such variables before embarking on such actions, which may at times give rise to considerable resistance. For example, if the influence of leadership on outcomes (in terms of both performance and satisfaction) is explained by climate, it may well be desirable for the leader to refrain from promoting change unless the climate is positive (and particularly in a climate that support the innovation). Where this is not so, is recommend undertaking prior actions to improve the different dimensions of the climate (e.g. objectives, participation, task orientation, support to innovation, etc.).

Change-oriented leadership

323

JMP 20,3/4

324

Various strategies oriented to these ends exist, such as management by objectives, delegation and empowerment and so on, all of which are widely recognised and actually applied within the organisations, including healthcare institutions. On the other hand, as this study itself shows, the behaviour of the change-oriented leader may contribute to foster a positive climate, especially in teams with high group potency. Other studies (Gonza´lez-Navarro et al., 1993; Gonza´lez-Roma´ et al., 1995; Man˜as et al., 1999) have demonstrated that the leader can mould the climate perceptions held by team members through social interactions, which may in itself improve the quality of working life. Since the influence of leadership on performance is moderated by potency, any attempt by the leader to promote changes in low potency teams (i.e. where members are not confident of their potential) will be an exercise in futility, even where conditions are favourable (positive climate). In such cases, could be recommend any intervention to boost team potency before going ahead with change. This could be achieved, for example, by developing the skills of team members (including the skills necessary to take on new tasks and to work together as a team in a coordinated manner), or by fostering the self-confidence of the team. Training actions might be planned on the one hand and, on the other, activities and tasks could be designed that were, not only attractive and innovative, but also easily carried out by the team, providing a challenge within the range of its potential. To the extent that the team may have the necessary skills to undertake new tasks and has the opportunity to test them appropriately and obtain feedback and reinforcement, it is likely that self-confidence will increase (Guzzo et al., 1993). Since satisfaction appears strongly associated with climate, it is also essential to improve the dimensions of climate in the manner we have just described. In the case of high potency teams, where self-confidence is likely to be a major source of satisfaction, we would consider to implement previous interventions centred in strengthening potency. Limitations and future directions. The present research is subject to certain limitations, which should be considered in future research. First, the sample; despite the relatively large number of individuals involved, the sample shrinks when the analysis is performed at the group level. It would also be of interest to use samples differentiated by service within the healthcare field, and from other sectors. It would also be interesting to fill out this correlative and cross-level study by carrying out experimental and longitudinal research to establish the direction of causality and explore the possible influence of team development over time. The findings also invite to explore the conditions under which change-oriented leadership, insofar as it represents a strategic issue, will have the greatest impact, analysing the environment, external relations and so on, as well as the moderating role of leadership between these variables and climate, as proposed by West and Hirst (2003). Similarly, the relationships between other important variables need further examination, as task and objectives interdependence, empowerment and team autonomy, etc. It would likewise be of interest to study their effects on new forms of work organisation and on virtual teams. To conclude, the importance of the leader’s role in anticipating change and providing the team with guidance in fluid situations is beyond doubt, but his/her real influence will depend on having the appropriate allies, on a favourable team climate

and on the existence of teams that are confident of their own potential. In this context, we may cite the metaphor that West (2002) applies to innovation: teams may be “sparkling fountains” instead of “stagnant ponds”.

References Alcover, C.M. and Gil, F. (1999), “The effects of member change and continuity on the productive efficiency of work teams”, Psychology in Spain, Vol. 3, pp. 88-97. Alcover, C.M. and Gil, F. (2000), “Potencia en grupos: un constructo entre la autoeficacia y la motivacio´n colectiva, (“Potency in groups: a construct between self-efficacy and collective motivation”), Apuntes de Psicologı´a, Vol. 18, pp. 123-43. Ancona, D. and Caldwell, D.F. (1992), “Bridging the boundary: external activity and performance in organizational teams”, Administrative Science Quarterly, Vol. 37, pp. 634-65. Anderson, N.R. and West, M.A. (1994), The Team Climate Inventory, Berks ASE Press, Windsor. Anderson, N.R. and West, M.A. (1998), “Measuring climate for work group innovation: development and validation of the team climate inventory”, Journal of Organizational Behavior, Vol. 19, pp. 235-58. Avolio, B.J., Jung, D.I., Murry, W.D. and Sivasubramaniam, N. (1996), “Building highly developed teams: focused on shared leadership process, efficacy, trust and performance”, in Beyerlein, M.M., Johnson, D.A. and Beyerlein, S.T. (Eds), Advances in Interdisciplinary Studies of Work Teams, JAI, Greenwich, CT, pp. 173-209. Baron, R.M. and Kenny, D.A. (1986), “The moderator-mediator distinction in social psychological research: conceptual, strategic, and statistical considerations”, Journal of Personality and Social Psychology, Vol. 51, pp. 1173-82. Bass, B.M. (1985), Leadership and Performance beyond Expectations, Free Press, New York, NY. Bass, B.M., Avolio, B.J., Jung, D.I. and Berson, Y. (2003), “Predicting unit performance by assessing transformational and transactional leadership”, Journal of Applied Psychology, Vol. 88, pp. 207-18. Burningham, C. and West, M.A. (1995), “Individual, climate, and group interaction processes as predictors of work team innovation”, Small Group Research, Vol. 26, pp. 106-17. Campion, M.A., Medsker, G.J. and Higgs, C.A. (1993), “Relations between group characteristics and effectiveness: implication for designing effective work groups”, Personnel Psychology, Vol. 46, pp. 823-50. Campion, M.A., Papper, E.A. and Medsker, G.J. (1996), “Relations between team characteristics and effectiveness: a replication and extension”, Personnel Psychology, Vol. 49, pp. 429-52. Carron, A.V. (1982), “Cohesiveness in sport groups: interpretations and considerations”, Journal of Sport Psychology, Vol. 4, pp. 123-38. Chan, D. (1998), “Functional relations among constructs in the same content domain at different levels of analysis: a typology of composition models”, Journal of Applied Psychology, Vol. 83, pp. 234-46. Conger, J.A. and Kanungo, R. (1988), Charismatic Leadership: The Elusive in Organizational Settings, Jossey-Bass, San Francisco, CA. Ekwall, G. and Arvonen, J. (1991), “Change-centered leadership: an extension of the two-dimensional model”, Scandinavian Journal of Management, Vol. 7, pp. 17-26. Foels, R., Driskell, J.E., Mullen, B. and Salas, E. (2000), “The effects of democratic leadership on group member satisfaction: an integration”, Small Group Research, Vol. 31, pp. 676-701.

Change-oriented leadership

325

JMP 20,3/4

326

Gil, F. and Alcover, C.M. (2002), “Work teams as complex social systems: temporal factors, sociocognitive processes and work team empowerment as basis of group dynamics at work”, 13th General Meeting of European Association of Experimental Social Psychology, San Sebastian, pp. 153-4. Gil, F., Ares, A. and Barrasa, A. (2003), “Direccio´n y liderazgo organizacional: gestio´n y cambio, (“Management and organizational leadership: management and change”), Encuentros en Psicologı´a Social, Vol. 1, pp. 95-100. Gladstein, D. (1984), “Groups in context: a model of task group effectiveness”, Administrative Science Quarterly, Vol. 29, pp. 499-517. Gonza´lez-Navarro, P., Bravo, M.J., Gonza´lez-Roma´, V., Zurriaga, R. and Peiro´, J.-M. (1993), “Liderazgo y percepciones de clima, (“Leadership and perceptions of climate”), in Munduate, L. and Baro´n, M. (Eds), Psicologı´a del Trabajo y de las Organizaciones, Eudema, Sevilla, pp. 75-85. Gonza´lez-Roma´, V., Ramos, J., Peiro´, J.-M., Rodrı´guez, I. and Mun˜oz, P. (1995), “Formacio´n del clima de los equipos de trabajo e interaccio´n social (“Formation of work team climate and social interaction”), in Zurriaga, R. and Sancerni, M.D. (Eds), Experiencias Laborales en Organizaciones de Trabajo, Nau Llibres, Valencia, pp. 65-78. Gully, S.M., Joshi, A., Incalcaterra, K.A. and Beaubien, J.M. (2002), “A meta-analysis of team-efficacy, potency and performance: interdependence and level of analysis as moderators of observed relationships”, Journal of Applied Psychology, Vol. 87, pp. 819-32. Guzzo, R.A. and Dickson, M.W. (1996), “Teams in organizations: recent research on performance and effectiveness”, Annual Review of Psychology, Vol. 47, pp. 307-38. Guzzo, R.A., Yost, P.R., Campbell, R.J. and Shea, G.P. (1993), “Potency in groups: articulating a construct”, British Journal of Social Psychology, Vol. 3, pp. 87-106. Heinemann, G.D. and Zeiss, A.M. (Eds) (2002), Team Performance in Health Care: Assessment and Development, Kluwer Academic/Plenum Publishers, New York, NY. House, R.J. (1977), “A 1976 theory of charismatic leadership”, in Hunt, J.G. and Larson, L.L. (Eds), Leadership: The Cutting Edge, Southern Illinois University Press, Carbondale, IL, pp. 189-207. Jordan, M.H., Feild, H.S. and Armenakis, A.A. (2002), “The relationship of group process variables and team performance: a team-level analysis in a field setting”, Small Group Research, Vol. 33 No. 1, pp. 121-50. Kahai, S.S. and Sosik, J.J. (1998), “Computer-supported work group potency and effectiveness: the role of transformational leadership, anonymity, and task interdependence”, Computers in Human Behaviour, Vol. 14 No. 3, pp. 491-511. Katzenbach, J.R. (1997), Teams at Top: Unleashing the Potential of Both Teams and Individual Leaders, Harvard Business School Press, Boston, MA. Kenny, D.A. and La Voie, L. (1985), “Separating individual and group effects”, Journal of Personality and Social Psychology, Vol. 48, pp. 339-48. Kotter, J.P. (1990), A Force for Change, Free Press, New York, NY. Kozlowski, S.W. and Doherty, M.L. (1989), “Integration of climate and leadership: examination of a neglected issue”, Journal of Applied Psychology, Vol. 74, pp. 546-53. Kozlowski, S.W., Gully, S.M., Salas, E. and Cannon-Bowers, J.A. (1996), “Team leadership and development: theories, principles, and guidelines for training leaders and teams”, in Beyerlein, M.M., Johnson, D.A. and Beyerlein, S.T. (Eds), Advances in Interdisciplinary Studies of Work Teams, JAI, Greenwich, CT, pp. 253-91.

Lester, S.W., Meglino, B.M. and Korsgaard, M.A. (2002), “The antecedents and consequences of group potency: a longitudinal investigation of newly formed work groups”, Academy of Management Journal, Vol. 45 No. 2, pp. 352-68.

Change-oriented leadership

Lewin, K., Lippitt, R. and White, R.K. (1939), “Patterns of aggressive behavior in experimentally created climates”, Journal of Social Psychology, Vol. 10, pp. 271-99. Litwin, G.H. and Stringer, R.A. (1968), Motivation and Organizational Climate, Harvard University Press, Boston, MA. Lowe, K.B., Kroeck, K.G. and Sivasubramaniam, N. (1996), “Effectiveness correlates of transformational and transactional leadership: a meta-analytic review of the MLQ literature”, Leadership Quarterly, Vol. 7, pp. 385-425. Man˜as, M.A., Gonzalez-Roma´, V. and Peiro´, J.-M. (1999), El Clima de los Equipos de Trabajo: Determinantes y Consecuencias (Work Team Climate: Determinants and Consequences), Servicio de Publicaciones de la Universidad, Almerı´a. Mohrman, S., Cohen, S. and Mohrman, A. (1995), Designing Team-based Organizations, Jossey-Bass, San Francisco, CA. Pearce, C.L., Gallagher, C.A. and Ensley, M.D. (2002), “Confidence at the group level of analysis: a longitudinal investigation of the relationship between potency and team effectiveness”, Journal of Occupational and Organizational Psychology, Vol. 75, pp. 115-19. Poole, M.S. and Real, K. (2003), “Groups and teams in health care: communication and effectiveness”, in Thompson, T.L., Dorsey, A.M., Miller, K. and Parrott, R. (Eds), Handbook of Health Communication, Lawrence Erlbaum, Mahwah, NJ, pp. 369-402. Salas, E., Stagl, K.C. and Burke, C.S. (2004), “25 years of team effectiveness in organizations: research themes and emerging needs”, in Cooper, C.L. and Robertson, I.T. (Eds), International Review of Industrial and Organizational Psychology, Vol. 19, John Wiley & Sons, Chichester, pp. 47-91. Shamir, B. (1999), “Leadership in boundaryless organizations: disposable or indispensable?”, European Journal of Work and Organizational Psychology, Vol. 8, pp. 49-71. Shamir, B., House, R.J. and Arthur, M.B. (1993), “The motivational effects of charismatic leadership: a self-concept-based theory”, Organizational Science, Vol. 4, pp. 577-94. Sivasubramaniam, N., Murry, W.D., Avolio, B.J. and Jung, D.I. (2002), “A longitudinal model of the effects of team leadership and group potency on group performance”, Group and Organization Management, Vol. 27, pp. 66-96. Sobel, M.E. (1982), “Asymptotic confidence intervals for indirect effects in structural equation models”, in Leinhart, S. (Ed.), Sociological Methodology, Jossey-Bass, San Francisco, CA, pp. 290-312. Sosik, J.J., Avolio, B.J. and Kahai, S.S. (1997), “Effects of leadership style and anonymity on group potency and effectiveness in a group decision support system”, Environment Journal of Applied Psychology, Vol. 82 No. 1, pp. 89-103. Sosik, J.J., Avolio, B.J., Kahai, S.S. and Jung, D.I. (1998), “Computer-supported work group potency and effectiveness: the role of transformational leadership, anonymity, and task interdependence”, Computers in Human Behavior, Vol. 14 No. 3, pp. 491-511. Spink, K.S. (1998), “Mediational effects of social cohesion on the leadership behavior intention to return relationship in sport”, Group Dynamics, Vol. 2, pp. 92-100. Stewart, G.L. and Barrick, M.R. (2000), “Team structure and performance: assessing the mediating role of intra-team process and the moderating role of task type”, Academy of Management Journal, Vol. 43, pp. 135-48.

327

JMP 20,3/4

328

Tushman, M.L. and Romanelli, E. (1985), “Organization evolution: a metamorphosis model of convergence and reorientation”, in Cummings, L.L. and Staw, B.M. (Eds), Research in Organizational Behavior, Vol. 7, JAI Press, Greenwich, CT, pp. 171-222. West, M.A. (2002), “Sparkling fountains or stagnant ponds? An integrative model of creativity and innovation implementation in work groups”, Applied Psychology: An International Review, Vol. 51, pp. 355-424. West, M.A. and Hirst, G. (2003), “Cooperation and teamwork for innovation”, in West, M.A., Tjosvold, D. and Smith, K.G. (Eds), International Handbook of Organizational Teamwork and Cooperative Working, John Wiley, Chichester, pp. 297-319. Yukl, G. (1998), “An evaluative essay on current conceptions of effective leadership”, European Journal of Work and Organizational Psychology, Vol. 8, pp. 33-48. Yukl, G. (2002), Leadership in Organizations, Prentice-Hall, Saddle River, NJ. Yukl, G. (2004), “Tridimensional leadership theory: a road-map for flexible, adaptive leaders”, in Burke, R.J. and Cooper, C. (Eds), Leading in Turbulent Times, Blackwell, Oxford, pp. 75-91. Yukl, G., Gordon, A. and Taber, T. (2002), “A hierarchical taxonomy of leadership behavior: integrating a half century of behavior research”, Journal of Leadership and Organizational Studies, Vol. 9, pp. 15-32.