Person-Organization Fit 9781846633638, 9781846633621

Citation preview

jmp cover (i).qxd

07/02/2007

10:44

Page 1

ISSN 0268-3946

Volume 22 Number 2 2007

Journal of

Managerial Psychology Person-organization fit Guest Editor: Michael Morley

www.emeraldinsight.com

Journal of Managerial Psychology

ISSN 0268-3946 Volume 22 Number 2 2007

Person-organization fit Guest Editor Michael Morley

Access this journal online ______________________________

107

Editorial advisory board ________________________________

108

Person-organization fit Michael J. Morley ______________________________________________

109

A contingency perspective of the importance of PJ fit and PO fit in employee selection Tomoki Sekiguchi ______________________________________________

118

Attracting for values: an empirical study of ASA’s attraction proposition Jon Billsberry__________________________________________________

132

Using assessment centre performance to predict subjective person-organisation (P-O) fit: a longitudinal study of graduates Thomas N. Garavan ____________________________________________

150

Gender enactment at work: the importance of gender and gender-related behavior to person-organizational fit and career decisions Angela M. Young and David Hurlic _______________________________

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 150 additional Emerald journals in Emerald Management Xtra (www.emeraldinsight.com) See page following contents for full details of what your access includes.

168

CONTENTS

CONTENTS

Work value fit and turnover intention: same-source or different-source fit

continued

Annelies E.M. van Vianen, Irene E. De Pater and Floor Van Dijk _______

188

When person-organization (mis)fit and (dis)satisfaction lead to turnover: the moderating role of perceived job mobility Anthony R. Wheeler, Vickie Coleman Gallagher, Robyn L. Brouer and Chris J. Sablynski ____________________________

203

Call for papers ___________________________________________

220

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

Structured abstracts Emerald structured abstracts provide consistent, clear and informative summaries of the content of the articles, allowing faster evaluation of papers.

How to access this journal electronically

Additional complimentary services available

To benefit from electronic access to this journal, please contact [email protected] A set of login details will then be provided to you. Should you wish to access via IP, please provide these details in your e-mail. 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

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

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.

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

Emerald online training services Visit www.emeraldinsight.com/training and take an Emerald online tour to help you get the most from your subscription.

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 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.

Xtra resources and collections When you register your journal subscription online, you will gain access to Xtra resources for Librarians, Faculty, Authors, Researchers, Deans and Managers. In addition you can access Emerald Collections, which include case studies, book reviews, guru interviews and literature reviews.

Emerald Research Connections An online meeting place for the research community where researchers present their own work and interests and seek other researchers for future projects. Register yourself or search our database of researchers at www.emeraldinsight.com/ connections

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 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

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 785201

JMP 22,2

108

Journal of Managerial Psychology Vol. 22 No. 2, 2007 p. 108 # Emerald Group Publishing Limited 0268-3946

EDITORIAL ADVISORY BOARD

EDITORIAL REVIEW BOARD

Professor Neil Anderson University of Amsterdam, The Netherlands Professor Chris Argyris Harvard Graduate School of Business Administration, USA Professor Yehuda Baruch Norwich Businesss School, University of East Anglia, UK Professor Frank Bournois Universite´ Panthe´on-Assas, Paris II, France Professor Cary L. Cooper Lancaster University Management School, Lancaster University, UK Professor Dov Eden Tel Aviv University, Israel Martin Euwema University of Utrecht, The Netherlands Professor Adrian Furnham University College London, UK Professor Hugh P. Gunz Joseph L. Rotman School of Management, University of Toronto, Ontario Dr Frank Heller Tavistock Institute, UK Professor Geert Hofstede Institute for Research on Intercultura Co-operation, The Netherlands Professor Paul Iles Teesside Business School, UK Professor Jim Jawahar Illinois State University, USA 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, Germany Professor Harold J. Leavitt Stanford University, USA Professor Manuel London State University of New York, NY, USA Professor Dr Wolfgang Mayrhofer Vienna University of Economics and Business Administration, Austria Professor Greg Northcraft College of Business, University of Illinois, USA Dr Francisco Gil Rodriguez Universidad Complutense de Madrid, Spain

Professor Yochanan Altman London Metropolitan University, UK Prof. dr. Arnold Bakker Erasmus University Rotterdam, The Netherlands Dr Dean Bartlett London Metropolitan University, UK Dr Gayle Baugh University of West Florida, USA Professor Ce´leste Brotheridge Universite´ Du Que´bec a` Montre´al, Canada Dr Adrian Carr University of Western Sydney, Australia Dr Stuart Carr Massey University, New Zealand Professor Kerry Carson University of Louisiana at Lafayette, USA Dr 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 Christian Kiewitz University of Dayton, USA Dr Ute-Christine Klehe Programmagroep A&O Psychologie, Amsterdam, The Netherlands Professor Steven D. Maurer Old Dominion University, USA Dr Ioannis Nikolaou Athens University of Economics and Business, Greece Dr Chris Rees University of Manchester, UK Dr Ramon Rico Universidad Autonoma de Madrid, Spain Professor Alain M. Roger IAE de Lyon, Universite´ Jean Moulin, Lyon Dr Raymond Saner Centre for Socio-Economic Development, Geneva, Switzerland Dr Rene´ Schalk Tilburg University, The Netherlands Dr Ruth Simpson Brunel University, UK Dr Sherry E. Sullivan Bowling Green State University, USA Dr Shay Tzafrir University of Haifa, Israel Dr Daniel Vloeberghs University of Antwerp, Belgium Dr Lichia Yiu Centre for Socio-Economic Development, Geneva, Switzerland

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

Person-organization fit

Personorganization fit

Michael J. Morley University of Limerick, Limerick, Ireland Abstract

109

Purpose – The purpose of the paper is to introduce the special issue that brings together six papers exploring aspects of person-organization fit. Design/methodology/approach – This overarching paper contextualizes the theme and introduces the selected papers. Findings – The findings in this paper vary according to the core theme of each of the six contributions. Originality/value – Combined the papers explore new avenues of enquiry in the person-organization (P-O) fit domain and showcase international theoretical and empirical work on the P-O fit construct. Keywords Organizations, Selection, Recruitment Paper type Research paper

Introduction Since its inception, the Journal of Managerial Psychology has endeavored to promote a dialogue between psychologists and management scholars and between theoreticians and practitioners. There are few areas in which the promotion and safeguarding of a dialogue between these groups could be more fitting than in the area of person-environment fit (P-E fit) theory. While its etymology lies in interactional psychology, the umbrella notion of P-E fit is now a central plank of enquiry in several allied fields in the social and behavioral sciences. It is an intuitively appealing concept because of what we know about the desirability of good fit in the key domain aspects of our lives and the positive psychosomatic consequences that can accumulate when individuals perceive good fit between these aspects and their environment. Applied specifically to the workplace domain, the P-E fit construct has been hugely influential and has spawned a long line of investigations dedicated to exploring P-E fit as an overarching construct, and more recently, a series of influential contributions dedicated to dimensionalizing aspects of the P-E architecture and unearthing factual and counterfactual evidence of the associated constructs (Edwards et al., 1998; Holland, 1997; Schneider, 2001; Caplan, 1987; Dawis, 1992; Judge and Ferris, 1992). As Kristof-Brown et al. (2002, p. 985) note, P-E fit “is a comprehensive notion that necessarily includes one’s compatibility with multiple systems in the work environment”. Apposite to this it also has an ongoing and persistent “elusive” quality (Judge and Ferris, 1992; Wheeler et al., 2005). Within the P-E fit construct, distinct building blocks are seen to concomitantly exist, namely person-job fit (P-J fit), person-vocation fit (P-V fit), person-person fit (P-P fit), person-group fit (P-G fit) and person-organization fit (P-O fit) (Caplan, 1987; O’Reilly et al., 1991; Cable and Judge, 1996; Kristof, 1996; Werbel and Gilliland, 1999; Kristof-Brown et al., 2002; Kristof-Brown et al., 2005).

Journal of Managerial Psychology Vol. 22 No. 2, 2007 pp. 109-117 q Emerald Group Publishing Limited 0268-3946 DOI 10.1108/02683940710726375

JMP 22,2

110

In this special issue of the Journal of Managerial Psychology, our interest lies predominantly with the latter construct, person-organization fit (P-O fit). While there is a danger in isolating P-O fit for independent treatment of allied fit constructs because assessing multiple dimensions of fit simultaneously provides a “more realistic account of their relative influence” (Carless, 2005, p. 412), and because multi-level research bridging the macro (organizational) and micro (individual) perspectives is more likely to enable researchers to make linkages between constructs that might appear unconnected (Pappas and Flaherty, 2006), nonetheless because of the state of development of both theory and empirics in the area, it is justified. The nature of person-organization fit (P-O fit) Rynes and Cable (2003) in their account of recruiting for the twenty-first century note that serious job applicants are likely to demonstrate as much concern about choosing the most appropriate organization for them to work for as much as the most appropriate job for them to perform. From an organizational perspective, while conventional selection processes were centrally concerned with work oriented analyses and the determination of sets of knowledge, skills and abilities required for in-role behavior, more recent research has sought to look beyond the job to identify extra-role behavior. In this way, the priority is seen to have shifted from conventional models which are primarily based on “KSA’s” for “jobs” to hiring for organizational compatibility as manifest through a fit between an individuals personality, beliefs and values and the organization’s espoused culture, norms and values. Schneider (2001) advances an attraction-selection-attrition (ASA) model suggesting that work values are a core means by which individuals judge their P-O fit and individuals are attracted to and seek employment with organizations that exhibit characteristics similar to their own and organizations in turn tend to select individuals who are most similar to the organization. Values are an important aspect of both individuals and organizations that can be compared “directly and meaningfully” (Cable and Judge, 1997, p. 547). Arthur et al. (2006) note that if P-O fit is going to be used for employment decision making, as increasingly appears to be the case, then measures of P-O fit must be held to the same psychometric and legal standards as are other selection tests. Kristof (1996) in her integrative review defines P-O fit as the “compatibility between people and organizations that occurs when at least one entity provides what the other needs or they share similar fundamental characteristics, or both”. Derived from the central premise that different types of individuals are attracted to different types of organizations, and located within the broader debate on person-environment fit, person-organization fit emphasizes the importance of fit between employees and work processes and the importance of creating an organizational identity through the institutionalization of consistent values that permeate an organization’s culture (Werbel and DeMarie, 2005). Value congruence has become “widely accepted as the defining operationalization of P-O fit” (Kristof-Brown et al., 2005, p. 285). Westerman and Vanka (2005) note that P-O fit is underpinned by the assumption that attitudes, behavior and other person level outcomes result not from the person or the work environment independent of each other, but rather from the relationship between the two. In this context, value congruence between the individual and the organization, achieved through the determining of good fit when making the employment decision is the central plank of this aspect of person-environment fit (Chatman, 1989; Ostroff et al.,

2005), and during the recruitment and selection process as potential employees explore and develop perceptions of fit with the organization in terms of a congruence between their value set and that of the recruiting organization, they will potentially select themselves out of the recruitment process if they perceive a misfit between their enduring value set and that of the hiring organization. Progress and pessimism in the P-O fit agenda Achieving a high degree of P-O fit is viewed in many quarters as desirable in terms of positive work-related outcomes, especially in the context of a tight labor market and the war for talent (Ng and Burke, 2005) and in buttressing organizational culture. Much has been be claimed for it in terms of its potential impact on inter alia, job seeking intentions, both job and career satisfaction, psychological strain, organizational citizenship behaviors, knowledge acquisition and knowledge sharing, ethical conduct, organizational identification, job performance, and turnover. Thus, for example, in their meta-analytic review of the relationship between P-O fit and behavioral outcomes, Hoffman and Woehr (2006) indicate that P-O fit is weakly to moderately related to job performance, organizational citizenship behavior and turnover. In their most recent meta-analysis Kristof-Brown et al. (2005) find that P-O fit has strong correlations with job satisfaction and organizational commitment and a more moderate correlation with intention to quit. The relationship between P-O fit and attitudinal dimensions including satisfaction with coworkers, satisfaction with supervisors and trust in management was moderate, while the correlation with organizational satisfaction was substantially higher. Specifically on the issue of performance Kristof-Brown et al. (2005) find that P-O fit has low correlations with overall job performance and task performance and moderate correlations with contextual performance. Despite significant advances in what has been characterized as a “robust” stream of literature (Pappas and Flaherty, 2006), P-O fit as a concept remains somewhat problematic and as Kristof-Brown et al. (2005, p. 282) note there are a number of challenges to this kind of analysis, most especially “the proliferation of conceptualizations, measures and analytic approaches” P-O fit rather than having a singular platform for investigation has been conceptualized in several different ways. Kristof-Brown et al. (2005) note the dominant focus on values and value congruence as the main approach to operationalizing P-O fit, but they also note that it is not the only one. In some situations, goal congruence is used. Bretz and Judge (1994) identify four distinct conceptualizations which characterize this literature: the degree to which individual knowledge, skills and abilities match core job requirements; the degree of congruence between individual needs and organizational structures; the match between an individual’s value set and the organizations culture and value set; and the individual’s personality and perceived organizational image. In their recent review Piasentin and Chapman (2006) suggest that there is an underlying confusion in the literature on P-O fit generated through an inconsistency in investigations on how fit is conceptualized and how it is subsequently assessed. In acknowledging the multiple domain aspects of the concept of person-organization fit, they highlight the necessity to develop and validate uniform measures of person-organization fit in order to overcome some of the confusion inherent in this literature. They identify four common definitions of P-O fit, namely:

Personorganization fit

111

JMP 22,2

112

(1) Supplementary fit where an individual possesses characteristics that are similar to existing organizational characteristics. (2) Complementary fit where an individual fills a void or adds something that is missing in the organization. (3) Needs-supplies fit where an individuals needs are fulfilled by the organization. (4) Demand abilities fit where an individual’s abilities meet the demands of the organization. Arising from this, Piasentin and Chapman (2006, p. 203) note that: Although, theoretically, these definitions should represent distinct ways of perceiving fit, the dimensions have yet to be precisely defined or empirically tested. This lack of attention to conceptual issues not only impedes our ability to adequately measure the construct, but it may also lead us to draw faulty conclusions about the antecedents and consequences of subjective P-O fit.

The relative emphasis in many studies is also problematic. Arthur et al. (2006) draw attention to the absence of a strong theoretical or conceptual basis for a direct relation between P-O fit and job performance. Their meta analysis revealed that, contrary to the increased popularity of P-O fit in the selection decision, the volume of literature investigating the criterion related validity of P-O fit as a predictor of job performance and particularly, turnover was limited when compared with attitudinal criteria: Results of our meta-analysis of criterion-related validity of P-O Fit suggest that P-O Fit is not a good predictor of job performance, although it may hold more promise as a predictor of turnover (Arthur et al., 2006, p. 797).

Carless (2005) notes that a problem that plagues many of the fit studies of job choice is the heavy reliance on college students as a source of research data. Referring to the extent to which we can assume that the perceived fit and attraction ratings given in an experimental situation reflect the values attached to these during an actual job search, she cites Breaugh (1992, p. 83) who notes that “such an assumption seems implausible”. In proposing a strategic contingency framework dedicated to aligning strategic human resource management and overall P-O fit, Werbel and DeMarie (2005) draw attention to the problem of organizational sub-cultures and to the manifest difficulty attaching to achieving person-organization fit where different units in an organization will likely have different cultures. In this context they suggest an important role for human resource management where consistency in selection, performance appraisal and compensation systems across organizational units is necessary if firms are to compete through culture-based competencies. In this issue on P-O fit The importance of exploring and testing person-organization fit concepts and measures in a greater variety of cultural settings and with more diverse groups within the labor market has been acknowledge as an important academic endeavor and one which now forms part of the landscape of this area of enquiry (Van Hoof et al., 2006; Erdogan and Bauer, 2005; Yaniv and Farkas, 2005; Parkes et al., 2001; Turban et al., 2001), but one too which poses many significant challenges because of disparity in culturally derived value systems and wide variation in the meaning and centrally of

work in the cross-cultural, and increasingly heterogeneous workforce context. This issue of the Journal of Managerial Psychology brings together six papers dedicated to further exploring aspects of the P-O fit agenda and showcases theoretical and empirical contributions from Japan, the UK, Ireland, The Netherlands and the US. In our first paper Tomoki Sekiguchi of the Graduate School of Economics in Osaka University in Japan presents a contingency perspective of the importance of P-J fit and P-O fit in employee selection. The paper explores the different types of employees and employment relationships that impact the relative importance of P-J fit and P-O fit in the hiring decision. Drawing on several literatures, the centrality of fit as a selection criterion is affirmed and propositions advancing the relative importance of P-J fit over P-O fit and vice versa, depending on whether the firm wishes to establish a transactional psychological contract or a relational psychological contract, are set down. The paper offers insight into the value of fit constructs, depending on the preferred human resource management architecture of the firm. Our second paper by Jon Billsberry of the Open University Business School in the UK reports an empirical test of Schneider’s attraction-selection-attrition (ASA) model. Noting the cross-level conceptualization of P-O fit common in the extant literature, and combining them with Schneider’s attraction proposition, and drawing upon data gathered among applicants for graduate entry managerial trainee posts in nine utility companies in the UK and from a comparison group seeking work through university career services, three core hypotheses are tested relating to whether individuals who apply to an organization are a better fit than a sample of individuals who were looking for similar work at the same time through the same process. Billsberry’s results fail to support Schneider’s attraction proposition with graduates appearing to make vocational choices over organizational ones. Controlling for P-V fit, the P-P, P-G and P-O fit of job applicants to the firms in the study were not significantly different from the comparison group seeking work through the university career services. Our third paper by Thomas Garavan of the Kemmy Business School in the University of Limerick in Ireland uses a panel design to examine the potential of assessment performance to predict high potential graduate P-O fit over time. Three core hypotheses relating to the assessment center performance and subjective P-O fit are advanced and tested using hierarchical regression analysis. The results indicate that assessment center performance has an important role to play in the selection of graduates providing incremental variance over and above that provided by other measures employed. Our fourth paper by Angela Young of California State University in Los Angeles and David Hurlic of Pepperdine University in Los Angles explores the importance of gender and gender related behavior to P-G and P-O fit and career decisions. Drawing upon multiple literatures, a model is advanced in order to provide an exposition of gender enactment and fit in relation to career decisions. Gender enactment is viewed as the “composite of an individual’s awareness that situations demand gender-related behavior adjustments and an individual’s orientation toward masculine or feminine behaviors. A series of P-G fit and P-O fit propositions and their impact on career decisions are advanced from the enfolding literatures and offered as potential avenues of enquiry in order that we might better understand the nuances of gender-related behavior and its likely impact on interactions and perceptions in both the workgroup and organization context.

Personorganization fit

113

JMP 22,2

114

Our fifth paper by Annelies van Vianen, Irene De Pater and Floor Van Dijk of the University of Amsterdam in The Netherlands explores same source/different source fit, investigating the linkage between fit regarding work values and turnover intentions. The authors link employees work values to their own perceptions of the work values of the organization, the aggregated perceptions of team members and the aggregated perceptions of a larger group of organizational members and advance and test three hypotheses using data from 105 employees grouped into sixteen teams in a liquor trading company in The Netherlands Antilles. The results suggest that personal work values were not significantly related to turnover intention. Neither was the same-source fit measure related to turnover intention. However, the different source fit measure did reach significant. Both the scientific and practical implications of the work are discussed. Our final paper in this special issue by Anthony Wheeler of Bradley University, Vickie Coleman Gallagher and Robyn Brouer both of Florida State University and Chris Sablynski of California State University examines the relationships between P-O fit, job satisfaction, perceived job mobility and intent to turnover. Drawing upon Lee and Mitchell’s unfolding model of voluntary turnover and Wheeler et al.’s theory of multidimensional fit, four interlinked hypotheses are advanced and tested using data from a web based survey. The analysis points to a statistically significant relationship between P-O fit and job satisfaction, and as job satisfaction increases, respondents intention to turnover decreases. Perceived job mobility is found to moderate the relationship between job satisfaction and intent to turnover. Thus, decreases in P-O fit, which leads to decreases in job satisfaction, are more likely to result in increases in intent to turnover if the individual also perceives alternative job opportunities. Special issue reviewer acknowledgements This special issue was made possible by the support of a large number of collaborators, most especially a cadre of committed reviewers who willingly assisted in reviewing manuscripts submitted for consideration. I place on record my thanks to the following reviewers who gave of their time and expertise: . Jon Billsberry, Open University; . Jeanette Cleveland, Pennsylvania State University; . Christine Cross, University of Limerick; . Ali Dastmalchian, University of Victoria; . Brian Dineen, University of Kentucky; . Joan Finegan, University of Western Ontario; . Sandra Fisher, Clarkson University; . Thomas Garavan, University of Limerick; . Noreen Heraty, University of Limerick; . Margaret Linehan, Cork Institute of Technology; . Brian Lyons, University of Albany, State University of New York; . Wenzel Matiaske, University of Flensburg; . Cameron Newton, Queensland University of Technology;

. . . . . . . . .

Emma Parry, Cranfield University; Judy Pate, University of Glasgow; Elizabeth Ravlin, University of South Carolina; Robert Renn, University of Memphis; Lynn Shore, San Diego State University; Aharon Tziner, Netanya Academic College; Annelies van Vianen, University of Amsterdam; Anthony Wheeler, Bradley University; and Jonathan Ziegert, Drexel University.

A special word of thanks to the Editor of the Journal of Managerial Psychology, James Werbel for his encouragement in translating the first ideas for the Special Issue into a coherent call for papers and to Kay Sutcliffe, the journal Editorial Administrator, for ongoing support. References Arthur, W., Bell, S.T., Villado, A.J. and Doverspike, D. (2006), “The use of person-organization fit in employment decision making: an assessment of its criterion-related validity”, Journal of Applied Psychology, Vol. 91 No. 4, pp. 786-801. Breaugh, J.A. (1992), Recruitment: Science and Practice, PWS Kent, Boston, MA. Bretz, R.D. and Judge, T.A. (1994), “Person-organization fit and the theory of work adjustment: implications for satisfaction, tenure and career success”, Journal of Vocational Behavior, Vol. 44, pp. 43-54. Cable, D.M. and Judge, T.A. (1996), “Person-organization fit, job choice decisions, and organizational entry”, Organizational Behavior and Human Decision Processes, Vol. 67, pp. 294-311. Cable, D.M. and Judge, T.A. (1997), “Interviewers’ perceptions of person-organization fit and organizational selection decisions”, Journal of Applied Psychology, Vol. 82 No. 4, pp. 546-61. Caplan, R.D. (1987), “Person-environment fit theory and organizations: commensurate dimensions, time perspectives, and mechanisms”, Journal of Vocational Behavior, Vol. 31, pp. 248-67. Carless, S. (2005), “Person-job fit versus person-organization ft as predictors of organizational attraction and job acceptance intentions: a longitudinal study”, Journal of Occupational and Organizational Psychology, Vol. 78, pp. 411-29. Chatman, J.A. (1989), “Improving interactional organizational research: a model of person-organization fit”, Academy of Management Review, Vol. 14, pp. 333-49. Dawis, R. (1992), “Person-environment fit and job satisfaction”, in Cranny, C., Smith, P. and Stone, E. (Eds), Job Satisfaction: How People Feel about Their Jobs and How it Affects Their Performance, Lexington, New York, NY. Edwards, J., Caplan, R. and Harrison, R. (1998), “Person-environment fit theory: conceptual foundations, empirical evidence, and directions for future research”, in Cooper, C. (Ed.), Theories of Organizational Stress, Oxford University Press, Oxford. Erdogan, B. and Bauer, T. (2005), “Enhancing career benefits of employee proactive personality: the role of fit with jobs and organizations”, Personnel Psychology, Vol. 58 No. 4, pp. 859-91.

Personorganization fit

115

JMP 22,2

116

Hoffman, B.J. and Woehr, D.J. (2006), “A quantitative review of the relationship between person-organization fit and behavioral outcomes”, Journal of Vocational Behavior, Vol. 68 No. 3, pp. 389-99. Holland, J.L. (1997), Making Vocational Choices: A Theory of Vocational Personalities and Work Environments, Psychological Assessment Resources, Odessa, FL. Judge, T. and Ferris, G. (1992), “The elusive criterion of fit in human resource staffing decisions”, Human Resource Planning, Vol. 15 No. 4, pp. 47-66. Kristof, A.L. (1996), “Person-organization fit: an integrative review of its conceptualizations, measurement, and implications”, Personnel Psychology, Vol. 49, pp. 1-49. Kristof-Brown, A.L., Jansen, K.J. and Colbert, A.E. (2002), “A policy-capturing study of the simultaneous effects of fit with jobs, groups, and organizations”, Journal of Applied Psychology, Vol. 87 No. 5, pp. 985-93. Kristof-Brown, A.L., Zimmerman, R.D. and Johnson, E.C. (2005), “Consequences of individuals’ fit at work: a meta-analysis of person-job, person-organization, person-group, and person-supervisor fit”, Personnel Psychology, Vol. 58, pp. 281-342. Ng, E.S.W. and Burke, R.J. (2005), “Person-organization fit and the war for talent: does diversity management make a difference?”, International Journal of Human Resource Management, Vol. 16 No. 7, pp. 1195-210. O’Reilly, C.A., Chatman, J. and Caldwell, D.F. (1991), “People and organizational culture: a profile comparison approach to assessing person-organization fit”, Academy of Management Journal, Vol. 34, pp. 487-516. Ostroff, C., Shin, Y. and Kinicki, A.J. (2005), “Multiple perspectives of congruence: relationships between value congruence and employee attitudes”, Journal of Organizational Behavior, Vol. 26 No. 6, pp. 591-623. Pappas, J.M. and Flaherty, K.E. (2006), “The moderating role of individual-difference variables in compensation research”, Journal of Managerial Psychology, Vol. 21 No. 1, pp. 19-35. Parkes, L., Bochner, S. and Schneider, S. (2001), “Person-organization fit across cultures: an empirical investigation of individualism and collectivism”, Applied Psychology: An International Review, Vol. 50 No. 1, pp. 81-108. Piasentin, K.A. and Chapman, D.S. (2006), “Subjective person-organization fit: bridging the gap between conceptualization and measurement”, Journal of Vocational Behavior, Vol. 69 No. 2, pp. 202-21. Rynes, S.L. and Cable, D.M. (2003), “Recruiting research in the 21st century: moving to a higher level”, in Borman, W., Ilgen, D. and Klimoski, R. (Eds), The Handbook of Psychology, Vol. 12, John Wiley, New York, NY, pp. 55-77. Schneider, B. (2001), “Fits about fit”, Applied Psychology: An International Review, Vol. 50 No. 1, pp. 141-52. Turban, D.B., Lau, C.M., Ngo, H.Y., Chow, I.H.S. and Si, S.X. (2001), “Organizational attractiveness of firms in the People’s Republic of China: a person-organization fit perspective”, Journal of Applied Psychology, Vol. 86 No. 2, pp. 194-206. Van Hoof, E.A.J., Born, M., Taris, T.W. and Van der Flier, H. (2006), “Ethnic and gender differences in applicants’ decision-making processes: an application of the theory of reasoned action”, International Journal of Selection and Assessment, Vol. 14 No. 2, pp. 156-66. Werbel, J.D. and DeMarie, S.M. (2005), “Aligning strategic human resource management and person-environment fit”, Human Resource Management Review, Vol. 15 No. 2, pp. 247-62.

Werbel, J.D. and Gilliland, S.W. (1999), “Person-environment fit in the selection process”, Research in Personnel and Human Resource Management, Vol. 17, pp. 209-43. Westerman, J.W. and Vanka, S. (2005), “A cross-cultural empirical analysis of person-organization fit measures as predictors of student performance in business education: comparing students in the Unites States and India”, Academy of Management Learning and Education, Vol. 4 No. 4, pp. 409-20. Wheeler, A.R., Buckley, M.R., Halbesleben, J.R., Brouer, R.L. and Ferris, G.R. (2005), “The elusive criterion of fit revisited: toward an integrative theory of multidimensional fit”, in Martocchio, J. (Ed.), Research in Personnel and Human Resource Management, JAI Press, Greenwich, CT, pp. 265-304. Yaniv, E. and Farkas, F. (2005), “The impact of person-organization fit on the corporate brand perception of employees and of customers”, Journal of Change Management, Vol. 5 No. 4, pp. 447-61. Further reading Dineen, B.R., Ash, S.R. and Noe, R.A. (2002), “A web of applicant attraction: person-organization fit in the context of web-based recruitment”, Journal of Applied Psychology, Vol. 87 No. 4, pp. 723-34. Lauver, K.J. and Kristof-Brown, A.L. (2001), “Distinguishing between employees’ perceptions of person-job and person-organization fit”, Journal of Vocational Behavior, Vol. 59 No. 3, pp. 454-70. About the author Michael J. Morley is Head of the Department of Personnel and Employment Relations, Kemmy Business School, University of Limerick, Ireland, and Director of the Master of Science in Work and Organizational Psychology. He is Consulting Editor of the Journal of Managerial Psychology and a member of the Editorial Board of several other international journals. His current research interests include international human resource management and expatriate transfers, intercultural transitional adjustment and convergence and divergence in European HRM. Michael J. Morley can be contacted at: [email protected]

To purchase reprints of this article please e-mail: [email protected] Or visit our web site for further details: www.emeraldinsight.com/reprints

Personorganization fit

117

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

JMP 22,2

A contingency perspective of the importance of PJ fit and PO fit in employee selection

118

Tomoki Sekiguchi Graduate School of Economics, Osaka University, Toyonaka, Osaka, Japan Abstract Purpose – The purpose of this paper is to propose a contingency perspective that describes the relative importance of person-job (PJ) fit and person-organization (PO) fit as selection criteria for hiring various types of employees. Design/methodology/approach – Drawing on the theories of psychological contracts, human capital and cosmopolitan-local perspective, propositions are developed regarding the relative importance of PJ fit and PO fit in specific hiring situations. Findings – The propositions developed in this paper suggest that PJ fit will be more important than PO fit when organizations hire employees to form transactional psychological contracts, to obtain general human capital, and/or who are categorized as cosmopolitans. On the other hand, it is suggested that PO fit will be more important than PJ fit when organizations hire employees to form relational psychological contracts, to develop firm-specific human capital, and/or who are likely to become locals. Research limitations/implications – Further empirical and theoretical work should be conducted to elaborate the contingency perspective. A more comprehensive contingency theory of person-environment (PE) fit could include other types of PE fit, other contingency variables, and cover a wider range of management practices. Practical implications – Organizations should be aware of the trade-off between PJ fit and PO fit in the selection process, and carefully examine what type of employees they will hire in order to determine the relative weights of PJ fit and PO fit as selection criteria. Originality/value – This paper is the first attempt to consider the different types of employees and employment relationships that determine the relative importance of PJ fit and PO fit in selecting employees. It is done through incorporating different theoretical perspectives. Keywords Recruitment, Selection, Contingency planning Paper type Conceptual paper

Introduction The concept of fit in employee selection has received growing attention in recent years. The overarching concept of fit in this field stems from person-environment (PE) congruence or PE fit in the interactionalist theory of behavior (e.g. Lewin, 1951). Among various forms of PE fit, employee selection researchers have extensively studied person-job (PJ) fit and person-organization (PO) fit (Adkins et al., 1994; Cable Journal of Managerial Psychology Vol. 22 No. 2, 2007 pp. 118-131 q Emerald Group Publishing Limited 0268-3946 DOI 10.1108/02683940710726384

An early version of this manuscript was presented at the Annual Meeting of the Academy of Management, Seattle, WA, August 2003. The author would like to thank Vandra Huber, Xiao-Ping Chen and Chad Higgins for their helpful comments on the earlier version of this manuscript. The author would also like to thank anonymous reviewers for their insightful feedback in the review process.

and Judge, 1997; Edwards, 1991; Kristof, 1996; Kristof-Brown, 2000; Werbel and Gilliland, 1999). PJ fit is defined as the compatibility between a person’s characteristics and those of the job or tasks that are performed at work. Edwards (1991) mentions that PJ fit has both a demands-supplies relationship (i.e. demands of the job and the abilities of the person) and a needs-supplies relationship (i.e. needs of the person and supplies from the job). PO fit is defined as the compatibility between people and entire organizations. Kristof (1996) summarizes several conceptualizations of PO fit, including supplementary PO fit (i.e. fit as similarity) and complementary PO fit (i.e. fit between demands and supplies). Cable and DeRue (2002) propose a three-dimensional model of PO value congruence, PJ needs-supplies fit and PJ demands-abilities fit. In the employee selection literature, the demands-abilities dimension of PJ fit and the supplementary dimension of PO fit are most frequently used (Adkins et al., 1994; Kristof-Brown, 2000; Werbel and Gilliland, 1999), and this paper also focuses on these two types of fit. Traditional research on employee selection has focused on PJ fit or the match between individual knowledge, skills and abilities (KSAs) and the requirements of the job as the major selection criteria; however, a growing number of practitioners and researchers advocate that this is not enough (Bowen et al., 1991; Kristof, 1996; Montgomery, 1996; Werbel and Gilliland, 1999). For example, Bowen et al. (1991) argue that organizations with self-motivated and committed employees tend to select employees based on PO fit as well as PJ fit. Werbel and Gilliland (1999) propose that employee selection should include PO fit and person-group (PG) fit as well as PJ fit. These researchers also propose ways to assess PO fit in selecting employees, such as conducting organizational analyses and assessing applicants’ personality, values and needs (Bowen et al., 1991; Werbel and Gilliland, 1999). Other researchers suggest that PO fit has already been included in actual employee selection practices such as in the selection interviews (Chatman, 1989; Ferris and Judge, 1991; Judge and Ferris, 1992; Rynes and Gerhart, 1990). Recent meta-analyses conducted by several researchers found that both PJ fit and PO fit are related to many positive individual outcomes, indicating that both PJ fit and PO fit should be considered in selecting employees (Hoffman and Woehr, 2006; Kristof-Brown et al., 2005; Verquer et al., 2003). For example, Kristof-Brown et al. (2005) report that PJ fit has strong correlations with job satisfaction (0.56), organizational commitment (0.47) and intent to quit (2 0.46), and moderate correlations with overall performance (0.20). Similarly, they report that PO fit has strong correlations with job satisfaction (0.44) and organizational commitment (0.51), and moderate correlations with intent to quit (2 0.35) and contextual performance (0.27). While research to date indicates that both PJ fit and PO fit are important as selection criteria, Werbel and Gilliland (1999) further suggest that the relative importance of fit is contingent on the work context. However, there has been little research into such a work context. That is, the PE fit literature has not provided theoretical guidance as to the relative importance of different types of fit when organizations hire new employees. In today’s world, hiring and managing a diverse workforce appear to be the keys to an organization gaining competitive advantage. For example, there has been a gradual movement away from full-time and ongoing employment arrangements towards an increased use of contingent workers (Kalleberg, 2000; McLean Parks et al., 1998).

PJ fit and PO fit

119

JMP 22,2

120

Another trend is that the number of professionals, such as lawyers, doctors and engineers, working in organizations has increased dramatically (Barley, 1996). Therefore, it is now common for many organizations to hire different types of employees, such as contingent employees and professional employees, as well as traditional full-time/regular employees. In such a situation, organizations should use different weights of PJ fit and PO fit as selection criteria for different types of employees, because the effect of each type of fit on various employee and organizational outcomes may differ according to the type of employees or employment relationships. The purpose of this paper is to provide a contingency perspective on the relative importance of PJ fit and PO fit in the context of employee selection by focusing on the differences in the types of employees and employment relationships. Specifically, this paper focuses on the characteristics of different types of employees and employment relationships that determine the relative importance of PJ fit and PO fit in selection. In developing theoretical arguments and propositions, this paper uses the theories of psychological contracts (Rousseau, 1995), human capital (Becker, 1964) and cosmopolitan-local perspective (Gouldner, 1957), which are critical when discussing employment relationships between organizations and employees. Note that the contingency perspective presented in this paper does not suggest that either PJ fit or PO fit is unimportant when the other type of fit is important. Rather, what the contingency perspective focuses on is the “relative importance.” That is, although both PJ fit and PO fit would be important in selecting employees (e.g. Kristof-Brown et al., 2005), one type of fit could be relatively more important than the other. Psychological contracts Psychological contracts are defined as an individual belief in mutual obligations between a person and another party such as an employer (Rousseau, 1995). Researchers commonly differentiate between transactional and relational psychological contracts (e.g. Hulin and Glomb, 1999; McLean Parks et al., 1998; Rousseau, 1989). In transactional psychological contracts, job requirements and expectations are likely to be clear and specified in advance, which allows individuals to assess personal costs and benefits associated with the exchange and calibrate their contributions accordingly. In relational psychological contracts, the details of the exchange are unlikely to be specified in advance, and the monitoring of inducements and contributions is less relevant. Instead, relational trust leads individuals into social exchange relationships (Rousseau et al., 1998). An example of the transactional psychological contracts is the employment relationship with contingent employees. Typical psychological contracts between contingent workers and organizations are, “You tell me specifically what tasks you want done in what period and I will do them. Don’t expect me to go beyond those tasks. I will not expect you to go beyond what is spelled out in the contract.” (Hulin and Glomb, 1999). Therefore, contingent workers are likely to carry out their assigned work and tasks with few modifications. On the other hand, relational psychological contracts are likely to occur in long-term employment relationships such as a life-long employment practice. The relational psychological contracts are likely to be more complex in terms of both what is expected from employees and what is expected from organizations, and are based more on social exchange and reciprocity in employment

relationships. Organizations expect commitment beyond the simple execution of assigned tasks, including citizenship duties that are necessary for organizations to function smoothly but that cannot be reasonably codified and standardized (Organ, 1988). Individual employees may expect organizations to “owe” them for their loyalty and commitment to nonjob tasks. Such debts may be paid in a number of non-financial ways, such as advanced training courses or informal flextime arrangements (Hulin and Glomb, 1999). The theory of psychological contracts suggests that the importance of PJ fit and PO fit in selecting employees may differ according to the type of psychological contracts that organizations want to develop with specific types of employees. When organizations hire employees to form transactional psychological contracts, the jobs that employees will be assigned are likely to be clearly defined with the detailed specifications of KSAs to do the job. Therefore, PJ fit will play the major role in selecting such employees. Transactional psychological contracts do not assume a deep social exchange relationship that requires citizenship behaviors that go beyond the simple execution of assigned tasks. Because PO fit plays an important role when organizations require employees to conduct citizenship behaviors aligned with organizational goals and values, PO fit may be relatively less important in selecting employees with transactional psychological contracts. P1a. When organizations hire employees with whom transactional psychological contracts are to be formed, PJ fit will be more important than PO fit. When organizations hire employees to form relational psychological contracts, the details of the exchange are not likely to be specified in advance, which suggests that job requirements are kept broad and flexible without the detailed specifications of KSAs. Employees with relational psychological contracts may be expected to adapt to changes in their tasks and to obtain new KSAs if they are necessary during their long-term employment relationship. Therefore, a PJ fit approach that is specific to the employee’s immediate job may be relatively less important at the time of hiring. On the other hand, the PO fit will be more important when selecting such employees. As discussed previously, PO fit plays an important role when organizations require employees to conduct citizenship behaviors so that organizations function smoothly. Empirical research has demonstrated that a high level of PO fit is positively related to citizenship behaviors (e.g. Goodman and Svyantek, 1999; O’Reilly and Chatman, 1986). According to meta-analysis, the correlation between PO fit and contextual performance (a similar construct to citizenship behaviors) was 0.27 (Kristof-Brown et al., 2005). In general, a high level of PO fit is supposed to contribute to a good long-term relationship between employees and the organizations because the congruence and similarity of goals and values increase the mutual understanding and trust between the two parties. P1b. When organizations hire employees with whom relational psychological contracts are to be formed, PO fit will be more important than PJ fit. Human capital theory According to the human capital theory, there are two main types of human capital: firm-specific human capital and general human capital (e.g. Becker, 1964). Firm-specific human capital is associated with KSAs that can be applied at a

PJ fit and PO fit

121

JMP 22,2

122

specific firm only and are not easily transferred to outside the firm. General human capital is associated with KSAs that are applicable to a broad range of jobs across organizations. This typology is relevant to the organizations’ “make or buy” decisions about their human capital (Lepak and Snell, 1999). The more firm specific the human capital is, the less likely it is to be available on the external market. In this case, organizations will “make” their own human capital by developing it internally thorough company sponsored training. On the other hand, the more general the human capital is, the easier it is to obtain from the external market. In this case, organizations will “buy” their human capital through employment externalization (e.g. by using a contingent workforce). The human capital theory suggests that firm-specific human capital is likely to become valuable, rare, inimitable and non-substitutable assets for the firm, which would be the source of sustainable competitive advantage (Barney, 1991; Becker, 1976). This attracts job applicants who want to remain in the same organization for a long time and obtain benefits by acquiring firm-specific human capital. For general human capital, employees would pay for the skills training and enjoy most of the benefits from their own training investment. That is, individuals who want to obtain general human capital and sell it to the organizations would do so at their own cost (Lazear, 1998). Employees who are expected to acquire firm-specific human capital will not need to have specific KSAs at the time of organizational entry. Because such KSAs are firm specific, job applicants have little chance of obtaining them in advance. They will obtain the firm-specific KSAs after they are hired through firm-specific skills training. Therefore, the PJ fit approach that focuses on the applicant’s current KSAs may be less important in selecting such employees. On the other hand, PO fit may be more important when hiring employees for firm-specific human capital. To maximize their return on investment in extensive training and development, organizations need to motivate employees to obtain and utilize the firm-specific KSAs, and to retain these employees for a long time. Employees’ commitment and loyalty to the organization and their understanding of the organization’s goals, values and strategy are critical for maintaining their motivation so that they obtain firm-specific KSAs and use those KSAs appropriately. Empirical evidence suggests that a high level of PO fit increases employees’ organizational commitment (Bretz and Judge, 1994; Chatman, 1991; O’Reilly et al., 1991). A high level of PO fit was also found to reduce employees’ intent to quit and actual turnover (Chatman, 1991; O’Reilly et al., 1991; Vancouver et al., 1994). Meta-analysis has shown that PO fit has a strong relationship with organizational commitment (0.51) and a moderate correlation with intent to quit (2 0.35) (Kristof-Brown et al., 2005). Although employees who have obtained firm-specific KSAs are less likely to leave because such KSAs are not easily applicable outside of the company, PO fit will have an additional effect on retaining employees (Mitchell et al., 2001). These discussions suggest that the importance of PO fit will be greater than that of PJ fit in selecting employees for firm-specific human capital. P2a. When organizations hire employees for firm-specific human capital, the PO fit will be more important than PJ fit. Contrary to the arguments on firm-specific human capital, organizations may opt to “buy” human capital from the external market if it is more general. Some types of

general human capital are so easy to purchase from the market that they can be treated essentially as commodities (Lepak and Snell, 1999). In this case, job applicants usually obtain the necessary KSAs in advance through general training and experience such as schooling. Because a wide-range of general human capital may be available on the external labor market, organizations are able to find people who have specific KSAs that would match the task demands. Furthermore, it is relatively easy to calculate the market value of such specific KSAs, which enables organizations to obtain them at fair prices. Therefore, when selecting employees for general human capital, it will be effective to focus on PJ fit. On the other hand, PO fit may be relatively less important in selecting such employees. Because general human capital is easily applicable in other organizations, employees with such KSAs are more likely to leave organizations than are those with firm-specific human capital, because they can “sell” their KSAs relatively easily in the external market (Lepak and Snell, 1999). From the employers’ point-of-view, organizations can use a “just-in-time” employment relationship for general human capital through the flexibility of hiring and firing (Tsui et al., 1995). Therefore, organizations may be able to manage such employees without them having strong organizational attachment (e.g. organizational commitment and retention), to which PO fit is theoretically relevant. Thus, the importance of PJ fit will be stronger than that of PO fit in hiring employees for general human capital. P2b. When organizations hire employees for general human capital, PJ fit will be more important than PO fit. Cosmopolitan-local perspective The cosmopolitan-local perspective was originally proposed by Gouldner (1957), and this perspective is closely related to the professional-bureaucratic conflict (Hall, 1968) and the professional-manager dichotomy (Golden et al., 2000; Kerr et al., 1977; Wallace, 1995). Gouldner (1957) made the distinction between “cosmopolitans”, who represent the interest of the occupation and have little loyalty to the organization, and “locals”, who are expected to represent the often-conflicting interests of the organization. Professionals, such as lawyers and engineers, employed by organizations are often seen as cosmopolitans, whereas executives and managers of organizations are usually considered locals (Golden et al., 2000). In a similar vein, the professional-bureaucratic conflict model suggests that there is an inherent tension between professional and bureaucratic goals and values. The professional value system is believed to emphasize values such as professional autonomy, conformity to professional standards and ethics, collegial authority and client orientation and loyalty. In contrast, the bureaucratic value system is said to emphasize hierarchical authority and control, conformity to organizational norms and regulations, and organizational loyalty (Hall, 1968; Wallace, 1993). Because the interests and values of cosmopolitans and locals often differ significantly, organizations that hire both groups of employees may have to use different employment practices to manage them effectively. This perspective suggests that the distinction between cosmopolitans and locals will be related to the relative importance of PJ fit and PO fit in selection. Cosmopolitans, like many salaried professionals, usually share values such as occupational autonomy, conformity to occupational standards and ethics, collegial authority and client orientation and loyalty. They tend to devote years of study to learn the technical idiosyncrasies of their discipline (Raelin, 1994). Therefore, their KSAs are

PJ fit and PO fit

123

JMP 22,2

124

often highly standardized within occupational societies. Hiring such employees may require the descriptions of necessary KSAs that are consistent with their occupational standards and norms. Organizations may also need to provide such employees with collegiality, autonomy and discretion after their organizational entry in order to satisfy their needs (Wallace, 1995). PJ fit will play the major role in selecting such employees because existing professional or occupational groups usually evolved around specific KSAs, and a focus on PJ fit is associated with the assessment of the applicant’s professional qualifications. On the other hand, PO fit may be relatively less important when selecting such employees. Cosmopolitans tend to seek fit with their occupation rather than with their employers. Therefore, for cosmopolitans, a high level of person-vocation (PV) fit (i.e. the match between individual and vocational characteristics) may be more important for employee outcomes than would PO fit (Kristof, 1996). Even when the level of PO fit is not high, organizations can structure their jobs so that they can maximize their professional value systems (e.g. maintain their professional autonomy by creating an independent department), which would maintain and/or enhance PV fit and positively affect their satisfaction and performance. Thus, the importance of PJ fit will be greater than that of PO fit in selecting cosmopolitans. P3a. When organizations hire employees who are categorized as cosmopolitans, PJ fit will be more important than PO fit. Corporate executives and managers are typical examples of locals as opposed to cosmopolitans. Contrary to the cosmopolitans, locals are generally more dependent on and accountable to their employing organization than to their occupations (Zajac et al., 1991). Tasks of locals are often deeply embedded in the organizational context. For example, the essential task of the corporate executive is to devise strategies and formulate policies to ensure that the organization’s specific goals and objectives are met (US Department of Labor, Bureau of Labor Statistics, 2002). The manager’s major function is to plan, organize, command, coordinate and control in order to attain organizational goals (Fayol, 1916). Their interpersonal, informational and decisional roles, the major set of roles determined by Mintzberg (1973), may be deeply embedded in the organizational context. Managers should also resolve conflicts with cosmopolitans in order to protect the interest of the organization. A high level of PO fit in terms of value and goal congruence is essential for such tasks and roles. Considering the original meaning of the cosmopolitan-local construct, non-managers can also be locals if they identify strongly with their employing organizations rather than with their occupation. In general, behaviors that are consistent with organizational goals, values and strategies are especially critical for locals to be effective. Thus, PO fit will be important in selecting employees who are locals. On the other hand, PJ fit as a selection criterion appears to be narrow, considering the tasks and roles of employees who are locals in organizations. Because their tasks and roles are deeply embedded in the organizational context, they need to obtain knowledge of the organizational context to perform their tasks and roles effectively. In this sense, the essential nature of the locals’ job tends to be broad, which suggests that the specific PJ fit approach is less meaningful. This argument might be especially true when organizations need to be adaptive and responsive to rapid technological, economic and social changes (Carson and Stewart, 1996). These

discussions suggest that the relative importance of PO fit in selecting locals is greater than that of PJ fit.

PJ fit and PO fit

P3b. When organizations hire employees who are likely to become locals, PO fit will be more important than PJ fit. Discussion The major contribution of this paper is that it provides the theoretical basis for a contingency perspective regarding the relative importance of PJ fit and PO fit in selecting employees. Little research to date has explicitly theorized the role of PE fit in employee selection from a contingency perspective. This paper is the first attempt to consider the different types of employees and the employment relationships that determine the relative importance of PJ fit and PO fit in selecting employees. In the following subsections, the implications of the contingency perspective developed in this paper are discussed from both theoretical and practical standpoints, along with a discussion of future research directions. Contribution to the PE fit theory The contingency perspective presented in this paper was developed by applying theoretical concepts from different fields of management. The theory of psychological contracts primarily focuses on the perceived psychological relationship between two parties. It is closely related to trust and mutual understanding between individuals and organizations. Human capital theory was originally developed in economics. It is occasionally applied at the macro level of human resource management research (e.g. strategic human resource management) to analyze competitive advantage through people (e.g. Lepak and Snell, 1999). Cosmopolitan-local perspective has been discussed traditionally in the sociology of professions. It also illustrates the inherent conflict between cosmopolitans and locals within the organization, which needs to be managed. The use of these different theoretical concepts enables us to see PE fit issues from a variety of perspectives. Most past research on PE fit has focused on the conceptualizations and validations of various forms of fit and their relationships between antecedents and consequences. For this reason, PE fit research to date seems narrowly focused, and there seem to be many opportunities to elaborate the theory of PE fit through applying more diverse perspectives. Sekiguchi (2006) suggests that research on PE fit should be extended from fundamental and nomological understandings to more applied and practical fields. As discussed below, the contingency perspective developed in this paper would contribute to the theory of PE fit not only by applying diverse perspectives, but also by considering practical usefulness. PJ fit and PO fit as a trade-off As stated earlier, the contingency perspective developed in this paper does not suggest that either PO fit or PJ fit is unimportant when the other type of fit is important. What this contingency perspective focuses on is the relative importance of PJ fit and PO fit. This implicitly assumes that there may be a trade-off between PJ fit and PO fit in actual hiring situations. The statement that PJ fit and PO fit are equally important for employees to be successful is intuitively appealing. However, organizations may not always be able to

125

JMP 22,2

126

find a person who is a “perfect fit.” The job applicant pool may not be large enough to find job candidates who are high in both PJ fit and PO fit. Some applicants may be high in one type of fit but not high enough in another type of fit. It might also be costly in some situations to attract a large number of job applicants and spend a long time in the selection process to find the person who fits with everything. In such cases, it is not only necessary but also efficient and effective for organizations to weigh a particular type of fit more than others in hiring decisions. Past research did not consider fully this kind of trade-off in developing models and hypotheses about applicant fit. The contingency perspective, which takes into account the relative importance of PE fit, may be a realistic approach to theorizing fit issues in employee selection. Hiring and managing a diverse workforce The contingency perspective proposed in this paper is based on the view that, to manage human resources effectively, organizations should have combinations of different employment practices for different groups of employees (Lepak and Snell, 1999; Osterman, 1987; Sonnenfeld and Peiperl, 1988). Changing the relative weights of PJ fit and PO fit as selection criteria according to the type of employees is consistent with this view. Past literature on PE fit in selection appears to assume only the case of hiring traditional (e.g. regular, full-time or permanent) employees, and most empirical studies are conducted using such selection contexts (e.g. Bretz et al., 1993; Cable and Judge, 1997; Kristof-Brown, 2000; Rynes and Gerhart, 1990). Although these studies are valuable, the consideration of different types of employees or employment relationships as contingency factors would contribute to the development of more elaborate theories of PE fit. Relationship with general intelligence and conscientiousness One of the robust findings in the employee selection research is the strong relationship between general intelligence and performance in almost all jobs (e.g. Schmidt and Hunter, 1998). General intelligence refers to the individual’s repertoire of knowledge and facility for acquiring, storing in memory, retrieving, combining, comparing and using information and conceptual skills in new contexts (Kanfer and Ackerman, 1989). In short, general intelligence represents a person’s learning ability. Findings from meta-analysis show that when job performance is measured objectively, the correlation with general intelligence measures is about 0.70, and when performance is measured using supervisor ratings, the correlation with general intelligence measures is over 0.60 for all jobs (Schmidt and Hunter, 2000). In addition, researchers contend that conscientiousness, one of the big-five personality dimensions, predicts job performance consistently across jobs (e.g. Barrick and Mount, 1991). Barrick and Mount (1991) concluded that the conscientiousness dimension significantly predicted all job performance criteria for all occupational groups represented in the data they analyzed. The integrity test, which mostly measures conscientiousness as well as some components of agreeableness and emotional stability, predicts job performance quite well, with a validity coefficient of 0.46 (Ones et al., 1993). The aforementioned research stream seems to conclude that, regardless of the degree of PJ fit and PO fit, higher intelligence and conscientiousness will lead to better

job performance. However, this does not deny the usefulness of PJ fit and PO fit as selection criteria. Whereas general intelligence and conscientiousness would predict job performance better, PJ fit and PO fit, as well as other types of PE fit, may predict various employee attitudes such as satisfaction and commitment more precisely. This indicates that the findings on general intelligence and conscientiousness and the contingency perspectives of PJ fit and PO fit presented in this paper would contribute to the effectiveness of employee selection in a complementary way. For example, in practice, it seems useful to prescreen job applicants by measuring general intelligence and conscientiousness, and to make final hiring decisions by assessing PJ fit and PO fit after considering the relative importance of each type of fit. The use of paper and pencil tests to measure general intelligence and conscientiousness is less costly and thus would be suited for the early stage of the selection process, whereas the careful assessment of PJ fit and PO fit is more costly and would be suited for the final stages of the selection process. Furthermore, assessing general intelligence and consciousness may sometimes work as preliminary assessments of PJ fit and PO fit. For example, when a job consists of complex cognitive tasks, an applicant’s general intelligence may be one of the antecedents of PJ fit, and when employee conscientiousness is part of the organizational culture, an applicant’s conscientiousness may be one of the antecedents of PO fit. Relationships between different theoretical perspectives The contingency perspective developed in this paper suggests that organizations should carefully examine what type of employees they will hire and what type of employment relationship they will develop before determining the appropriate criteria for selection. The three theoretical perspectives, namely, psychological contracts, human capital and cosmopolitan-local perspective, may often lead to the same conclusions. However, it may also be the case that different theoretical perspectives lead to conflicting conclusions. For example, when organizations hire contingent employees, they usually develop transactional psychological contracts with such employees. Furthermore, contingent employees are likely to be general human capital for organizations. In this case, it is clear from the propositions that organizations should consider PJ fit more than PO fit in selecting such employees. However, organizations may sometimes want to develop relational psychological contracts with professional employees who are usually cosmopolitans. In this case, P1b suggests that PO fit will be more important than PJ fit, but P3a suggests that PJ fit will be more important than PO fit. As a result, organizations should examine which of the contingency factors have stronger effects on post-hire outcomes and determine which theoretical perspective should be applied with higher priority when determining the relative importance of fit. Future research This paper contributes to the PE fit literature by providing the contingency perspective on the relative importance of fit in the context of employee selection. This perspective should encourage further empirical and theoretical work toward a more comprehensive contingency theory of PE fit. First, although a large amount of empirical research on PE fit has been conducted and findings have been accumulated, there is still a lack of research into the effects of

PJ fit and PO fit

127

JMP 22,2

128

employee types or employment relationships on the relationship between PE fit and various outcomes. Therefore, it is necessary to investigate empirically such moderating effects of employee types or employment relationships on the PE fit-outcome relationship, which would lead to the empirical examination of the contingency propositions offered in this paper. In addition, research is needed to investigate what organizations or decision makers actually do when they select employees for different types of job positions or with different employment relationships. Do organizations or individual decision makers actually give more weight to particular types of PE fit than to others when selecting particular types of employees? What type of cognitive process causes such a tendency? This line of research is useful for understanding the cognitive process of combining different types of PE fit information when making selection decisions and for examining whether the actual decision-making process is consistent with the contingency propositions developed in this paper. Second, further theoretical development and elaboration of the contingency perspective of PE fit is expected. To simplify the theoretical framework, this paper tends to focus on the opposite ends of the theoretical continuum (e.g. transactional vs relational psychological contracts), and focuses on the relative importance of PE fit rather than examining the importance of each type of fit separately. However, in some situations, the characteristics of a certain type of employee or employment relationship may lie somewhere between the opposite ends of the continuum or contain both opposite components. For example, some professional employees may have both cosmopolitan and local characteristics, and the latter would become greater as they are promoted to higher positions in the organizational hierarchy. Moreover, some employees may develop both general and firm-specific human capital, although the proportions of each type would be different. To examine these cases, a more integrative model of PE fit in selection could be developed in which different (often opposite) characteristics can be simultaneously considered and not only the relative importance of fit, but also the absolute importance of each type of fit, can be determined separately. Another way to elaborate the contingency perspective is to introduce more variables and wider contexts to the contingency model. A more comprehensive contingency model could include other types of PE fit such as PG fit and PV fit, and could cover a wider range of management practices, including external staffing, internal staffing and human resource development. Other contingency factors that affect the relative importance of PE fit in each practice could also be included. Possible candidates of such contingency variables include industry characteristics, labor-market variations, the legal and institutional environment, organizational culture and structure, and national culture. References Adkins, C.L., Russell, C.J. and Werbel, J.D. (1994), “Judgments of fit in the selection process: the role of work value congruence”, Personnel Psychology, Vol. 47, pp. 605-23. Barley, S.R. (1996), The New World of Work, British-North American Research, London. Barney, J. (1991), “Firm resources and sustained competitive advantage”, Journal of Management, Vol. 17, pp. 99-128. Barrick, R.M. and Mount, M.K. (1991), “The big five personality dimensions and job performance: a meta-analysis”, Personnel Psychology, Vol. 44, pp. 1-26.

Becker, G.S. (1964), Human Capital, Columbia University Press, New York, NY. Becker, G.S. (1976), The Economic Approach to Human Behavior, University of Chicago Press, Chicago, IL. Bowen, D.E., Ledford, G.E. and Nathan, B.R. (1991), “Hiring for the organization, not the job”, Academy of Management Executive, Vol. 5 No. 4, pp. 35-51. Bretz, R.D. and Judge, T.A. (1994), “Person-organization fit and the theory of work adjustment: implications for satisfaction, tenure, and career success”, Journal of Vocational Behavior, Vol. 44, pp. 32-54. Bretz, R.D., Rynes, S.L. and Gerhart, B. (1993), “Recruiter perceptions of applicant fit: implications for individual career preparation and job search behavior”, Journal of Vocational Behavior, Vol. 43, pp. 310-27. Cable, D.M. and DeRue, D.S. (2002), “The convergent and discriminant validity of subjective fit perceptions”, Journal of Applied Psychology, Vol. 87, pp. 885-93. Cable, D.M. and Judge, T.A. (1997), “Interviewers’ perception of person-organization fit and organizational selection decisions”, Journal of Applied Psychology, Vol. 82, pp. 546-61. Carson, K.P. and Stewart, G.L. (1996), “Job analysis and the sociotechnical approach to quality: a critical examination”, Journal of Quality Management, Vol. 1, pp. 49-66. Chatman, J.A. (1989), “Improving interactional organizational research: a model of person-organization fit”, Academy of Management Review, Vol. 14, pp. 333-49. Chatman, J.A. (1991), “Matching people and organizations: selection and socialization in public accounting firms”, Administrative Science Quarterly, Vol. 36, pp. 459-84. Edwards, J.R. (1991), “Person-job fit: a conceptual integration, literature review, and methodological critique”, in Cooper, C.L. and Robertson, I.T. (Eds), International Review of Industrial and Organizational Psychology, Vol. 6, Wiley, New York, NY, pp. 283-357. Fayol, H. (1916), Industrial and General Administration, Dunod, Paris. Ferris, G.R. and Judge, T.A. (1991), “Personnel/human resource management: a political influence perspective”, Journal of Management, Vol. 17, pp. 447-88. Golden, B.R., Dukerich, J.M. and Fabian, F.H. (2000), “The interpretation and resolution of resource allocation issues in professional organizations: a critical examination of the professional-manager dichotomy”, Journal of Management Studies, Vol. 37, pp. 1158-87. Goodman, S.A. and Svyantek, D.J. (1999), “Person-organization fit and contextual performance: do shared values matter”, Journal of Vocational Behavior, Vol. 55, pp. 254-75. Gouldner, A.W. (1957), “Cosmopolitans and locals: toward an analysis of latent social roles – I”, Administrative Science Quarterly, Vol. 2, pp. 281-306. Hall, R.T. (1968), “Professionalization and bureaucratization”, American Sociological Review, Vol. 33, pp. 92-104. Hoffman, B.J. and Woehr, D.J. (2006), “A quantitative review of the relationship between person-organization fit and behavioral outcomes”, Journal of Vocational Behavior, Vol. 68, pp. 389-99. Hulin, C.L. and Glomb, T.M. (1999), “Contingent employees: individual and organizational considerations”, in Illgen, D.R. and Pulakos, E.D. (Eds), The Changing Nature of Performance: Implications for Staffing, Motivation, and Development, Jossey-Bass, San Francisco, CA, pp. 87-118. Judge, T.A. and Ferris, G.R. (1992), “The elusive criterion of fit in human resource staffing decisions”, Human Resource Planning, Vol. 154, pp. 47-67.

PJ fit and PO fit

129

JMP 22,2

130

Kalleberg, A.L. (2000), “Nonstandard employment relations: part-time, temporary and contract work”, Annual Review of Sociology, Vol. 26, pp. 341-65. Kanfer, R. and Ackerman, P.L. (1989), “Motivation and cognitive abilities: an integrative/aptitude-treatment interaction approach to skill acquisition”, Journal of Applied Psychology, Vol. 74, pp. 657-90. Kerr, S., Von Glinow, A. and Schriesheim, J. (1977), “Issues in the study of professionals in organizations: the case of scientists and engineers”, Organizational Behavior and Human Performance, Vol. 18, pp. 329-45. Kristof, A.L. (1996), “Person-organization fit: an integrative review of its conceptualizations, measurement, and implications”, Personnel Psychology, Vol. 49, pp. 1-49. Kristof-Brown, A.L. (2000), “Perceived applicant fit: distinguishing between recruiters’ perceptions of person-job and person-organization fit”, Personnel Psychology, Vol. 53, pp. 643-71. Kristof-Brown, A.L., Zimmerman, R.D. and Johnson, E.C. (2005), “Consequences of individuals’ fit at work: a meta-analysis of person-job, person-organization, person-group, and person-supervisor fit”, Personnel Psychology, Vol. 58, pp. 281-342. Lazear, E. (1998), Personnel Economics for Managers, Wiley, New York, NY. Lepak, D.P. and Snell, S.A. (1999), “The human resource architecture: toward a theory of human capital allocation and development”, Academy of Management Review, Vol. 23, pp. 3-48. Lewin, K. (1951), Field Theory in Social Science, Harper & Row, New York, NY. McLean Parks, J., Kidder, D.L. and Gallagher, D.G. (1998), “Fitting square pegs into round holes: mapping the domain of contingent work arrangements onto psychological contract”, Journal of Organizational Behavior, Vol. 19, pp. 697-730. Mintzberg, H. (1973), The Nature of Managerial Work, Harper & Row, New York, NY. Mitchell, T.R., Holtom, B.C., Lee, T.W., Sablynski, C.J. and Erez, M. (2001), “Why people stay: using job embeddedness to predict voluntary turnover”, Academy of Management Journal, Vol. 44, pp. 1102-21. Montgomery, C.E. (1996), “Organization fit is a key to success”, HRM Magazine, January, pp. 94-6. Ones, D.S., Viswesvaran, C. and Schmidt, F.L. (1993), “Comprehensive meta-analysis of integrity test validities: findings and implications for personnel selection and theories of job performance”, Journal of Applied Psychology, Vol. 79, pp. 679-703. O’Reilly, C.A. and Chatman, J. (1986), “Organization commitment and psychological attachment: the effects of compliance, identification and internalization on prosocial behavior”, Journal of Applied Psychology, Vol. 71, pp. 492-9. O’Reilly, C.A., Chatman, J. and Caldwell, D.F. (1991), “People and organizational culture: a profile comparison approach to assessing person-organization fit”, Academy of Management Journal, Vol. 34, pp. 487-516. Organ, D.W. (1988), Organizational Citizenship Behavior: The Good Soldier Syndrome, Lexington Books, Lexington, MA. Osterman, P. (1987), “Choice of employment systems in internal labor markets”, Industrial Relations, Vol. 26, pp. 46-67. Raelin, J.A. (1994), “Three scales of professional deviance within organizations”, Journal of Organizational Behavior, Vol. 15, pp. 483-501. Rousseau, D.M. (1989), “Psychological and implied contracts in organizations”, Employee Responsibilities and Rights Journal, Vol. 2, pp. 121-39.

Rousseau, D.M. (1995), Psychological Contracts in Organizations: Understanding Written and Unwritten Agreements, Sage, Newbury Park, CA. Rousseau, D.M., Sitken, S.B., Burt, R.S. and Camerer, C. (1998), “Not so different at all: a cross discipline view of trust”, Academy of Management Review, Vol. 23, pp. 393-404. Rynes, S.L. and Gerhart, B. (1990), “Interviewer assessments of applicant ‘fit’: an exploratory investigation”, Personnel Psychology, Vol. 43, pp. 13-35. Schmidt, F.L. and Hunter, J.E. (1998), “The validity and utility of selection methods in personnel psychology: practical and theoretical implications of 85 years of research findings”, Psychological Bulletin, Vol. 134 No. 2, pp. 262-74. Schmidt, F.L. and Hunter, J.E. (2000), “Select on intelligence”, in Locke, E.A. (Ed.), The Blackwell Handbook of Organizational Behavior, Blackwell Publishing, Malden, MA, pp. 3-14. Sekiguchi, T. (2006), “How organizations promote person-environment fit: using the case of Japanese firms to illustrate institutional and cultural influences”, Asia Pacific Journal of Management, Vol. 23, pp. 47-69. Sonnenfeld, J.A. and Peiperl, M.A. (1988), “Staffing policy as a strategic response: a typology of career systems”, Academy of Management Review, Vol. 13, pp. 588-600. Tsui, A.S., Pearce, J.L., Porter, L.W. and Hite, J.P. (1995), “Choice of employee-organization relationship: influence of external and internal organizational factors”, in Ferris, G.R. (Ed.), Research in Personnel and Human Resource Management, Vol. 13, JAI Press, Stamford, CT, pp. 117-51. US Department of Labor, Bureau of Labor Statistics (2002), Occupational Outlook Handbook, 2002-2003 Edition, US Department of Labor, Bureau of Labor Statistics, Washington, DC. Vancouver, J.B., Millsap, R.E. and Peters, P.A. (1994), “Multilevel analysis of organizational goal congruence”, Journal of Applied Psychology, Vol. 79, pp. 666-79. Verquer, M.L., Beehr, T.A. and Wagner, S.H. (2003), “A meta-analysis of relations between person-organization fit and work attitudes”, Journal of Vocational Behavior, Vol. 63, pp. 473-89. Wallace, J.E. (1993), “Professional and organizational commitment: compatible or incompatible?”, Journal of Vocational Behavior, Vol. 42, pp. 333-49. Wallace, J.E. (1995), “Organizational and professional commitment in professional and nonprofessional organizations”, Administrative Science Quarterly, Vol. 40, pp. 228-55. Werbel, J.D. and Gilliland, S.W. (1999), “Person-environment fit in the selection process”, in Ferris, G.R. (Ed.), Research in Personnel and Human Resource Management, Vol. 17, JAI Press, Stamford, CT, pp. 209-43. Zajac, E.J., Golden, B.R. and Shortell, S.M. (1991), “New organizational forms for enhancing innovation: the case of internal corporate joint ventures”, Management Science, Vol. 37, pp. 170-84. Corresponding author Tomoki Sekiguchi can be contacted at: [email protected]

To purchase reprints of this article please e-mail: [email protected] Or visit our web site for further details: www.emeraldinsight.com/reprints

PJ fit and PO fit

131

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

JMP 22,2

Attracting for values: an empirical study of ASA’s attraction proposition

132

Jon Billsberry Centre for Human Resource and Change Management, Open University Business School, The Open University, Milton Keynes, UK Abstract Purpose – This paper seeks to report an empirical test of Schneider’s attraction proposition that organizations attract similar types of people. Design/methodology/approach – The person-people (PP), person-group (PG) and person-organization (PO) fit of applicants to eight utility firms were compared with the similar fits of members of a suitable comparison group. Findings – The results show an effect for person-vocation (PV) fit but, once this is controlled for, all significant effects disappear. In other words, the PP, PG and PO fits of applicants to the utilities were no different from those in the comparison group once PV fit was controlled for. Research limitations/implications – These results suggest that applicants choose which organization to apply to based on their desire for a particular type of work rather than their attraction for particular companies, which is contrary to Schneider’s attraction proposition. One reason for this might be the nature of graduates who are largely unaware of the organizational environments to which they are applying. Drawing from the interpersonal attraction literature, it is argued that applicants need familiarity, proximity and exposure to the organization for Schneider’s attraction proposition to appear. Practical implications – These results suggest graduates choose vocations over organizations and that, if organizations wish to attract people who share their values, they need to put more effort into their recruitment efforts than those described in this study. The absence of an attraction effect suggests that, when applicants make attraction decisions based on their perceived fit, they may not be supported by an actual congruence. Originality/value – This study extends previous work by including multiple conceptualizations of fit, by including person-group and person-people fits in addition to person-organization fit. Keywords Recruitment, Organization, Graduates Paper type Research paper

ASA theory was developed by Schneider (1983a, b, 1985, 1987) (Schneider et al., 1995, 1998, 2000) as an explanation of why organizations look, feel and behave as they do. The main idea in this framework is that organizations attract, select, and retain similar types of people. Schneider (1987) argues that this similarity limits the actions of the organization owing to the fact that it occupies a constrained niche of like-minded employees sharing similar values, personalities and attitudes. Schneider et al. (1995) call this effect the “homogeneity hypothesis” and they predict that it will cause organizational dysfunctionality, as they become increasingly ingrown and resistant to change. Journal of Managerial Psychology Vol. 22 No. 2, 2007 pp. 132-149 q Emerald Group Publishing Limited 0268-3946 DOI 10.1108/02683940710726401

The author would like to acknowledge and thank Kevin Daniels, Jo Silvester, Chris Coupland, Amy Kristof-Brown, Peter Herriot, Tim Clark, Nathalie van Meurs, Michael Morley and anonymous reviewers for their help and advice. The usual disclaimer stands.

A number of studies have directly tested and found support for the homogeneity hypothesis (e.g. Denton, 1999; Jackson et al., 1991; Jordan et al., 1991; Schneider et al., 1998), leading to the conclusion that it is phenomenon worthy of interest especially given the proposed nature of its dire organizational consequences. Of the three processes involved in the ASA cycle that contribute to the formation of homogeneity, much is known about the socialization phase and its influence (e.g. Chatman, 1991; O’Reilly et al., 1991; Kristof, 1996), but much less is known about the first two phases: attraction and selection. The studies that have been conducted on these two phases of the cycle have either been conducted in laboratory settings or on perceptions of fit, rather than actual fit. The current study addresses this gap by focusing on the attraction phase of the cycle and, in particular, on whether the actual fit of applicants to the values of the recruiting organization predicts their application. This study also extends previous work by including multiple conceptualizations of fit, by including person-group (PG) and person-people (PP) fits in addition to person-organization (PO) fit. The paper begins with a review of the published studies on the attraction phase of the cycle that draws out considerations for the present study, before moving on to the report of empirical study itself. Tests of Schneider’s attraction proposition Initial studies of Schneider’s attraction proposition were conducted in laboratory settings and were based on the well-known effect that similarity leads to attraction (e.g. Bretz et al., 1989; Judge and Bretz, 1992; Turban and Keon, 1993; Cable and Judge, 1994). Bretz et al. (1989) conducted a laboratory experiment that measured personality and then presented the participants with descriptions of organizations with reward systems that had been manipulated. They found that people with a high need for achievement disproportionally chose organizations with individually-focused reward systems. Judge and Bretz (1992) surveyed students for their values and then presented them with a number of work scenarios in which 11 organizational and value variables were manipulated. They were asked the probability of whether or not they would accept a job offer from the organization in each scenario. As predicted by the researchers, value alignment between students and the manipulated scenarios was positively related to the students’ job choice decisions. Turban and Keon (1993) asked management students to indicate their attraction to paper descriptions of organizations in which various characteristics were manipulated. They found that people with a high need to achieve were more attracted to organizations that offered a merit-based reward structure (i.e. those that rewarded performance over seniority) than people with a low need to achieve. They also showed that people with low self-esteem were more attracted to decentralized organizational structures (and larger firms) than people with high self-esteem. These results suggest that people are attracted to organizations that mirror their personality. In a similar study, Cable and Judge (1994) examined whether congruence of one aspect of the organizational environment, the pay and compensation system, with individual personality traits influenced the job search decisions of engineering and hotel administration students approaching graduation. The researchers asked engineering and hotel administration students to evaluate 32 different pay and compensation scenarios. They found that the attractiveness of the pay policies of organizations was

Attracting for values

133

JMP 22,2

134

heightened by greater levels of “fit” between individual personality traits and the characteristics of the compensation system. These studies provide interesting glimpses of the consequences of applicants’ similarity to various organizational constructs on their attractiveness of organizations supporting the “similarity leads to attraction” hypothesis. However, they all have a laboratory form in which non-applicants rate the attractiveness of hypothetical organizations. Such laboratory experiments are likely to over-emphasize situational variables and mask individual differences (Bowers, 1973; Schneider, 1987). Moreover, whilst these studies provide support for the hypothesis that similarity leads to attraction, the decisions of people in artificial situations does not necessarily accord with actual behaviour. Other studies, therefore, have explored the behaviour and perceptions of “real” applicants. Cable and Judge (1996) investigated the PO fit perceptions of job seekers. The job seekers were undergraduate applicants to 35 organizations who were recruiting at a large university in the north east of the US. The job seekers were asked to report their perceptions of the attractiveness of the job attributes for which they had just been interviewed, their perceived fit with the company and the job, and they completed a modified version of the organizational culture profile (OCP) (O’Reilly et al., 1991) to report their perceptions of the company’s values. The researchers found that applicants’ perceived value congruence was predictive of their perceptions of their own fit. The researchers also found that job seekers’ perceived value congruence (OCP derived) was associated with their job choice intentions (r ¼ 0:24, p , 0:05, n ¼ 273), albeit weakly. A year later, the same authors published a follow-up study that employed a similar design (Judge and Cable, 1997). In this follow-up, they also captured students’ big five personality traits. The authors found similar findings with both subjective (direct questions) and objective (OCP derived) PO fit being associated with organizational attractiveness. Returning to a laboratory design, Dineen et al. (2002) extended Cable and Judge’s (1996, 1997) work by capturing the subjective and objective (OCP derived) PO fit when attraction was conducted through web-based methods. The results of their experiment in which feedback was manipulated showed that attraction was related to both subjective and objective PO fit. Kristof-Brown et al.’s (2005) recent meta-analysis usefully summarizes studies between PO fit and organizational attraction. Where PO fit was calculated through direct measures (i.e. perceived fit), they found that the average correlation between PO fit and organizational attraction (n ¼ 10) was 0.52, thereby suggesting that people are attracted to organizations where they believe they will fit in, which is just a short step away from demonstrating that applicants choose to join organizations that they “actually” fit. This is the gap that the current study addresses. Specifically, this study examines whether or not people apply to organizations whose values they share. Multiple conceptualizations of fit PO fit is a specific form of person-environment (PE) fit focusing on the interaction of people and organizational environments. Like PE fit, PO fit is a cross-levels construct that can be conceptualized in many forms (Cable and DeRue, 2002; Kristof, 1996; Kristof-Brown et al., 2005; Van Vianen, 2000). The most common form of cross-level conceptualisation is to compare something at the individual level (e.g. values or

personality) to something at the organizational level (e.g. values, goals, mission) to arrive at a measure of PO fit. In addition, researchers have differently conceptualised the “E” part of the interaction to include individuals’ fit to vocations (person-vocation fit; PV fit), groups (person-group fit; PG fit) and other individuals (person-people fit; PP fit). These different conceptualizations of fit are thought to have differing salience to individuals. In addition to PO fit, both PG and PP forms of fit seem particularly relevant during attraction. Many PO fit researchers have captured the values of their target organizations via Top Team (TT) members (e.g. Chatman, 1991; O’Reilly et al., 1991). There are several problems with this approach. One of which is a fear of anthropomorphism; can organizations have values? An alterative method is for the top team members to consider the values that organizational members hold, but this leads to a generalized or average approach. Research in other fields shows that top team members can be detached from the day-to-day operation of the organization. This, combined with their seniority, can skew the nature of the behaviours they observe and experience (Ambrosini and Bowman, 2002; Hambrick and Mason, 1984). In such situations, the resulting assessments of the “O” element may distort PO fit calculations away from value congruence as experienced by people in the organization. In contrast, department members are, by definition, involved in the day-to-day work of the organization and thereby intimately exposed to, familiar with, and proximal to work values. Their assessments of work values are likely to be grounded in the realism of actual work that they experience and interact with on a daily basis. As a result, it could be argued that by having departmental members rather than TT members define the “O” element, a version of PO fit that is more grounded in the reality of organizational life will be arrived at. Versions of PO fit where the ‘O’ element of PO fit is determined by a subgroup is commonly termed person-group (PG) fit even though it might be more accurate to call it person-department or person-subculture fit (Kristof, 1996). Simply for convenience, the term PG fit will be used in this study to describe this form of fit. PP fit is interesting because it compares “like with like”, e.g. it compares the values of applicants to the values of existing employees. As such, PP fit is much closer to Schneider’s original attraction proposition that organizations attract similar types of people creating homogeneity amongst employees than PO fit is. Combining Schneider’s attraction proposition with these multiple conceptualizations of PO fit, three hypotheses are advanced: (1) Applicants’ PO fit will be greater than the PO fit of suitably qualified applicants contemporaneously looking for a similar type of work, but who did not apply. (2) Applicants’ PG fit will be greater than the PG fit of suitably qualified applicants contemporaneously looking for a similar type of job, but who did not apply. (3) Applicants’ PP fit will be greater than the PP fit of suitably qualified applicants contemporaneously looking for a similar type of job, but who did not apply. Method Sample and procedure Data were gathered by distributing questionnaires to applicants for graduate entry managerial trainee posts in nine utility companies in the UK. Each of these companies was functionally structured. Only a small number of departments in each company

Attracting for values

135

JMP 22,2

136

sought graduate entrants. These functional departments included finance, marketing and sales, engineering, information technology, and human resources. In total, 19 different departments spread across the nine utility companies sought graduates. Research questionnaires were sent to applicants by staff in the human resources departments. The procedure was as follows. If someone was interested in applying to one of the organizations they would get a brochure containing an application form from their university’s careers centre, at a corporate presentation, or by phoning the organization to request one. Every application was acknowledged with a letter. It was with these acknowledgement letters that the research questionnaire was sent to applicants. To reinforce the point that the questionnaire was only being used for academic purposes, applicants were asked to send their completed questionnaires to the researcher at a university address in a pre-paid, pre-addressed envelope. In total, the companies received applications from 825 different people. Of these, 621 applicants returned completed questionnaires, which is a response rate of 72.3 per cent. A total of 54 per cent were male and 46 per cent female. The average age of applicants was 23 years and 6 months with a standard deviation of 4 years and 5 months. The youngest applicant was 19 years and the oldest was 49 years and 6 months. To capture the values of the general population that applications might come from, questionnaires were sent to the careers services of the 11 UK universities who historically had supplied a disproportionately large percentage of applicants to these organizations. The process for distributing the questionnaires was as follows. A total of 50 copies of the questionnaire were sent to each university careers centre. The questionnaires were placed on the counter of the careers service; each was accompanied by an explanatory letter and a prepaid envelope. Careers service staff asked users of the service, i.e. people looking for work as graduate entrants, to complete a questionnaire. These questionnaires were distributed at the same time that people were applying to the organizations. The questionnaires asked job seekers for their values in the same manner as applicants to the corporations. In addition, they were asked to report their degree course, the type of work they were looking for, their gender, date of birth, whether they have had full-time or part-time work and, if so, for how long. In total, 550 questionnaires were distributed and 171 (31.1 per cent) were returned completed (two were not completed properly). There was a mixed response rate from the 11 universities. Only four questionnaires (8 per cent) were returned from one university, whereas the most responsive returned 28 (56 per cent). Amongst these questionnaires, there were 136 responses that could be categorized as looking for work in a similar type of department to those being offered by the companies (38 per cent male, 62 per cent female). Some eight people did not respond to this part of the questionnaire and 27 expressed a desire for a different type of work. This comparison group contained the same average age and age spread as the applicants. Measures Applicant values. Previous researchers have used the OCP to assess hypotheses associated with Schneider’s selection proposition (e.g. Chatman, 1991). However, the original card sort is impractical when there are a large number of remote and geographically dispersed respondents such as in this study (Block, 1978; Kerlinger, 1986; Nunnally, 1978). Cable and Judge’s (1996, 1997) reworked version of the OCP as a paper and pencil test fails to replicate the cognitive processes of the original and is

difficult to complete without assistance (Barber and Wesson, 1998) rendering it impractical with geographically dispersed respondents. Following Barber and Wesson’s (1998) advice, 23 items from the OCP were reworded for context-specificity (Schwab, 1980; Schwartz, 1992, 1994; Stackman et al., 2000) and presented to applicants and organizational members as a Likert-scaled questionnaire. Items, prompts and scales can be found in the Appendix. Corporate values. Three sets of values were gathered from organizational members. The first and second set of values came from employees in the departments seeking graduate trainees (e.g. marketing, finance). These people were asked to report their own values (to enable the calculation of PP fit) and the values of their department (to enable the calculation of PG fit). The third set of values came from the members of the top team of each utility (to enable the calculation of PO fit). In all circumstances, where there was evidence that the department members or senior managers did not agree on the nature of values of their department or organization, i.e. when reliability coefficients fell below 0.7 (George and Mallery, 1995), the department or organization was removed from all subsequent analysis. Statistical analysis. Fit was calculated using the sum of absolute differences method. This method was chosen because the comparative nature of the hypotheses prevented the use of polynomial regression. Although there are 621 applicants in this study, once they are separated into their various organizations and groups, sample sizes fall dramatically to a point where it is inappropriate to use multivariate statistics. This sample size issue is further complicated by the organizational and group specificity of the “O”, “G” and “P” parts of the PO, PG and PP fit equations making it impossible to combine organizations or groups to improve statistical leverage. As a result, the hypotheses were tested using t-tests, which have the benefit of being easily interpreted. Results To test the first hypothesis, an independent samples t-test was conducted to compare the PO fit of applicants and members of the comparison group who were seeking work in commercial organizations. In one of the organizations, there was a statistically significant difference in the proposed direction (applicants n ¼ 12, mean ¼ 231:24, SD ¼ 5:22; comparison group n ¼ 135, mean ¼ 234:87, SD ¼ 6:14; t ¼ 21:982, sig: ¼ 0:025), but in the other six there were no significant differences. It is possible that this way of conceptualizing the comparison group contains an element of PV fit, which might inflate the differences between the two constituencies. This is possible because there is an imbalance in the type of work (i.e. vocation) sought by the people in the two constituencies. To eliminate potential contamination by PV fit, the same analysis was run again, but this time it was conducted department by department and only those people in the comparison group wanting the type of work of the department were included. This leaves firm-specific PO fit. The reworked numbers are contained in Table I. A one-tailed significance test was employed because the differences in mean fit scores were hypothesized as being in one direction. The results displayed in Table I control for applicants’ choice of work: for example, it compares the PO fit of applicants wanting work as accountants, say, to the same fit of people in the comparison group wanting to work in finance and related fields. In nine of the fourteen cases, applicants’ fit is greater than non-applicants’ fit, although none of these differences are statistically significant. In the remaining five cases, the

Attracting for values

137

Function

Finance

Management and customer services

Engineering

IT

Finance

Sales and marketing

Finance

Finance

Management and customer services

Engineering

IT

Sales and marketing

Engineering

Strategy and dev.

1

1

1

2

3

4

5

6

6

6

6

6

7

7

Table I. Comparison of PO fit of applicants against people in the comparison group who want similar work

Company

group

group

group

group

group

group

group

group

group

group

group

group

group

group

18 9 38 36 18 21 22 47 18 6 14 12 18 25 18 14 38 13 18 11 22 41 14 23 18 22 12 26

n 2 30.57 2 28.11 2 31.22 2 31.93 2 28.96 2 30.64 2 32.89 2 30.75 2 34.85 2 33.85 2 31.68 2 31.24 2 34.94 2 34.10 2 38.44 2 36.29 2 37.38 2 36.04 2 36.72 2 39.00 2 38.70 2 38.39 2 33.21 2 35.43 2 29.40 2 31.58 2 32.96 2 30.20

Mean 6.45 4.83 7.17 5.99 5.18 6.64 5.50 5.95 5.85 4.73 3.98 5.22 6.90 5.93 6.20 7.28 6.67 6.57 5.61 3.15 6.30 7.10 5.78 7.51 4.99 5.64 7.90 5.47

SD

2 1.250

1.279

0.947

2 0.174

1.228

2 0.629

2 0.906

2 0.431

2 0.244

2 0.378

2 1.423

0.870

0.463

2 1.008

t

138

Comparison Applicants Comparison Applicants Comparison Applicants Comparison Applicants Comparison Applicants Comparison Applicants Comparison Applicants Comparison Applicants Comparison Applicants Comparison Applicants Comparison Applicants Comparison Applicants Comparison Applicants Comparison Applicants

Status

0.110

0.110

0.175

0.432

0.115

0.266

0.186

0.334

0.410

0.355

0.080

0.195

0.323

0.162

Sig. (1-tailed)

JMP 22,2

comparison groups’ fit are greater than applicants’ fit, although, yet again, none of these differences are statistically significant. Overall, these results provide no support for the first hypothesis. In short, these results do not support the idea that applicants have a better PO fit than the comparison group once PV fit is controlled for. The second hypothesis was tested in a similar fashion to the first. Again, the initial conceptualizing of the comparison group as a whole, rather than by splitting it by vocation, created effects with the two recruiting sales and marketing departments exhibiting statistically significant differences in the predicted direction. As a result, the comparison group was split with just those people wanting the same type of work as the applicants being included thereby controlling for PV fit. The results are displayed in Table II. Table II shows that just one of the eleven t-tests has a result at the p , 0:05 level and that is not in the hypothesized direction. This analysis is particularly interesting because it shows that when PV fit is controlled for, no attraction effect for PG fit can be found with this data set. Similar results are exhibited when the type of fit is PP fit with significant effects found for the sales and marketing and finance departments when the comparison group is not separated by vocation. Once PV fit is controlled for, all significant differences disappear causing the third hypothesis to be rejected. These results are displayed in Table III. Although these results reject the three hypotheses, they do not necessarily imply that Schneider’s attraction proposition should also be rejected. Measurement error and low statistical power could both account for the lack of findings. Nevertheless, there are some reasons to accept that Schneider’s attraction proposition may not be present with Company

Department

Status

1

Finance

1

Engineering

2

IT

4

Sales and marketing

5

Finance

6

Human resources

7

Finance

7

Engineering

7

IT

7

Sales and marketing

8

Engineering

Comparison Applicants Comparison Applicants Comparison Applicants Comparison Applicants Comparison Applicants Comparison Applicants Comparison Applicants Comparison Applicants Comparison Applicants Comparison Applicants Comparison Applicants

group group group group group group group group group group group

n

Mean

SD

t

Sig. (1-tailed)

18 9 18 21 22 47 14 12 18 25 13 39 18 14 18 11 22 41 14 23 18 22

2 37.92 2 35.02 2 34.45 2 37.71 2 33.98 2 33.06 2 29.96 2 28.29 2 35.13 2 33.85 2 24.80 2 26.84 2 36.99 2 36.01 2 40.83 2 42.51 2 34.10 2 32.83 2 29.46 2 30.35 2 31.88 2 33.51

5.33 5.23 4.99 6.13 4.86 6.04 4.74 4.72 5.66 3.66 5.86 5.23 5.84 5.34 4.89 2.79 4.83 5.92 4.11 5.61 3.92 6.12

21.340

0.096

1.803

0.040

20.628

0.262

20.899

0.189

20.902

0.181

1.180

0.122

20.486

0.316

1.034

0.155

20.864

0.196

0.511

0.307

0.975

0.168

Attracting for values

139

Table II. Comparison of PG fit of applicants against people in the comparison group who want a similar type of work

Department

Engineering

IT

Finance

Sales and marketing

Finance

Human resources

Finance

Management and customer services

Engineering

Sales and marketing

Engineering

1

2

3

4

5

6

7

7

7

7

8

Table III. Comparison of PP fit of applicants against people in the comparison group who want a similar type of work

Company

group

group

group

group

group

group

group

group

group

group

group

18 21 22 47 18 6 14 12 18 25 13 39 18 14 38 13 18 11 14 23 18 22

n 2 23.66 2 23.54 2 24.28 2 22.79 2 23.77 2 23.53 2 22.68 2 21.75 2 24.71 2 22.67 2 21.17 2 21.94 2 23.48 2 23.59 2 27.26 2 23.97 2 22.18 2 22.85 2 21.71 2 23.25 2 23.92 2 25.47

Mean 3.62 6.31 4.23 4.89 6.18 5.56 5.81 3.98 5.94 3.67 3.04 4.85 5.41 7.46 7.37 4.84 4.10 5.41 5.89 4.23 3.84 5.40

SD

1.019

0.922

0.382

2 1.497

0.047

0.539

2 1.388

2 0.467

2 0.082

2 1.234

2 0.074

t

140

Comparison Applicants Comparison Applicants Comparison Applicants Comparison Applicants Comparison Applicants Comparison Applicants Comparison Applicants Comparison Applicants Comparison Applicants Comparison Applicants Comparison Applicants

Status

0.157

0.182

0.353

0.071

0.481

0.298

0.087

0.323

0.468

0.111

0.471

Sig. (1-tailed)

JMP 22,2

these applicants rather than looking to statistical weaknesses. The presence of the vocation effect suggests that the instrument was sufficiently sensitive to capture effects, although, of course, this might just mean that it is attuned to PV fit rather than PO, PG and PP fit. Discussion This study looked at whether the people who apply to an organization are a better fit than a sample of people who were looking for similar types of work at the same time through the same process. This analysis was conducted by comparing the fit of applicants to the fit of (1) people in the comparison group contemporaneously seeking commercial work through university careers services and (2) people in the comparison group contemporaneously seeking a similar type of work through university careers services. The results of these tests suggest that once PV fit is controlled for, the applicants in this sample do not have a better fit than the people in the comparison group: these graduates appear to be making vocational rather than organizational choices. As a result, the data in this study failed to support Schneider’s attraction proposition. The researcher was able to talk to the heads of four of the university careers centres during the data-gathering period to discuss graduates’ approach to application. Generally speaking, graduates decide which types of jobs they want to do and then apply for all relevant openings that seem attractive (usually judged on starting salary and location), especially when application is allowed by standard application form. For example, someone decides that he or she wishes to train as an accountant within industry and then applies to all those organizations offering such training. He or she might exclude all those companies offering less than £20,000 starting salary that are not located in the south east of England. The heads of the university career centres quoted examples from all manner of professions and jobs to support this approach. Such an approach to job search has been widely reported in the literature (e.g. Herriot, 1984; Keenan, 1997; Rynes et al., 1997). As one careers centre head put it, “we’re called careers centres or services, not companies centres”. Not surprisingly, therefore, the data accords with applicants’ behaviour and reveals the vocational effect, but no organization-specific effect. Returning to Schneider’s original paper, it is interesting to note that he offers very little to support the notion of a firm-specific attraction effect; his main justification for advocating a firm-specific attraction effect is the vocational choice literature. He quotes Holland (1976, p. 533) thus, “vocational choice is assumed to be the result of a person’s type, or patterning of types and the environment” (Schneider, 1987, p. 441). Schneider concludes by suggesting that “the career environments people join are similar to the people who join them” (p. 441). Schneider extrapolates from the findings of Holland (1976), Tom (1971) and Owens (Neiner and Owens, 1985; Owens and Schoenfeldt, 1979) to posit that people will be attracted to “organizations of a particular sort” (p. 442) in the same way that they are attracted to vocations and jobs. By conducting the present study in utility companies with functional departments, it was possible both to observe this vocational choice effect and to control for it. Once controlled for, there was no firm-specific effect: these people were attracted to work in a sales and marketing department because it is a sales and marketing department, not because the company is particularly attractive to them.

Attracting for values

141

JMP 22,2

142

Schneider’s attraction hypothesis is so intuitive and compelling that it is interesting to discuss whether or not there are reasons why the present study might not have found the effect, when, in many other cases, it might be present (a form of type II error). Perhaps the most obvious study-specific constraint is the choice to study graduates entering work. Such an environment is compelling to the recruitment and selection researcher as it is one of the few planned opportunities to study organizational entry with the volumes required to produce statistical significance (Rynes et al., 1997). In addition, procedures for graduate entry tend to adhere to the principles of rigorous personnel selection, which informs the dominant selection paradigm (Schmitt and Chan, 1998). The allure of graduate entry is so powerful that it dominates the PO fit literature (Rynes et al., 1997). The influential studies of Rynes and Gerhart (1990), Chatman (1991), Cable and Judge (1996, 1997), and Kristof-Brown (2000) all considered graduates or graduate entry recruiters. Given the findings of the present study (i.e. that attraction is to the vocation not the organization), the focus on graduates seems inappropriate when the study seeks to discover PO fit influences. In the present study most of the graduates were seeking their first “real” job. There are implications of this, which are well recorded in the literature that considers the transition of graduates into the world of work. Nicholson and Arnold (1991), for example, demonstrated that graduates entered work with unrealistic expectations. Keenan and Newton (1984, 1986) found that engineering graduates quickly became frustrated because their aspirations were not met, which was also a finding of Mabey (1986), suggesting unrealistic expectations. Arnold (1985) reports that graduates found “the general atmosphere at work” (p. 308) the most commonly reported source of surprise. In addition, half of his graduate respondents said that “what the people are like with whom you work” (p. 313) was a surprise to them. The overall sense of this literature is that graduates enter organizations as “innocents abroad”. They know little of the world of work, working environments, or the employing organizations. Given the consistent finding that graduates hold unrealistic expectations, one has to wonder whether graduates are able to make objective assessments about the nature of the organizations to which they apply. The second study-specific factor that might account for the absence of an attraction effect is the choice of utility companies as the location of the study. Among the reasons why this was chosen was that they exemplify the “typical” large organizations that recruit graduates. Graduates would almost certainly have heard of the names of these companies, and graduates may have been more aware of them than usual due to the news flow surrounding the privatization of the utilities in the preceding decade. But whilst the names will have been recognizable to graduates, these companies are largely indistinguishable from other large corporations. If Schneider was right, this would not matter; his framework, after all, does not contain any provisos or clauses limiting it to a particular type of organization or industry (except for the enigmatic phrase “organizations of a particular sort”, (Schneider, 1987, p. 442)). As a representative of the “grey corporations” of the UK, the organizations providing the location for the present study provide a neutral reference to explore for a general attraction effect. That such an effect was not found with this corporation makes extrapolation to other “grey corporations” reasonably straightforward. However, it does beg the question of what sort of organization might exhibit an attraction effect. Presumably one with strong values that are visible to the applicant population.

The last point raises a third study-specific issue. This study used an instrument that captured 23 work values. Generally speaking, these values were operationalised as items depicting the behaviour of employees following the guidance of Stackman et al. (2000); i.e. as instrumental values. Such values may not, of course, correspond with the values that applicants use to make judgments about the attraction of organizations. Schneider (1987) cites the example of the YMCA as an organization that might exhibit an attraction effect. The YMCA has a strong and visible set of values connected to religion, social welfare and support, which are linked to the particular organization. This produces several avenues for researchers wishing to find the attraction effect. The first of these might be to take a more conceptual approach to values. By focusing on the behaviours of employees, applicants need an understanding of work within organizations, which previous studies have shown that graduates have unrealistic expectations of. Perhaps looking at the behaviour of organizations and conceptualizing values at this level, although it risks the anthropomorphism trap, might be a fruitful way forward. A second avenue might be to adopt an individual approach with the researcher studying the particular values of an organization and then measuring these values in employees, applicants, and relevant non-applicants. Such an approach has been suggested by Kristof-Brown (1997) for the capture of organizations’ values. The third avenue might be to look for an alternative currency altogether. Values were used in this study as they represented a logical extension of previous work (e.g. Cable and Judge, 1996, 1997; Chatman, 1991; O’Reilly et al., 1991) and because Schneider’s own clarifications suggested that values were paramount. Perhaps, goals, personality, climate, vision or some other dimension might show up an attraction effect. Finally, it is interesting to return to the attraction literature for an explanation of why the attraction effect might not have been found. Other than referencing the vocational choice literature, Schneider’s other justification for the attraction effect is the well-demonstrated “similarity leads to attraction” phenomenon. However, similarity does not stand in isolation as the reason for attraction. Other factors, such as proximity, exposure, and familiarity, all influence the ability of similarity to influence attraction (e.g. Festinger et al., 1950; Gross, 1992; Saegert et al., 1973; Turban, 2001). In the case of the behaviour of graduate applicants during recruitment, their exposure to hiring organizations, other than their general life experience, is the brochure and a short presentation. The brochures for the utilities in the present study were very brief with just a few sides of marketing material and a page describing each business unit and the opportunities they offered. The presentations were “manicured” to project the organizations in the most favourable light, as was the case for most organizations’ presentations to graduates. These presentations occurred midway through the application process and so might have only had a marginal effect on enticing more applications. Hence, most applicants’ exposure to the recruiting companies is likely to have been very slight. The earlier discussion on graduates’ entry into work demonstrated that they have little familiarity with organizational life. And proximity is not relevant given the clustering of applicants in locations remote to the recruiting organization. As a result, it seems unlikely that these remote applicants can gain sufficient understanding of the recruiting organizations’ values for the congruence of

Attracting for values

143

JMP 22,2

144

values to have any significant effect on their application behaviour. Not surprisingly, therefore, no attraction effect based on value congruence could be found. This discussion prompts the question of what circumstances it might be possible to find the attraction effect. From the above, several conditions can be advanced: (1) Potential applicants must understand the nature of work and organizational behaviour. (2) The values of recruiting organizations must be visible. (3) Potential applicants must be able to assess recruiting organizations’ values, which highlights, proximity, exposure and familiarity. Two environments suggest themselves. The first is the largely unresearched world of internal recruitment (Harris, 2000). In this environment, potential applicants are already employed by the recruiting organization and therefore are likely to be fully aware of the environment to which they would like to move. They are likely to understand exactly how the environment differs to their current environment and are able to find out more about the environment by talking to colleagues to fill any holes in their knowledge. The second environment is those industries or professions where there is considerable movement between a relatively small number of organizations, and opportunities for potential applicants to discover what work in other organizations will be like. An example of such a profession would be academics in higher education (HE). In this profession there are structural mechanisms, e.g. conferences, external examining, visiting lectureships, that give potential applicants opportunities to find out more about places that they might apply to. Moreover, such is the circulation of academic staff that potential applicants often know people who have intimate knowledge of the recruiting establishment whom they can contact to find out more. Also, in the HE environment, the type of work is likely to be broadly similar in most institutions. As a result, the decision to apply elsewhere is likely to be influenced to a greater extent by PO fit than job moves where roles might be expected to be very different. These observations suggest that it is possible to refine ASA theory in the following ways. The present study demonstrated that it was not enough to say that “similarity leads to attraction”. The discussion highlighted two reasons for this phenomenon. First, these applicants were concerned with vocational choice, not organizational choice. People chose to apply to these organizations because the employers had vacancies in particular types of work; i.e. there was no firm-specific element to applicants’ choosing. Second, the applicants in this study were remote with little opportunity to assess the values of the recruiting organization. Drawing from the interpersonal attraction literature, these people did not have proximity, exposure and familiarity to the recruiting organization, which are viewed as necessary precursors for the similarity leads to attraction effect. From these two issues, two attraction propositions can be advanced: (1) Applicants must have made, have embarked on, and be settled upon their choice of vocation before an organization-specific attraction effect can appear. (2) Applicants must have proximity, exposure, and familiarity with the recruiting organization’s values for value congruence to influence attraction.

Managerial implications The present study has remained neutral on Schneider’s fears (Schneider, 1987; Schneider et al., 1995, 1998, 2000) that greater homogeneity of workforces leads to organizational dysfunction. It has simply explored whether or not the attraction phase of the proposed ASA cycle contributes towards organizational homogeneity. Consequently, no views can be offered on whether or not it is right for organizations to recruit and select for fit. However, some thoughts can be offered to those organizations that have decided that they do, or do not, wish to do so. One managerial implication that emerges from this study is the ineffectiveness of recruitment interventions to alter the value profile of applicants. The recruiting companies had expressed a desire to communicate their corporate values during the attraction phase. Accordingly, they designed their brochure to feature the values they wished to promote – achievement, excellence, growth, employee-friendly personnel policies, community involvement – and reiterated these values in their presentations at universities. Despite these efforts, the profile of applicants was unaltered and did not differ from the general population from which applicants came. There are many ways to interpret this finding (perhaps the companies were ineffective in their efforts, perhaps applicants can see through the “hype” of brochures and presentations, or perhaps applicants are “blind” to the messages), but the most likely explanation seems to be that potential applicants knew little of the recruiting organization and applied to it because it offered the type of work they desired. If so, the natural managerial implication is that efforts by organizations to attract applicants with particular values are likely to be fruitless unless considerable effort – i.e. much more than demonstrated by the companies in the present study – is put into the exercise. References Ambrosini, V. and Bowman, C. (2002), “Mapping successful organizational routines”, in Huff, A.S. and Jenkins, M. (Eds), Mapping Strategic Knowledge, Sage, London, pp. 19-45. Arnold, J. (1985), “Tales of the unexpected: surprises experienced by graduates in the early months of employment”, British Journal of Guidance and Counselling, Vol. 13, pp. 308-19. Barber, A.E. and Wesson, M.J. (1998), “Using verbal protocol analysis to assess the construct validity of an empirical measure: an examination of the OCP”, in Wagner, J.A. III (Ed.), Advances in Qualitative Organization Research, Vol. 1, JAI Press, Greenwich, CT, pp. 67-104. Block, J. (1978), The Q-Sort Method in Personality Assessment and Psychiatric Research, Consulting Psychologists Press, Palo Alto, CA. Bowers, K.S. (1973), “Situationism in psychology: an analysis and a critique”, Psychological Review, Vol. 80, pp. 307-36. Bretz, R.D., Ash, R.A. and Dreher, G.F. (1989), “Do people make the place? An examination of the attraction-selection-attrition hypothesis”, Personnel Psychology, Vol. 42, pp. 561-81. Cable, D.M. and DeRue, D.S. (2002), “The convergent and discriminant validity of subjective fit perceptions”, Journal of Applied Psychology, Vol. 87, pp. 875-84. Cable, D.M. and Judge, T.A. (1994), “Pay preferences and job search decisions: a person-organization fit perspective”, Personnel Psychology, Vol. 47, pp. 317-48.

Attracting for values

145

JMP 22,2

146

Cable, D.M. and Judge, T.A. (1996), “Person-organization fit, job choice decisions, and organizational entry”, Organizational Behavior and Human Decision Processes, Vol. 67, pp. 294-311. Cable, D.M. and Judge, T.A. (1997), “Interviewers’ perceptions of person-organization fit and organizational selection decisions”, Journal of Applied Psychology, Vol. 82, pp. 546-61. Chatman, J. (1991), “Matching people and organizations: selection and socialization in public accounting firms”, Administrative Science Quarterly, Vol. 36, pp. 459-84. Denton, D.W. (1999), “The attraction-selection-attrition model of organizational behavior and the homogeneity of managerial personality”, Current Research in Social Psychology, Vol. 4, pp. 146-59. Dineen, B.R., Ash, S.R. and Noe, R.A. (2002), “A web of applicant attraction: person-organization fit in the context of web-based recruitment”, Journal of Applied Psychology, Vol. 87, pp. 723-34. Festinger, L., Schachter, S. and Back, K. (1950), Social Pressures in Informal Groups: A Study of Human Factors in Housing, Stanford University Press, Stanford, CA. George, D. and Mallery, A. (1995), SPSS/PC þ Step by Step: A Simple Guide and Reference, Wadsworth, Belmont, CA. Gross, R.D. (1992), Psychology: The Science of Mind and Behaviour, 2nd ed., Hodder & Stoughton, London. Hambrick, D.C. and Mason, P.A. (1984), “Upper echelons: the organization as a reflection of its top managers”, Academy of Management Review, Vol. 9, pp. 193-206. Harris, L.M. (2000), “Issues of fairness in recruitment processes: a case study of local government practice”, Local Government Studies, Vol. 26, pp. 31-46. Herriot, P. (1984), Down from the Ivory Tower: Graduates and their Jobs, Wiley, Chichester. Holland, J.L. (1976), “Vocational preferences”, in Dunnette, M.D. (Ed.), Handbook of Industrial and Organizational Psychology, Rand McNally, Chicago, IL, pp. 521-70. Jackson, S.E., Brett, J.F., Sessa, V.I., Cooper, D.M., Julin, J.A. and Peyronnin, K. (1991), “Some differences make a difference: individual dissimilarity and group heterogeneity as correlates of recruitment, promotions and turnover”, Journal of Applied Psychology, Vol. 76, pp. 675-89. Jordan, M., Herriot, P. and Chalmers, C. (1991), “Testing Schneider’s ASA theory”, Applied Psychology: An International Review, Vol. 40, pp. 47-54. Judge, T.A. and Bretz, R.D. (1992), “Effects of work values on job choice decisions”, Journal of Applied Psychology, Vol. 77, pp. 261-71. Judge, T.A. and Cable, D.M. (1997), “Applicant personality, organizational culture, and organization attraction”, Personnel Psychology, Vol. 50, pp. 359-94. Keenan, A. (1997), “Selection for potential: the case of graduate recruitment”, in Anderson, N. and Herriot, P. (Eds), International Handbook of Selection and Assessment, Wiley, Chichester, pp. 507-23. Keenan, A. and Newton, T.J. (1984), “Some antecedents and consequences of frustration in organizations”, Journal of Occupational Psychology, Vol. 57, pp. 57-66. Keenan, A. and Newton, T.J. (1986), “Work aspirations and experiences of young graduate engineers”, Journal of Management Studies, Vol. 23, pp. 224-37. Kerlinger, F.N. (1986), Foundations of Behavioral Research, Holt, Rinehart and Winston, Orlando, FL.

Kristof, A.L. (1996), “Person-organization fit: an integrative review of its conceptualizations, measurement, and implications”, Personnel Psychology, Vol. 49, pp. 1-19. Kristof-Brown, A.L. (1997), “Repertory grid technique: an alternative method for assessing fit in organizational settings”, paper presented at the Academy of Management Annual Meeting, Boston, MA, August. Kristof-Brown, A.L. (2000), “Perceived applicant fit: distinguishing between recruiters’ perceptions of person-job and person-organization fit”, Personnel Psychology, Vol. 53, pp. 643-71. Kristof-Brown, A.L., Zimmerman, R.D. and Johnson, E.C. (2005), “Consequences of individuals’ fit at work: a meta-analysis of person-job, person-organization, person-group and person-supervisor fit”, Personnel Psychology, Vol. 58, pp. 281-342. Mabey, C. (1986), Graduates into Industry, Gower, Aldershot. Neiner, A.G. and Owens, W.A. (1985), “Using biodata to predict job choice among college graduates”, Journal of Applied Psychology, Vol. 70, pp. 127-36. Nicholson, N. and Arnold, J. (1991), “From expectation to experience: graduates entering a large corporation”, Journal of Organizational Behavior, Vol. 12, pp. 413-29. Nunnally, J.C. (1978), Psychometric Theory, 2nd ed., McGraw-Hill, New York, NY. O’Reilly, C.A., Chatman, J. and Caldwell, D.F. (1991), “People and organizational culture: a profile comparison approach to assessing person-organization fit”, Academy of Management Journal, Vol. 34, pp. 487-516. Owens, W.A. and Schoenfeldt, L.F. (1979), “Toward a classification of persons”, Journal of Applied Psychology Monograph, Vol. 65, pp. 569-607. Rynes, S.L. and Gerhart, B. (1990), “Interviewer assessments of applicant ‘fit’: an exploratory investigation”, Personnel Psychology, Vol. 43, pp. 13-35. Rynes, S.L., Orlitzky, M.O. and Bretz, R.D. (1997), “Experienced hiring versus college recruiting: practices and emerging trends”, Personnel Psychology, Vol. 50, pp. 309-39. Saegert, S.C., Swap, W. and Zajonc, R.B. (1973), “Exposure context and interpersonal attraction”, Journal of Personality and Social Psychology, Vol. 25, pp. 234-42. Schmitt, N. and Chan, D. (1998), Personnel Selection: A Theoretical Approach, Sage, Thousand Oaks, CA. Schneider, B. (1983a), “Interactional psychology and organizational behavior”, in Cummings, L.L. and Staw, B.M. (Eds), Research in Organizational Behavior, Vol. 5, JAI Press, Greenwich, CT, pp. 1-31. Schneider, B. (1983b), “An interactional perspective on organizational effectiveness”, in Cameron, K.S. and Whetten, D.S. (Eds), Organizational Effectiveness: A Comparison of Multiple Models, Academic Press, New York, NY, pp. 27-54. Schneider, B. (1985), “Organizational behavior”, Annual Review of Psychology, Vol. 36, pp. 573-611. Schneider, B. (1987), “The people make the place”, Personnel Psychology, Vol. 40, pp. 437-53. Schneider, B., Goldstein, H.W. and Smith, D.B. (1995), “The ASA framework: an update”, Personnel Psychology, Vol. 48, pp. 747-73. Schneider, B., Smith, D.B. and Goldstein, H.W. (2000), “Attraction-selection-attrition: toward a person-environment psychology of organizations”, in Walsh, B.W., Craik, K.H. and Price, R.H. (Eds), Person-Environment Psychology: New Directions and Perspectives, 2nd ed., Erlbaum, Mahwah, NJ, pp. 61-86.

Attracting for values

147

JMP 22,2

148

Schneider, B., Smith, D.B., Taylor, S. and Fleenor, J. (1998), “Personality and organizations: a test of the homogeneity of personality hypothesis”, Journal of Applied Psychology, Vol. 83, pp. 462-70. Schwab, D.P. (1980), “Construct validity in organizational behavior”, in Staw, B.M. and Cummings, L.L. (Eds), Research in Organizational Behavior, JAI, Greenwich, CT, pp. 3-43. Schwartz, S.H. (1992), “Universals in the content and structures of values: theoretical advances and empirical tests in 20 countries”, in Zanna, M.P. (Ed.), Advances in Experimental Social Psychology, Vol. 25, Academic Press, San Diego, CA, pp. 1-65. Schwartz, S.H. (1994), “Are there universals in the structure and contents of human values?”, Journal of Social Issues, Vol. 50, pp. 19-45. Stackman, R.W., Pinder, C.C. and Connor, P.E. (2000), “Values lost: redirecting research on values in the workplace”, in Ashkanasy, N.M., Wilderom, C.P.M. and Peterson, M.F. (Eds), Handbook of Organizational Culture and Climate, Sage, Thousand Oaks, CA, pp. 37-54. Tom, V.R. (1971), “The role of personality and organizational images in the recruiting process”, Organizational Behavior and Human Performance, Vol. 6, pp. 573-92. Turban, D.B. (2001), “Organizational attractiveness as an employer on college campuses: an examination of the applicant population”, Journal of Vocational Behavior, Vol. 58, pp. 293-312. Turban, D.B. and Keon, T.L. (1993), “Organizational attractiveness: an interactionalist perspective”, Journal of Applied Psychology, Vol. 78, pp. 184-93. Van Vianen, A.E.M. (2000), “Person-organization fit: the match between newcomers’ and recruiters’ preferences for organizational cultures”, Personnel Psychology, Vol. 53, pp. 1-32.

Further reading Boudreau, J.W. and Rynes, S.L. (1985), “Role of recruitment in staffing utility analysis”, Journal of Applied Psychology, Vol. 70, pp. 354-66. Byrne, D. and Nelson, D. (1965), “Attraction as a linear function of proportion of positive reinforcements”, Journal of Personality and Social Psychology, Vol. 1, pp. 659-63. Cohen, J. (1988), Statistical Power Analysis for the Behavioral Sciences, 2nd ed., Lawrence Erlbaum Associates, Hillsdale, NJ. Edwards, J.R. (1993), “Problems with the use of profile similarity indices in the study of congruence in organizational research”, Personnel Psychology, Vol. 46, pp. 641-65. Edwards, J.R. (1994), “The study of congruence in organizational behavior: critique and a proposed alternative”, Organizational Behavior and Human Decision Processes, Vol. 58, pp. 51-100. Hatfield, E., Traupmann, J. and Walster, G.W. (1978), “Equity and extramarital sexuality”, Archives of Sexual Behavior, Vol. 7, pp. 127-42. Meglino, B.M. and Ravlin, E.C. (1998), “Individual values in organizations: concepts, controversies, and research”, Journal of Management, Vol. 24, pp. 351-89. Myers, I.B. and McCaulley, M.H. (1985), Manual: A Guide to the Development and Use of the Myers-Briggs Type Indicator, Consulting Psychologists Press, Palo Alto, CA. Newcomb, T.M. (1961), The Acquaintanceship Process, Holt, Rinehart & Winston, New York, NY. Sheridan, J.E. (1992), “Organizational culture and employee retention”, Academy of Management Journal, Vol. 35, pp. 1036-56.

Attracting for values

Appendix Original OCP item

Reworked item

Being innovative Being quick to take advantage of opportunities A willingness to experiment Respect for the individual’s rights Action-oriented Developing friends at work Working in collaboration with others Working long hours Risk taking Autonomy Paying attention to detail Taking initiative Being demanding Offers praise for good performance Fitting in An emphasis on quality Being results-oriented Being precise Fairness Being people-oriented Opportunities for professional growth Being highly organized

Staff are continually being innovative People are quick to take advantage of opportunities Staff experiment with new ways of doing things 149 People have respect for the rights of others Employees are very busy at work People develop friendships at work People work in collaboration with others People work long hours Employees take risks Staff have a lot of autonomy Staff pay attention to detail People act on their own initiative Staff have considerable demands made of them Employees are given praise for good performance People try to fit in People make quality a priority People focus on profits Staff are precise Being fair is a priority for people in the organization Managers are concerned that people are treated well Table AI. There are opportunities for growth and development Conversion of OCP items, Staff approach their work in a very organized prompts and anchored manner scales People are competitive

Being competitive

To capture environmental vales (e.g. of the company or department), the following prompt was adopted: “How characteristic of your organization’s [changed to department’s when PG fit was being captured] culture are the following items?” The items were scored on the following anchored seven-point Likert-scale: (1) “very uncharacteristic”, (2) “uncharacteristic”, (3) “neutral”, (4) “sometimes found”, (5) “characteristic”, (6) “very characteristic”, and (7) “a defining characteristic”. To capture individuals’ values (e.g. applicants, department members), the following prompt was used: “How desirable is it for each of the following items to be a part of the organization you work for?” The items were scored on following anchored seven-point Likert-scale: (1) “very undesirable”, (2) “undesirable”, (3) “neutral”, (4) “desirable”, (5) “very desirable”, (6) “important”, and (7) “essential”.

About the author Jon Billsberry is Senior Lecturer in Organizational Behaviour at the Open University Business School. He holds degrees from the universities of Manchester, Birmingham and Nottingham. He is currently Chair of the Organizational Psychology division of the British Academy of Management and Secretary of the Management Education and Development division of the Academy of Management. His research interests are in the fields of person-organization fit, recruitment and selection, leadership, cinematic portrayals of work and working life, and the psychology of the penalty shoot-out. Jon Billsberry can be contacted at: [email protected] To purchase reprints of this article please e-mail: [email protected] Or visit our web site for further details: www.emeraldinsight.com/reprints

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

JMP 22,2

Using assessment centre performance to predict subjective person-organisation (P-O) fit

150

A longitudinal study of graduates Thomas N. Garavan Department of Personnel and Employment Relations, Kemmy Business School, University of Limerick, Limerick, Ireland Abstract Purpose – This research aims to contribute to the discussion of P-O fit by examining the potential of assessment centre performance to predict graduate P-O fit over time. Does assessment centre performance provide predictive value over and above that provided by personality, work experience, level of degree, socialisation tactics, and quality of training and development? Design/methodology/approach – The study utilises a panel design and surveyed graduates at three points over a six-year period. The sample size used for analysis was 137. Findings – The findings revealed that assessment centre performance predicted P-O fit, even after controlling for personality, level of initial degree, work experience, socialisation tactics and quality of training and development. The findings indicated that assessment centre performance had predictive value over the six-year period of the study. Research limitations/implications – The study indicates that the assessment centre has long-term value as a selection tool in the context of predicting P-O fit. The implications for future research include the extent to which this finding will hold for objective as well as subjective P-O fit. It is also appropriate to investigate the link between assessment centre performance and P-O fit, based on the perceptions of others such as boss and subordinates. Future research should examine these relationships on a different sample of employees. Originality/value – Examines the potential of assessment centre performance to predict graduate P-O fit over time. Keywords Graduates, Assessment centres Paper type Research paper

Journal of Managerial Psychology Vol. 22 No. 2, 2007 pp. 150-167 q Emerald Group Publishing Limited 0268-3946 DOI 10.1108/02683940710726410

Introduction Graduates are costly to hire and develop, however their skills are increasingly demanded by organisations (Baruch, 1999; Viney et al., 1997). They are hired to provide organisations with a pool of potential managers and enhance succession possibilities. The emergence of talent management has led organisations to hire what they considered future stars. Graduate stars are frequently labelled “high flier”, “fast track” or “high potential”, and represent a special category of human resources, capable of making a unique contribution to the strategic success of the organisation (MacKenzie-Davey and Arnold, 1995, 2000). Organisations invest significant resources in development and career management processes for high potential graduates. Graduates expect early responsibility for complex tasks, high visibility work opportunities, quality development opportunities and fast career advancement (Coupland, 2001; Lievans et al., 2001). Organisations expect high potential graduates to

achieve superior performance and to demonstrate flexibility, loyalty and commitment to the organisation (Connor et al., 1990; Hendry et al., 1995). They are expected to integrate quickly into the organisation’s structure and culture (Clarke et al., 1994; Fournier, 1998). Organisations utilise a number of HR tools to achieve P-O fit. These include: the use of sophisticated selection processes (Arnold and McKenzie-Davey, 1999); high intensity organisational socialisation processes and tactics (Baruch, 1999; Autry and Wheeler, 2005); leading edge training and development processes (Cheng and Ho, 2001) and structured career planning and development processes (Graham and McKenzie, 1995). Assessment centre (AC) methodologies are used to hire high-potential graduates and identify those who will best fit the organisation (Arnold, 1989). Assessment centres have a number of important advantages in this context. They enable the measurement of a range of competencies and skills and they particularly capture the strength of management and interpersonal competencies associated with managerial success, job satisfaction and person-organisational fit (Boyatzis et al., 2002). Assessment centre scores have demonstrated predictive validity (Tziner, 2002), however there is some doubt concerning their predictive power with the elapse of time. Tziner et al. (1993) found, for example, that the predictive validity of assessment centre scores decreases with time. This study has two unique features. It includes subjective P-O fit as the criterion and investigates its predictive power over a period of six years. The study of P-O fit is relevant in the context of graduates. The subjective assessment that graduates make concerning the match between their personal characteristics and those of the organisation are likely to determine their decision to remain in the organisation. The theory of reasoned action (Azjen, 2001) would suggest that a graduate’s beliefs about the organisation will be manifest in either positive or negative attitudes about the organisation and will be influenced by subjective criteria. Where high-potential graduates perceive that there is a mismatch they are likely to engage in job-search activities. It is well established that the subjective perceptions of P-O fit will influence whether a graduate stays in the organisation (Judge and Cable, 1997; Adkins et al., 1994). The use of P-O fit as a criterion variable is in itself important. There is evidence in the context of graduates that perceptions of P-O fit are important in determining whether the graduate stays, performs effectively and achieves career success (Bretz and Judge, 1994). The extent of P-O fit may also explain the extent of career progression (Bretz et al., 1993). If graduates are to realise their potential and participate more fully in the organisation, how they perceive the extent of fit becomes important. This study therefore seeks to address the following questions: do assessment centre scores have value in predicting the P-O fir of high potential graduates over time? Theory and hypothesis development Assessment centres are widely used for graduate selection and they show considerable success in predicting a range of describable outcomes such as high performance, career satisfaction and advancement (Thornton and Cleveland, 1990; Tziner et al., 2005). They possess a number of important characteristics that make them appropriate to use for the selection of graduates. They are standardised. They focus on behavioural and performance-based exercises and measure multiple skill dimensions using multiple

Assessment centre performance 151

JMP 22,2

152

exercises. They use multiple assessors to measure skill dimensions (Gaugler et al., 1987). Hsin-Chih (2006) suggests that assessment centres have the capacity to measure unique characteristics of human capital and match them to organisational strategy and culture. This occurs because the AC development process is rooted in organisational processes such as competency modelling and job analysis. The behaviours typically assessed in an assessment centre are closely associated to real job performance and organisation culture requirements (Thornton et al., 1997). Organisational competency models reflect the culture and strategic priorities of the organisation (Burnett and Dutsch, 2006). A competency-based model of sustainable advantage designed by Lado et al. (1992) placed emphasis on managerial competencies and their impact on the culture and strategic success of the organisation. Organisational competency models are built on particular values and articulate a profile of a successful manager (LeDeist et al., 2005). Howard (1997) has suggested that where the AC incorporates a broad range of exercises that capture a multiplicity of performance and organisational dimensions, the potential of the AC to predict P-O fit should be enhanced. Assessment centres have potentially particular predictive value for a graduate population. The assessment process may itself project positive perceptions concerning the professionalism of the organisation and how progressive it is. Furthermore, graduates are likely to use criteria related to characteristics of the organisation when deciding to join it. Graduates are likely, on the point of entry to the organisation, to make attributions concerning the favourability of the organisation from career and work perspectives (Mayrhofer et al., 2005) and they will place particular value on how they are likely to fit in. High potential graduates are likely to be exposed to various practices that are designed to enhance P-O fit. These processes may change the graduate’s initial perceptions of the organisation and it is possible that the graduate’s personal vales may change as a result of exposure to organisational processes. Various studies (Ryan and Ployhart, 2000; Anderson et al., 2001) highlight the importance of focusing on the graduate’s future fit with the organisation on the grounds that it is a describable outcome for organisations. Selection processes in organisations are very much concerned with assessing the graduate’s future fit with the organisation. I predict that: H1. AC performance will positively predict P-O fit over time. Impact of personality and level of initial degree The evidence suggests some relationships between personality and P-O fit. Judge et al. (1995) suggest that graduates who are highly conscientious should be more satisfied with their work. They will as a result be rewarded through greater development opportunities, praise and positive feedback, which will in turn lead to greater satisfaction, organisational commitment and P-O fit. Tokar and Subich (1997) found a negative relationship between emotional immaturity and satisfaction. Graduates who score high on extroversion are likely to experience higher levels of job satisfaction especially where the work involves a significant number of interpersonal interactions. Emotionally mature graduates are likely to respond effectively to feedback, stretch assignments, setbacks and failure, and the challenges of dealing with complex interpersonal situations (Hough, 1997). In contrast, emotionally unstable graduates are more likely to experience performance problems and a lack of promotion opportunities

(Salgado, 1997). Personality is also likely to be an important determinant of interpersonal skills, leadership capability and potential to work as part of a team. Personality characteristics are likely to be important on explaining the extent of P-O fit. Siebert and Kraimer (2001) suggest that personality characteristics such as adaptability, flexibility, emotional maturity and need for achievement may impact on a graduate’s ability to adjust his or her behaviour to match the requirements of the organisation. Therefore graduates who scores highly on these characteristics are likely to have stronger subjective P-O fit. It is therefore relevant to explore whether the AC performance is predictive of P-O fit beyond that provided by personality measures. The level of the initial degree is also important. Some commentators argue that the quality of the degree is itself sufficient because it indicates a certain level of discipline or content knowledge (Bartels et al., 2000; Cohen, 1984). The assessment centre process however has potential to demonstrate skills in areas such as problem solving, team working, oral and written communication and strategic analysis over an above that potentially provided by measures of personality and level of degree. I predict that: H2. AC performance will explain additional variance in P-O fit beyond that explained by personality and level of initial degree.

Organisation socialisation tactics, quality of development opportunities and P-O fit Louis (1980, 1990) suggests that when graduates join an organisation they experience shock. They have to cope with differences between expectations and reality, they have to learn new routines and their precise contribution to the organisation is often poorly defined. Organisations use a range of socialisation tactics designed to reduce these ambiguities and provide a common set of learning experiences and structured development events to facilitate adjustment to their new environment (Cooper-Thomas et al., 2004). Kim et al. (2005) indicate that structured and highly institutionalised socialisation processes provide the graduate with explicit information about the organisation. They also provide positive social support and exposure to experienced organisational members (Chalman, 1991; Westerman and Vanka, 2006). The socialisation process is usually aligned with structured training and development processes such as mentoring, planned work experience and formal programmes of training and development. The developmental resources provided to the graduate are particularly salient in explaining P-O fit. Fast-track graduate development programmes incorporate activities such as skill-building seminars, workshops, special assignments, team projects, mentoring, coaching and multi-source feedback processes. Greenhaus (1988) suggests that the quality of training and developmental resources and opportunities to participate in development are important in explaining P-O fit. Sturges et al. (2002) suggests that career management and development activities are associated with increased organisational commitment, job satisfaction and P-O fit. I predict that: H3. AC Performance will explain additional variance in P-O fit, beyond that explained by socialisation tactics and quality of training and development activities.

Assessment centre performance 153

JMP 22,2

Method Participants and response rates Participants were drawn from four Irish organisations employing 500 þ employees. The study organisations recruited large numbers of graduates on an annual basis. A description of the four organisations is provided below:

154

Organisation 1 was a publicly-owned financial services organisation which employed graduates in a wide range of managerial and financial specialist roles. Graduates worked in a range of divisions and specialisms. Organisation 2 was a publicly-owned agri-business organisation which recruited graduates into a range of functional roles. Graduates worked in areas such as finance, marketing, logistics and human resources. Organisation 3 was a multinational electronics company that manufactured electronic components for the car industry. This organisation recruited graduates with engineering and technical backgrounds. They worked in areas such as production control, quality management, engineering support and logistics. Organisation 4 was a computer manufacturer that recruited graduates for a range of financial and generalist manager type roles. Graduates worked in a very broad range of functions such as marketing and sales, engineering support, quality control, information support and finance.

Participants were initially hired to undertake a one-year structured fast-track graduate training and development programme. A standardised assessment centre process was used to select graduates for each organisation. Participants were educated in a variety of disciplines including: business, agri-business, humanities, engineering, and sciences. All graduates selected by the organisations were included in the study (n ¼ 290). The study was conducted over a six-year period, utilising three measurement points; time 1 (t1) three months after joining the programme; time 2 (t2) two years after joining the organisation; and time 3 (t3) six years after joining the organisation. The response rates for the study are high. A total of 240 (83 per cent) usable questionnaires were returned at t1, 160 (55 per cent) usable questionnaires were returned at t2 and 137 (47 per cent) usable questionnaires were returned at t3. I included in the analysis only those respondents for whom I received three completed questionnaires. Graduates were aged between 21-27 at t1. The mean age was 23.8 years. A total of 57 per cent of respondents were male. A total of 45 per cent of graduates had a first level degree and the remainder had a higher second. Thirdly, 1 percent of graduates had studied business or economics; 28 per cent engineering; 27 per cent agri-business and 10 per cent humanities/arts. A total of 57 per cent of graduates had no prior working experience prior to joining the graduate development programmes. Assessment centre process and exercises All study respondents participated in a one-day assessment centre (AC). AC sessions were undertaken within an eight-week period. Graduates were informed that the AC had multiple purposes including, helping the organisation to evaluate the suitability of the graduate and providing graduates with feedback on strengths and development areas. Each AC was managed by a lead assessor and six trained assessors. A total of ten graduates participated on each AC. Feedback to participants was provided by the trained assessors. Graduates participated in multiple exercises during the AC process. For all AC sessions, graduates participated in an in-basked simulation, leaderless group discussion, a case study, an interview simulation and an oral presentation. These exercises were chosen because they were the most appropriate way of capturing

the dimensions described in Table I. Boyatzis et al. (2002) suggests that this combination of exercises is appropriate to assess team-based management and interpersonal skills as well as individual problem-solving and decision-making skills, and written communication skills. Detailed job analysis was conducted to determine the assessment dimensions to be included in the assessment centre process. Successful graduates within the four study organisations were requested to provide performance incidents. These incidents were analysed by the researcher and personnel specialists within the organisations to determine the characteristics of successful graduates. Consensus was achieved on the priority qualities required for successful performance. Seven dimensions emerged such as problem solving and decision-making, planning and organising, interpersonal skills, and teamwork Based on these dimensions a number of situational exercises were designed. Mental ability test were not included as part of the assessment centre process. The exercises included in the AC are described as follows: . In-basket. The in-basket exercise involved one of two scenarios assigned to different AC sessions. In the first, graduates were placed in the role of a new regional manager of a major bank. The graduate had two hours to respond to thirty memos and letters waiting in the person’s in-basket. The memos and letters dealt with problems pertaining to customers, sales performance and human resources. In the second scenario, graduates were placed in the role of a production manager of a manufacturing company. In both scenarios, graduates had to prioritise their in-basket and prepare a written report that was typed and emailed to the lead assessor for subsequent distribution to assessors. Analysis revealed no significant performance differences between graduates engaged in the two in-basket scenarios and hence, analysis of the scenarios was combined. . Leaderless group discussion (LGD). The LGD involved placing graduates randomly into four or five person groups. Each group was given one of two unstructured Identifying, analysing and solving problems, including the willingness to make timely decisions Planning, organising and controlling the work of oneself and others through such means as effective prioritisation and delegation Interpersonal skills (leaderless group discussion, The ability to communicate one-on-one and in interview simulation) small groups, by showing persuasiveness coupled with sensitivity and ability to listen Leadership skills (leaderless group discussion) Demonstrating self-confidence and the ability to influence others by helping them to deal with uncertainty and change Teamwork skills (leaderless group discussion, Demonstrating a willingness to cooperate with interview simulation) others in a group setting and work toward attaining shared group goals Written communication (in basket, case scenario) Demonstrating appropriate content, organisation, and style in one’s written work Presentational skills (oral presentation) Conducting presentations in such a manner as to generate interest and enthusiasm, while being organised, structured and including convincing and informative arguments

Assessment centre performance 155

Problem solving and decision making (in basket, leaderless group discussion, case scenario) Planning and organising (in basket, oral presentations)

Table I. Definitions of graduate assessment centre dimensions and associated exercises

JMP 22,2

156

.

.

.

problems that were randomly assigned to different AC sessions. For each scenario, graduates in the group were allowed 15-minutes to read a two page summary of the problem and write down issues and potential solutions. They were brought together for 40 minutes to discuss the problem as a group. They were also instructed that at the end of the 40 minutes, they would be given 10 minutes to present their analysis of the problem. Each group was observed by two assessors. Case scenarios. Graduates analysed one of two case scenarios, each dealing with management issues in an actual organisation. The cases were randomly assigned to the various AC sessions. Graduates were given a total of 65-minutes to read over the case and provide a written analysis. All responses were typed and e-mailed to the principle investigator for subsequent distribution to assessors. There is no significant performance differences between the two case scenarios, therefore analysis of the cases were combined. Interview simulation. Graduates participated in a one-to-one interview with a trained interviewer. Each interview lasted between 30 and 40 minutes. The interviewer asked a series of structured questions focusing on the competencies identified by the participating organisations. Interviewers were trained to probe for behavioural specifics. Each interview was observed by two assessors. Oral presentation. The oral presentation involved a ten-minute presentation from among several topics. Each graduate was given 20 minutes to prepare. Each presentation was observed by two observers. Table I describes the main dimensions and associated exercises.

The assessment process and training of assessors Six teams of six assessors – three senior managers and three psychologists – each conducted the observation and evaluation process on groups of eight assessees. Tziner et al. (1993) reported that a combination of psychologists and managers as assessors provided more valid outcomes than those employing managers alone. Each assessor observed the behaviour of each group situational exercise and studied the performance of each assessee in each individual exercise. Each assessor independently rated each assessee in the dimensions included in the study (Table I) after all the exercises were completed. Assessors used five-point behavioural scales with multiple anchors at the low, mid-point and high ends of the scales. These anchors were determined after the assessors came to agreement regarding commonly observed behaviours illustrating low, mid-point and high ends of respective scales. On the day following the assessment centre, the assessors convened to discuss discrepancies in the ratings given to each assess and to form a consensus on a single rating on each of the dimensions for each of the candidates that they observed and evaluated. They also determined an overall assessment rating representing the likelihood of a candidate performing successfully in the organisation. The assessors were trained to evaluate the participant’s performance in three four-hour training sessions. Assessors received training on different examples of good and poor performance in the assessment centre. Samples of videotaped leaderless group discussions were shown during the training and assessors viewed actual samples of in-baskets. A key focus of the training was to ensure that assessors maintained a common frame of reference when they approached the rating task.

Measures Socialisation tactics. I used Jones’s (1986) 26-item scale of institutionalised socialisation tactics, to which graduates were asked to report the degree to which they experienced different socialisation tactics since they started to work in the organisation. Example items include: . “I have been through a set of training experiences which are specifically designed to give graduates a thorough knowledge of job related skills”. . “Almost all of my colleagues have been supportive of me personally”. This scale was included in the three questionnaires. The internal consistency reliability estimates for the scale were t1 0.82, t2 0.84 and t3 0.81. Perceptions of quality of training and development activities. I developed a scale that asked respondents to indicate the quality of training and development received. I included six items related to training and six items related to development. The items related to training included quality of technical skills training, coaching processes, interpersonal skills training and computer skills training. The items on development included the quality of work experience, quality of project assignments, mentoring, and career planning activities, formal development programmes and multi-source feedback processes. The internal consistency reliability estimates for the scale were t1 0.78, t2 0.81 and t3 0.77. Subjective P-O fit. Graduates reported their subjective P-O fit by responding to a five-item scale (Judge and Cable, 1997). Sample items include: . “My personal values match my organisation’s values and culture”. . “My organisation’s values and culture provide a good fit with the things that I value in life”. Responses for all of the scale items were on a five-point scale with anchors ranging from 1 – strongly agree to 5 – strongly disagree. All items were reverse scored in order to have higher scores indicate greater P-O fit. The internal consistency reliability estimates for the scale were t1 0.86, t2 0.81 and t3 0.83. Control variables. I included a personality profile in the AC to measure personality. The 16PF (Cattell) measures five global personality factors: extraversion, anxiety, self-control, independence, and tough mindedness. The resulting alpha reliabilities were 0.88, 0.87, 0.87, 0.85, and 0.82 respectively. I also included work experience as an additional control variable. Experience is important in predicting performance in a variety of organisational settings (McDaniel et al., 1988). The average reported level of experience at t1 was 1.37 years. This includes part time work. Results Means, standard deviations and correlations among variables are shown in Tables II to IV for the three measurement points. The correlations highlight a number of key findings. Personality, work experience, level of degree and AC performance scores are relatively independent of each other and therefore measured distinct constructs. This trend holds for the three sets of correlations at t1, t2 and t3. At t1, work experience is significantly correlated with gender (r ¼ 0:48 p , 0.01), socialisation tactics (r ¼ 0:24 p , 0.05), quality of training and development (r ¼ 0:31 p , 0.05) and P-O fit r ¼ 0:23 p , 0.05). It is also correlated with the level of initial degree (r ¼ 0:21 p , 0.05). At t1 gender is

Assessment centre performance 157

0.45 0.50 0.61 0.72 0.51 0.79 0.87 0.50 0.96 0.89 1.10 0.91

SD 0.42 * * 0.07 0.11 0.16 0.08 0.10 0.21 * 0.14 0.24 * 0.31 * 0.23 * 0.21 * 0.11 0.14 0.16 0.19 0.08 0.14 0.17 * 0.09 0.08

Work exp. Gender

(0.88) 0.31 * 0.37 * 0.28 * 0.27 * 0.10 0.11 0.07 0.02 0.10

Extra

(0.870 0.28 * 0.31 * 0.24 * 0.07 0.11 0.02 0.03 0.11

Anxiety

(0.87) 0.27 * 0.27 * 0.02 0.05 0.04 0.05 0.12

Self

(0.85) 0.29 * 0.04 0.06 0.07 0.09 0.01

(0.82) 0.04 0.07 0.08 0.03 0.04

0.14 * 0.07 0.04 0.09

Tough Independence Degree

SOC

QTD

P-O

0.16 * (0.82) 0.37 * * 0.49 * * (0.78) 0.41 * * 0.57 * * 0.31 * (0.80)

AC

Notes: Men are coded as 0 and women were coded as 1; Degree level: 1 ¼ 2:1, 2 ¼ first; Reliabilities are in parentheses; xp , 0.10; *p , 0.05; * *p , 0.01

1.35 0.42 3.75 3.86 3.45 3.95 4.11 0.62 3.75 4.12 3.82 3.67

Work experience Gender Extraversion Anxiety Self-control Tough mindedness Independence Degree level Overall AC score Socialisation tactics Quality of T&D P-O fit

Table II. Descriptive statistics and correlations (n ¼ 137) t1

M

158

Variables

JMP 22,2

3.35 0.42 3.75 3.86 3.45 3.95 4.11 0.62 3.75 3.86 3.41 3.41

Work experience Gender Extraversion Anxiety Self-control Tough mindedness Independence Degree level Overall AC score Socialisation tactics Quality of T&D P-O fit

0.45 0.5 0.61 0.72 0.51 0.79 0.87 0.50 0.96 0.81 1.20 0.89

SD – 0.09 0.06 0.07 0.08 0.10 0.10 0.11 0.13 0.21 * 0.37 * 0.28 * – 0.21 * 0.11 0.14 0.16 0.19 0.08 0.14 0.27 * 0.24 * 0.29 *

Work exp. Gender

(0.88) 0.31 * 0.37 * 0.28 * 0.27 * 0.10 0.11 0.09 0.10 0.10

Extra

(0.87) 0.28 * 0.31 * 0.24 * 0.07 0.11 0.07 0.05 0.11

Anxiety

(0.87) 0.27 * 0.27 * 0.02 0.05 0.11 0.11 0.11

Self

(0.85) 0.29 0.04 0.06 0.12 0.11 0.10 (0.82) 0.04 0.06 0.12 0.11 0.10

0.14 * 0.14 0.17 * 0.11

Tough Independence Degree

SOC

QTD

P-O

0.17 * (0.84) 0.24 * 0.47 * * (0.81) 0.27 * 0.39 * * 0.41 * * (0.81)

AC

Notes: Men are coded as 0 and women were coded as 1; Degree level: 1 ¼ 2:1, 2 ¼ first; Promotions are coded as follows: 1 ¼ 1, 2 ¼ 2, 3 ¼ 3 or more; Reliabilities are in parentheses; xp , 0.10; *p , 0.05; * *p , 0.01

M

Variables

Assessment centre performance 159

Table III. Descriptive statistics and correlations (n ¼ 137) t2

0.45 0.50 0.61 0.72 0.51 0.79 0.87 0.50 0.96 0.34 0.89 0.87

SD – 0.15 0.11 0.08 0.10 0.12 0.10 0.11 0.15 * 0.24 * 0.39 * 0.41 * * – 0.21 0.11 0.14 0.16 0.19 0.08 0.14 0.22 * 0.27 * 0.19 *

Work exp. Gender

(0.88) 0.31 * 0.37 * 0.28 * * 0.27 * 0.10 0.10 0.10 0.07 0.10

Extra

(0.87) 0.28 * 0.31 * 0.24 * 0.07 0.11 0.04 0.06 0.10

Anxiety

(0.87) 0.27 * 0.27 * 0.02 0.05 0.06 0.07 0.09

Self

(0.85) 0.29 * 0.04 0.06 0.08 0.10 0.09

(0.82) 0.04 0.07 0.09 0.10 0.11

0.14 * 0.08 0.06 0.03

Tough Independence Degree

SOC

QTD

P-O

0.14 * (0.81) 0.27 * 0.38 * (0.77) 0.24 * 0.40 * * 0.44 * (0.83)

AC

Notes: Men are coded as 0 and women were coded as 1; Degree level: 1 ¼ 2:1, 2 ¼ first; Reliabilities are in parentheses; xp , 0.10; *p , 0.05; * *p , 0.01; Promotions were coded as follows: 1 ¼ 1, 2 ¼ 2, 3 ¼ 3 or more promotions

6.35 0.42 3.75 3.86 3.45 3.96 4.12 0.62 3.75 3.41 3.87 3.79

Work experience Gender Extraversion Anxiety Self-control Tough mindedness Independence Degree level Overall AC score Socialisation tactics Quality of T&D P-O fit

Table IV. Descriptive statistics and correlations (n ¼ 137) t3

M

160

Variables

JMP 22,2

significantly correlated with extraversion (r ¼ 0:21 p , 0.05) and perception of socialisation tactics (r ¼ 0:17 p , 0.05). The overall assessment centre score is significantly correlated with perceptions of socialisation tactics (r ¼ 0:16 p , 0.05), perceptions of quality of training and development (r ¼ 0:37 p , 0.01) and person-organisation fit (r ¼ 0:41 p , 0.01). A strong significant correlation exists between perception of socialisation tactics and quality of training and development (r ¼ 0:49 p , 0.01) and P-O fit (r ¼ 0:57 p , 0.01). The mean score for socialisation tactics at t1 is high (4.12) which indicate that graduates experienced strong socialisation tactics. At t2 work experience is significantly correlated with perception of socialisation tactics (r ¼ 0:21 p , 0.05) perceptions of quality of training and development (r ¼ 0:37 p , 0.05) and perceptions of P-O fit (r ¼ 0:28 p , 0.05). Overall assessment centre score is significantly correlated with socialisation tactics (r ¼ 0:17 p , 0.05), quality of training and development (r ¼ 0:24 p , 0.05) and P-O fit (r ¼ 0:27 p , 0.05). The data at t2 also reveals significant correlations between socialisation tactics, quality of training and development (r ¼ 0:47 p , 0.01) and P-O fit (r ¼ 0:39 p , 0.01) and between quality of training and development and P-O fit (r ¼ 0:41 p , 0.01). At t3 the analysis shows relationships that are consistent with those revealed at t1 and t2. I tested the hypotheses using hierarchical regression analyses. The results are shown in Table V. Control variables were entered in the first step. They included work experience, gender, level of degree and personality characteristics. I ordered the entry of the predictor variables according to the establishment of the variables in the literature as a predictor of P-O fit. Then socialisation tactics and quality of training and development were entered following the control variables. I focused on overall AC performance in the regression analyses. There was a strong correlation between the various components of the AC, therefore it was appropriate to use the overall AC score. It was entered last in each of the analyses because of the relative lack of support in the literature for it as a predictor of P-O fit. A key question in this study was to determine

Predictor variables

Step 1

Work experience Gender Degree level Extraversion Anxiety Self-control Tough mindedness Independence Socialisation tactics Quality of training and development Overall AC score DR 2 Total R 2 DF

0.07 0.04 0.06 0.14 0.12 0.11 0.15 0.12

0.08

t1 Step 2

Step 3

Step 1

0.09 0.06 0.06 0.12 0.11 0.10 0.11 0.12 0.29 *

0.09 0.07 0.03 0.10 0.10 0.10 0.11 0.11 0.25 *

0.29 * 0.09 0.07 0.11 0.14 0.12 0.12 0.11

0.34 *

0.21 * 0.34 * 0.11 0.29 * 6.70 *

0.10 0.18 * 0.26

0.08

t2 Step 2

Step 3

Step 1

0.27 * 0.08 0.07 0.12 0.13 0.11 0.14 0.12 0.21 *

0.28 * 0.04 0.07 0.11 0.12 0.13 0.11 0.12 0.29 *

0.29 * 0.05 0.08 0.10 0.10 0.09 0.08 0.10

0.39 * *

0.37 * * 0.18 * 0.01 0.29 * 7.41 * *

0.20 0.28 * 0.95

0.10

Notes: Standardised regression coefficients are shown: *p , 0.05, * *p , 0.0

t3 Step 2

Step 3

0.31 * 0.04 0.09 0.09 0.11 0.12 0.11 0.07 0.14

0.32 0.05 0.07 0.11 0.12 0.11 0.10 0.10 0.16

0.28 *

0.36 * * 0.16 * 0.08 0.44 * * 6.95 * *

0.26 0.36 * * 0.87

Assessment centre performance 161

Table V. Hierarchical regression analysis predicting P-O fit, t1, t2, t3

JMP 22,2

162

whether AC performance would explain variance in P-O fit beyond that of the control variables, socialisation tactics and quality of training and development. The results for P-O fit are shown in Table V. The overall equation in the prediction of P-O fit, as revealed by the total R 2 values, was significant on the three regression steps for the three measurement points. The analyses suggest support for H1. The impact of the assessment centre score is most significant at t1, the DR 2 are smaller at t2 and t3. Socialisation tactics and quality of training and development are significant in explaining variance in P-O fit, however AC performance does add variance over and above these variables. This indicates support for H2a and H3a. Discussion The results indicate that AC performance has an important role to play in the selection of high potential graduates, because of its capacity over time to predict P-O fit. Very few studies have examined this relationship. The majority of studies use performance, promotions and salary as the criterion variables. Given the relative lack of research on AC performance and P-O fit, one important result arising from this study is that AC performance has the potential to provide incremental variance over and above that provided by other measures. The study revealed that socialisation tactics and the quality of training and development are important. The results for these dimensions suggested that graduates will perceive greater P-O fit with an organisation’s culture, where they are provided with a common and consistent message, positive signals and role models regarding what the organisation stands. Both socialisation tactics and the quality of training and development have lasting value. They were significant in explaining P-O fit, current salary and promotions over the six-year period of the study. The analysis suggests that high intensity training and development and socialisation tactics will lead to graduate greater P-O fit. The results for the lasting impact of AC performance are perhaps surprising. It is likely that AC performance would be valuable in explaining P-O fit early in the employment relationship. I took my first measure three months into the relationship. A possible explanation is that the skills that are assessed in an AC are relevant in understanding graduate job performance and the types of on-the-job behaviours that potentially lead to high performance and quick promotion opportunities. The potential of overall AC performance to predict P-O fit at later stages of the employment relationship was not anticipated or predicted in the literature. Over time, AC performance has value in explaining P-O fit. The reasons for this are difficult to ascertain. However, I would suggest that graduates were more likely to perceive that organisations that adopted such sophisticated selection processes as ACs are more likely to have a progressive corporate culture and to support management and career development initiatives. The study should be placed within the context of a growing body of research on P-O fit. I focused on a unique population, high potential graduates who were entering the labour market on a full-time basis for the first time. The graduates were employed in a variety of organisations across different industry sectors. The study showed that P-O fit, the number of promotions and current salary are best predicted incorporating a combination of achievement, competence, personality and organisational strategies such as socialisation and training and development. However, this study does reveal that AC performance have predictive power over and above these variables. ACs can be designed to address short-term as well as long-term concerns. They have value to a fast-track graduate

population in that they provide a strong measure of learning outcomes from education as well as providing an indication of long-term fit with an organisation. It is likely that the use of ACs may be viewed in a positive light by graduates and as a result they are more likely to make positive attributes concerning the suitability of the organisation as a place to work. ACs are likely to be perceived as having both greater face validity and acceptability (Howard, 1997). The feedback component of the AC process in this study may have served an important role in facilitating P-O fit, in that it helped the graduate determine suitability to the organisation as well as giving the graduate the perception that he/she was valued by the organisation. It should be noted that in this study I did not examine actual P-O fit. Instead I focused on graduate perceptions of P-O fit. I argue that perceptions of fitting-in are important in themselves. Subjective P-O fit will likely be influenced by the sophistication of selection methods as well as the intensity of socialisation and training and development processes. It is not however, clear in the present study, whether graduates actually developed greater P-O fit because of these training and socialisation processes. The study relied on self-report data; however this was collected at three points in time. This longitudinal feature suggests that I can be more confident concerning the direction of causality. Given the tenure of respondents, the results of the study may be biased by self-selection in that graduates who had low P-O fit had already quit. The average turnover for the four study organisations was moderate, 14 per cent over the six year period, and the results revealed considerable variance in P-O fit for the three measurement points [t1 M ¼ 3:67, SD ¼ 0:91; t2 M ¼ 3:41, SD ¼ 0:89; t3 M ¼ 3:79, SD ¼ 0:87]. This suggests that self-selection bias due to turnover was minimal in this study. The study had a high response rate that suggests that respondents may not have differed significantly from non-respondents. I conducted a non-response analysis that indicated that the data did not have a significant problem in this regard. The limitations of this study are countered by several important strengths. The exercises included in the AC were relevant in that they were derived from job analyses and competency frameworks within the four organisations. The study used a matched sample and had three sets of data for each respondent. The sample size is satisfactory and it was drawn from four different industries. This sample diversity increases my confidence that the results are not biased by idiosyncratic assessment centre and socialisation tactics, or on the specific expectations of entry into a particular industry. Thus the characteristics of the sample potentially increase the generalisability of the results. The sample consisted of genuine graduates who had applied to join a graduate development programme. Therefore, it is a field study. It differs from other studies of this type that have used students still in educational settings. The longitudinal design is valuable however it resulted in sample attrition. At each stage of the data collection, the sample was significantly reduced. Sample attrition occurred for a number of reasons including graduate turnover, and non-response. Further research could usefully examine other individual differences that may impact P-O fit. It is likely that the strength of an organisation’s culture is important in explaining P-O fit. Therefore culture strength could be included as a variable. It would also be useful to gather data from supervisors in order to verify salary, promotions and perceptions of P-O fit. The study focused on subjective P-O fit. It is possible that graduate subjective fit was more potent at t1 due to the relative lack of experience of each graduate and the limited opportunity to observe the actual organisation. It may

Assessment centre performance 163

JMP 22,2

have been appropriate at t2 and t3 to have investigated the link between objective fit (fit assessed by including the supervisor’s perception) and subjective fit. I used P-O fit as a criterion variable, however it would be interesting to investigate the impact of subjective P-O fit on work adjustment and other outcomes such as performance.

164

Conclusion This study examined relationships between graduate assessment centre performance and P-O fit. It also examined the moderating influence of graduate work experience, personality, level of degree, socialisation tactics and quality of training and development. The study found strong support for the contribution of assessment centre performance to explaining P-O fit. Further research may benefit from examining this relationship with a different sample of employees, and it would be valuable to investigate how other individual differences may impact P-O fit.

.

References Adkins, C.L., Russell, C.L. and Werbel, J.D. (1994), “Judgements of fit in the selection process: the role of work value congruence”, Personnel Psychology, Vol. 47, pp. 605-23. Anderson, N., Born, M. and Cunningham-Snell, N.A. (2001), “Recruitment and selection: application perspectives and outcomes”, in Anderson, N., Ones, D.S., Sinangil, H.K. and Viswesvaran, C. (Eds), Handbook of Industrial Work and Organizational Psychology, Sage, New York, NY, pp. 200-18. Arnold, J. (1989), “Career decidedness and psychological well-being: a two cohort longitudinal study of undergraduate students and recent graduates”, Journal of Occupational Psychology, Vol. 2, pp. 163-76. Arnold, J. and MacKenzie-Davey, K. (1999), “Graduates’ work experiences as predictors of organisational commitment, intention to leave and turnover: which experiences really matter?”, Applied Psychology: An International Review, Vol. 48, pp. 211-38. Autry, C.W. and Wheeler, A.R. (2005), “Post-hire human resource management practices and person-organisation fit: a study of blue-collar employees”, Journal of Managerial Issues, Vol. 17 No. 1, pp. 58-75. Azjen, I. (2001), “Nature and operation of attitudes”, Annual Review of Psychology, Vol. 52, pp. 27-58. Bartels, L.K., Bommer, W.H. and Rubin, R.S. (2000), “Student performance: assessment centres versus traditional classroom evaluation techniques”, Journal of Education for Business, Vol. 75, pp. 198-201. Baruch, Y. (1999), “Integrated career systems for the 2000s”, International Journal of Manpower, Vol. 20 No. 7, pp. 432-57. Boyatzis, R., Stubbs, E.C. and Taylor, S.N. (2002), “Learning cognitive and emotional intelligence competencies through graduate management education”, Academy of Management Learning and Education, Vol. 1 No. 2, pp. 150-62. Bretz, R.D. and Judge, T.A. (1994), “Person-organisation fit and the theory of work adjustment: implications for satisfaction, tenure and career success”, Journal of Vocational Behaviour, Vol. 44, pp. 32-54. Bretz, R.D., Rynes, S.L. and Gerhart, B. (1993), “Recruiter perceptions of applications fit: implications for individual career preparation and job search behaviour”, Journal of Vocational Behaviour, Vol. 44, pp. 32-54.

Burnett, M. and Dutsch, J.V. (2006), “Competency-based training and assessment centre: strategies, technology, process and issues”, Advances in Developing Human Resources, Vol. 8 No. 2, pp. 141-4. Chalman, J.A. (1991), “Matching people and organisations: selection and socialisation in public accounting firms”, Administrative Science Quarterly, Vol. 36, pp. 459-84. Cheng, E.W.L. and Ho, D.C.K. (2001), “The influences of job and career attitudes on learning motivation and transfer”, Career Development International, Vol. 6 No. 1, pp. 20-7. Clarke, D., Wiseman, L. and Garven, I. (1994), Review of Fast-Stream Recruitment, HMSO, London. Cohen, P.A. (1984), “College grades and adult achievement: a research synthesis”, Research in Higher Education, Vol. 20, pp. 281-93. Connor, H., Strebler, M. and Hirsch, W. (1990), You and Your Graduates: The First Few Years, Institute of Manpower Studies, Brighton. Cooper-Thomas, H.D., Van Vianen, A. and Anderson, N. (2004), “Changes in person-organisation fit: the impact of socialisation tactics and actual P-O fit”, European Journal of Work and Organisational Psychology, Vol. 13 No. 1, pp. 52-78. Coupland, C. (2001), “Accounting for change: a discourse analysis of graduate trainees’ talk of adjustment”, Journal of Management Studies, Vol. 38 No. 8, pp. 1103-19. Fournier, V. (1998), “Stories of development and exploitation: militant voices in an enterprise culture”, Organisation, Vol. 5 No. 1, pp. 55-80. Gaugler, B.B., Rossenthal, D.B., Thornton, S.C. III and Benston, V. (1987), “Meta analysis of assessment centre validity”, Journal of Applied Psychology, Vol. 72, pp. 493-511. Graham, C. and McKenzie, A. (1995), “Delivering the promise: developing new graduates”, Education þ Training, Vol. 37 No. 2, pp. 33-40. Greenhaus, J.H. (1988), “Career exploration”, in London, M. and Mono, E. (Eds), Career Growth and Human Resource Strategies, Quorum, New York, NY, pp. 17-30. Hendry, C., Arthur, M.B. and Jones, A.M. (1995), Strategy through People: Adaptations and Learning in the Small-Medium Enterprise, Routledge, London. Hough, L.M. (1997), “The millennium for personality psychology: new horizons or good old daze?”, Applied Psychology: An International Review, Vol. 47, pp. 233-61. Howard, A. (1997), “A reassessment of assessment centres: challenges for the 21st century”, Journal of Social Behaviour and Personality, Vol. 12, pp. 13-52. Hsin-Chih, C. (2006), “Assessment centre: a critical mechanism for assessing HRD effectiveness and accountability”, Advances in Developing Human Resources, Vol. 8 No. 2, pp. 247-64. Jones, G.R. (1986), “Socialisation tactics, self efficacy and newcomers adjustments to organisations”, Academy of Management Journal, Vol. 29, pp. 262-79. Judge, T.A. and Cable, D.M. (1997), “Applicant personality, organisational culture and organisational attrition”, Personnel Psychology, Vol. 50, pp. 359-94. Judge, T.A., Cable, D.M., Bourdreau, J.W. and Bretz, R.D. (1995), “An empirical investigation of the predictors of executive career success”, Personnel Psychology, Vol. 48, pp. 485-519. Kim, T.Y., Cable, D.M. and Kim, S.P. (2005), “Socialisation tactics, employee productivity and person-organisation fit”, Journal of Applied Psychology, Vol. 90 No. 2, pp. 232-41. Lado, A.A., Boyd, N.G. and Wright, P. (1992), “A competency model of sustainable competitive advantage, towards a conceptual integration”, Journal of Management, Vol. 18, pp. 779-99. LeDeist, F.D., Delamare, F. and Winterston, I. (2005), “What is competence?”, Human Resource Development International, Vol. 8 No. 1, pp. 27-46.

Assessment centre performance 165

JMP 22,2

166

Lievans, F., Decaesteker, C., Coetsier, P. and Geirnaert, J. (2001), “Organisational attractiveness for prospective applicants: a person-organisation fit perspective”, Applied Psychology: An International Review, Vol. 50 No. 1, pp. 30-51. Louis, M.R. (1980), “Surprise and sense-making: what newcomers experience on entering unfamiliar organisational settings”, Administrative Science Quarterly, Vol. 25, pp. 226-51. Louis, M.R. (1990), “Acculturation in the workplace, newcomers as lay ethnographers”, in Schneider, B. (Ed.), Organisational Climate and Culture, Jossey-Bass, San Francisco, CA, pp. 85-129. McDaniel, M., Schmidt, F. and Hunter, J. (1988), “Job experience correlates of job performance”, Journal of Applied Psychology, Vol. 73, pp. 327-30. MacKenzie-Davey, K. and Arnold, J. (1995), “Women’s responses to the organisation and accounts of their personal change since starting work”, Proceedings of the British Psychological Society Occupational Psychology Conference, Warwick. MacKenzie-Davey, K. and Arnold, J. (2000), “A multi-method study of accounts of personal change by graduates starting work: self-ratings, categories and women’s discourses”, Journal of Occupational and Organisational Psychology, Vol. 73, pp. 481-8. Mayrhofer, W., Steyer, J., Meyer, M., Strunk, G., Schiffinger, M. and Iellatchitch, A. (2005), “Graduates’ career aspirations and individual characteristics”, Human Resource Management Journal, Vol. 15 No. 1, pp. 1-19. Ryan, A.M. and Ployhart, R.E. (2000), “Applicants’ perceptions of selection procedures and decisions: a critical review and agenda for the future”, Journal of Management, Vol. 26, pp. 565-606. Salgado, J.F. (1997), “The five factor model of personality and job performance in the European community”, Journal of Applied Psychology, Vol. 82, pp. 30-43. Siebert, S.E. and Kraimer, M.L. (2001), “The five-factor model of personality and career success”, Journal of Vocational Behaviour, Vol. 58, pp. 1-21. Sturges, J., Guest, D., Conway, N. and MacKenzie-Davey, K. (2002), “A longitudinal study of the relationship between career management and organisational commitment among graduates in the first ten years at work”, Journal of Organisational Behaviour, Vol. 25 No. 3, p. 457. Thornton, G.C. III and Cleveland, J.N. (1990), “Developing managerial talent through simulations”, American Psychologist, Vol. 45, pp. 190-9. Thornton, G.C. III, Tziner, A., Dahan, M., Clevenger, J.P. and Meir, E. (1997), “Construct validity of assessment centre judgements: analyses of the behavioural reporting methods”, Journal of Social Behaviour and Personality, Vol. 12, pp. 109-28. Tokar, D.M. and Subich, L.M. (1997), “Relative contributions of congruence and personality dimensions of job satisfaction”, Journal of Vocational Behaviour, Vol. 50, pp. 482-91. Tziner, A. (2002), Human Resource Management and Organization Behaviour, Ashgate Publishing, Aldershot. Tziner, A., Murphy, K.R. and Cleveland, J.N. (2005), “Contextual and rater factors affecting rating behaviour”, Group and Organization Behaviour, Vol. 30 No. 1, pp. 89-98. Tziner, A., Ronen, S. and Hacohen, D. (1993), “A four-year validity study of an assessment center in a financial corporation”, Journal of Occupational Behaviour, Vol. 14, pp. 225-37. Viney, C., Adamson, S. and Doherty, N. (1997), “Paradoxes of fast-track career management”, Personnel Review, Vol. 26 No. 3, pp. 174-86. Westerman, J.W. and Vanka, S. (2006), “A cross-cultural empirical analysis of person-organisation fit measures as predictors of student performance in business

education: comparing students in the United States and India”, Academy of Management Learning and Education, Vol. 4 No. 4, pp. 409-20. Further reading Arnold, J. and Mackenzie-Davey, K. (1994), “Graduate experience of organisational career management”, International Journal of Career Management, Vol. 6 No. 3, pp. 14-18. Boyatzis, R.E. (1982), The Competent Manager: A Model for Effective Performance, John Wiley, New York, NY. Caplan, R.S. (1987), “Person-environment fit theory and organisations: commensurate dimensions, time perspectives and mechanisms”, Journal of Vocational Behaviour, Vol. 31, pp. 248-67. Carless, S.A. (2005), “Person-job fit versus person-organisation fit as job acceptance intentions: a longitudinal study”, Journal of Occupational and Organisational Psychology, Vol. 78, pp. 411-29. Gniatczyk, L.A. and Ladd, R.T. (2001), “Assessment centre validity in a downsizing organisation”, paper presented at the Conference of the Society for Industrial and Organisational Psychology, San Diego, CA, April. Howard, A. and Bray, D.W. (1988), Managerial Lives in Transition: Advancing Age and Changing Times, Guilford, New York, NY. Jansen, P.G.W. and Sloop, B.A.M. (2001), “The dynamics of assessment centre validity: results of a 7-year study”, Journal of Applied Psychology, Vol. 86, pp. 741-53. Kristof, A.L. (1996), “Person-organisational fit: an integrative review of its conceptualisations, measurement and implications”, Personnel Psychology, Vol. 49, pp. 1-49. Kristof-Brown, A.L. (2000), “Perceived applicant fit: distinguishing between recruiters’ perceptions of person-job and person-organisation fit”, Personnel Psychology, Vol. 53, pp. 643-71. Kudisch, J.D., Ladd, R.T. and Dobbins, G.H. (1997), “New evidence on the construct validity of diagnostic assessment centres: the findings may not be so troubling after all”, Journal of Social Behaviur and Personality, Vol. 12, pp. 129-44. Lauver, K.J. and Kristof-Brown, A. (2001), “Distinguishing between employees’ perceptions of person-job and person-organisation fit”, Journal of Vocational Behaviour, Vol. 59, pp. 454-70. Naquin, S. and Holton, E.F. III (2006), “Leadership and management competency models: a simplified process and resulting model”, Advances in Developing Human Resources, Vol. 8 No. 2, pp. 144-65. Schneider, B. (1987), “The people make the place”, Personnel Psychology, Vol. 40, pp. 437-53. Schneider, B., Goldstein, H.W. and Smith, A.B. (1995), “The ASA framework: an update”, Personnel Psychology, Vol. 48, pp. 747-73. Tae-Yeol, K., Cable, D.M. and Sang-Pyo, K. (2005), “Socialisation tactics, employee proactivity and person-organisation fit”, Journal of Applied Psychology, Vol. 90 No. 2, pp. 232-41. Corresponding author Thomas N. Garavan can be contacted at: [email protected] To purchase reprints of this article please e-mail: [email protected] Or visit our web site for further details: www.emeraldinsight.com/reprints

Assessment centre performance 167

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

JMP 22,2

Gender enactment at work

168

The importance of gender and gender-related behavior to person-organizational fit and career decisions Angela M. Young Department of Management, School of Business and Economics, California State University, Los Angeles, California, USA, and

David Hurlic Graduate School of Education and Psychology, Pepperdine University, Los Angeles, California, USA Abstract Purpose – This paper seeks to explore gender-related behavior in relation to person-group fit, person-organization fit, and career decisions in order to identify relevant constructs and relationships and present suggestions for further research. Design/methodology/approach – A model was developed based on a review of several literature bases including gender theories, gender enactment, person-group fit, and person-organization fit. Findings – Propositions for future research were developed and focus on the relationships influencing an individual’s perceptions of person-group fit, person-organization fit and career decision making. It is proposed that a deviation from a group’s collective gender orientation and accepted behaviors is likely to be met with unfavorable perceptions by group members and impact person-group fit, person-organization fit, perceived stress, self-efficacy, and career decision making. Practical implications – Managers and supervisors can use ideas presented in the model and paper to better understand nuances of gender-related behavior and its potential impact on behavior and perceptions in the workgroup and organization. Diversity managers and training professionals could incorporate aspects of gender-related differences into training programs, and individuals might use some ideas presented in this paper to consider the important implications of person-group and person-organization fit to career success. Originality/value – The model developed in this paper is a unique perspective combining historical perspectives on gender and gender-related behavior and workplace concepts such as person-group and person-organization fit to identify the potential impact of gender-related differences at work. Keywords Gender, Careers Paper type Conceptual paper

Journal of Managerial Psychology Vol. 22 No. 2, 2007 pp. 168-187 q Emerald Group Publishing Limited 0268-3946 DOI 10.1108/02683940710726429

Introduction There are several management studies that examine gender differences in the workplace (Heilman et al., 1995; Karakowsky and Elangovan, 2001; Miller and Karakowsky, 2005; Tepper et al., 1993) but underlying psychological explanations for these differences are not often the primary focus of gender research. Lilly et al. (2005) recently examined work-family conflict among men and women based on needs and roles, and used a sociological theory of gender reproduction to support proposed A previous version of this manuscript was presented at the conference for the Institute of Behavioral and Applied Management, 2006.

relationships. Gender reproduction theories have not been the focus of management research but could provide useful theoretical support for gender and gender-related behavior in relation to aspects of organization fit. The purpose of this paper is to explore gender-related behavior in relation to person-group and person-organization fit, and the resulting impact on career decisions. Specifically, we propose that individuals seek equilibrium between group and organizational environments and a personal tendency to manifest gender-related behavior. A deviation from a group’s collective gender-related orientation and accepted behaviors is likely to be met with unfavorable perceptions by group members, and impact person-group fit, person-organization fit, perceived stress, self-efficacy, and career decision making. The relationships among the variables of interest are illustrated in Figure 1. Gender-related behavior: theoretical approaches Laslett and Brenner (1989) discussed gender reproduction theory from an historical perspective and described gender as a socially constructed variable that is based on biological sex type. Thus there are certain behaviors that have been associated with masculinity and being a man and femininity and being a woman. Men and women reproduce gender-related behavior in many aspects of their lives. While sex type may not drive gender-related behavior, the two are highly correlated according to the authors. Gender is institutionalized and impacts how we think about all aspects of our lives and how members of society develop ideas and values about appropriate gender-related behavior, jobs, and activities. Ridgeway and Smith-Lovin (1999) examined gender-related behavior as well, and viewed gender as a system that is created versus a naturally occurring phenomenon in men and women. Gender systems, according to Ridgeway and Smith-Lovin, are historically rooted in inequality between men and women and comprised primarily of social practices. As such, gender must be viewed as an interaction between men and women within contexts of power and status. The active manifestation of behaviors associated with being male or female was termed “doing gender” by West and Zimmerman (1987). The authors base much of their work on Goffman (1977) who defined behaviors that denote masculinity or femininity as gender display. Goffman viewed gender display behaviors as choices, optional behaviors in which men and women may or may not engage. Thus when a man behaves in a manner that is deemed masculine by society, it is his choice to do so. Similarly, women have a choice in enacting behaviors that are viewed as particularly feminine. West and Zimmerman (1987) questioned the extent to which gender display was discretionary and presented two issues related to gender-reproduced behaviors. First, gender-reproduced behaviors must be viewed as an interaction between the actor and the group within which behaviors are enacted. And second, because “doing gender” goes beyond a single person’s behavior and extends into situations, groups, and activities, it is not always optional. While West and Zimmerman supported the idea that men and women may act in any gender-related manner they wish, to engage in gender-related behavior that is inappropriate based on group perception would be detrimental. Thus, discretion to engage in gender-related behavior is limited by the demands of specific situations.

Gender enactment

169

JMP 22,2

170

Figure 1. Gender enactment and fit in relation to career decisions

Society at large has determined that some situations and activities are more masculine or feminine just as society has concluded certain behaviors are more masculine or feminine (Laslett and Brenner, 1989). From a very early point in childhood development, people are inundated with feedback from others regarding behaviors (West and Zimmerman, 1987). We learn which behaviors are acceptable, which should be abandoned or hidden, and the contexts in which certain behaviors are appropriate. As we go through life, we participate in different situations and learn that to successfully navigate different situations and deal with different groups of people, diverse behaviors may be required. From a gender perspective, behaviors that are acceptable for men and women differ, thus we reach a dilemma when our desired gender-reproduced behaviors conflict with the expectations of the people around us. Gender enactment at work There is some evidence that expectations or preconceived notions for gender-related behavior exist in the workplace and when men and women act in ways incongruent with those expectations there may be negative consequences. For example, Heilman et al. (1995) provided research that supported the idea that stereotypes and expectations are formed about the work-related behavior of men and women. The authors found that definitions of successful management coincided with characteristics typically associated with masculinity, such as rationality and independence. Thus, because of preconceived ideas about gender, behaviors of men and women in management positions may not be interpreted similarly. Eagly and Johannesen-Schmidt (2001) focused on gender perceptions in leadership and explained that leadership characteristics can be categorized as agentic and communal. Agentic behaviors are considered assertive behaviors while communal behaviors are associated with nurturing behaviors. Men are typically described as naturally possessing agentic qualities while women are described in communal terms. The authors contend that, from the start, gender influences perceptions formed about men and women in leadership positions despite actual behaviors. These studies suggest that men are viewed as more naturally possessing qualities of successful leaders and managers. The detriment to women, in this case, is that they may be judged as less effective leaders and managers even before they begin work in a leadership position in an organization. There is further evidence that our gender and how we behave in terms of expected gender-related behavior impacts overall career success, and men and women may adjust behavior depending upon the context. For example, Tepper et al. (1993) found evidence that men and women receive different reactions and outcomes depending upon the influence strategies used with supervisors. Men were more successful when using strong influence tactics such as directness. Women were not as successful when using strong tactics, tactics considered incongruent with femininity, and received more positive outcomes when weaker tactics were utilized. Miller and Karakowsky (2005) discovered that men and women change their feedback-seeking behavior in groups depending upon the type of task presented and gender composition of the group. Further evidence is provided by Karakowsky and Elangovan (2001) who studied risk tolerance in groups of men and women and found that higher risk tolerance, typical of men, became the standard for group decision-making in mixed-gender groups.

Gender enactment

171

JMP 22,2

172

Beyond gender, the gender orientation of groups has been found to impact work outcomes and, in one study, teams with a common gender orientation experienced higher performance than teams without a common orientation (Scherer and Petrick, 2001). Findings suggest that, in some situations, men and women may be at a disadvantage if their tendency to engage in certain gender-related behavior is not aligned with the particular situation. However, people know that certain situations call for different behaviors, thus, people may try to suppress or exaggerate gender-reproduced behaviors depending upon the situation. Another aspect of gender reproduction theory is that there is no certainty that biological sex will be associated with typical gender-reproduced behaviors (Laslett and Brenner, 1989). A woman may engage in typically masculine behaviors while a man may act in a manner typically considered feminine. However, given other viewpoints, perhaps it is not only much a matter of tendency, but tendency combined with sensitivity to the demands of a given situation. Thus gender reproduction theories offer a possible explanation for why we develop more masculine or feminine behaviors and emphasize the point that gender-reproduced behaviors are present, to some degree, across situations and may be adjusted to fit situations or groups (Goffman, 1977; West and Zimmerman, 1987). Further, gender-related behavior congruent to the gender orientation of a group or situation may yield better outcomes for the individual and the group than behaviors that are incongruent with the situation or group (Scherer and Petrick, 2001; West and Zimmerman, 1987). West and Zimmerman’s (1987) phrase “doing gender” adds an important component to understanding gender-based behaviors in conjunction with a situation; however, there still lacks an element of specificity in defining the awareness, on the part of the individual, to act in gender-appropriate ways. Moreover, West and Zimmerman do not focus on an individual’s desire or tendency to engage in masculine or feminine behaviors. We contend that a person’s tendency to engage in masculine or feminine behaviors and awareness that gender display must be adjusted to fit others’ expectations are essential to understanding the impact of gender-related behavior at work. Young and Hurlic (2006) refer to the combination of awareness and purposely adjusted or engineered behavior as “genderneering”. We propose that awareness of the need to change one’s behavior and general tendencies toward specific behaviors interact with the gender orientation of the group and the group’s level of tolerance for incongruent gender-related behavior. The group level components, group gender orientation and tolerance, provide cues to individuals as to the importance and type of gender-related behavior adjustments required to fit into the group. We propose that group fit ultimately plays an important role in career decision making by impacting perceived stress and perceived self-efficacy at work. Depending upon an individual’s perceived group fit, and the resulting effect on job-related self-efficacy and perceived stress, he or she may decide to make a career change. Gender enactment Gender enactment is the composite of an individual’s awareness that situations demand gender-related behavior adjustments and an individual’s orientation toward masculine or feminine behaviors. Awareness of the situation and gender orientation,

exist within each person and are proposed to interact and manifest into actual behavior. Awareness. Awareness that a situation demands potential adjustment to gender-related behavior has not been the focus of research. The construct most closely representing awareness or sensitivity for the need to adjust gender-related behavior is self-monitoring (Snyder, 1974, 1987). Snyder explained that self-monitoring behavior occurs when individuals focus on and consider situational cues to adjust behavior. A high self-monitoring person will adjust his or her behavior based on perceived signals from people and situations. If a person is aware of the need to adjust gender-related behavior, it is more likely that he or she will be attuned to the situation and the reactions of others. Self-monitoring behavior in the workplace has been positively linked to organizational citizenship behaviors (Blakely et al., 2003), life satisfaction (Anderson and Randlet, 1993), and job stress (Fox and Dwyer, 1995). Measures of self-monitoring behavior address general perceptions about behavior adjustment, but do not focus specifically on gender-related behavior (Lennox and Wolfe, 1984). For example, if group members are asked if they sometimes behave in ways that will help them fit into a group, there is no guarantee that they will answer the question with gender-related behavior adjustments in mind. However, there is evidence that gender-related behavior may be considered when an individual self-monitors himself or herself. In a study in which interviewees were paired with female interviewers who enacted feminine or androgynous behaviors during the interview, it was found that high-self monitoring female interviewees paired with feminine interviewers reported higher femininity based on Bem’s (1974) sex role inventory (Smith et al., 1997). Development of a gender-specific construct would allow for a more precise determination of the extent to which people are aware of a need to adjust gender-related behavior. Another dimension of self-awareness that could be applicable to an individual’s ability to adjust or monitor his or her gender behavior can be found in various emotional intelligence (EI) constructs. For example, many authors have written that the awareness dimension of EI gives individuals the ability to use emotional information in a constructive and adaptive manner (Lane et al., 1990; Mayer and Salovey, 1997; Mayer et al., 2000). While other authors identify an individual’s level of self-awareness (i.e. perception and appraisal) as critical in any process that involves personal transformation or change (England, 2002; Goleman, 1998; Mayer and Salovey, 1997). Additionally, authors go even further by stating that self-awareness gives individuals the ability to deal effectively and successfully with environmental and organizational change and issues of diversity (Around-Thomas, 2004; Goleman, 2004; Shipper et al., 2003). Literature related to self-awareness suggests that awareness is the initiator or catalyst for not only appraising the organizational environment one works in, but taking the information gleaned from the surroundings and making the proper adjustments in order to successfully fit into a group or organization. Orientation. One of the more typical aspects of gender enactment in research is gender orientation (Bem, 1974), and many aspects of gender-related attributes and behavior have been examined in the workplace. Heilman et al. (1995) explained that women are typically believed to be less competent than men, particularly in terms of management responsibilities. The authors used supporting evidence from studies on

Gender enactment

173

JMP 22,2

174

gender and selection, in which women were found to be less desirable candidates for management positions. Terms used to describe successful managers included competence, independence, and rationality, typically masculine terms, according to the authors, and typically not used to describe women. Heilman, Block, and Martell found that attributes associated with successful managers, such as competence, emotional stability, independence, and rationality were higher for women who were specifically identified as successful managers than for women who were not identified in any particular job category. In addition, among male and female managers, women were scored lower than men on attributes of success. Yet, Johnson (1994) found no gender differences in actual managerial behavior when studying men and women. In a meta-analysis of gender and leadership literature, Eagly et al. (2003) found that women exhibited more tendencies of transformational or charismatic leadership than did men despite typical stereotypes of women as less effective leaders. Research has shown that perceptions of men and women in the workplace impact career outcomes, but so do actual behaviors. For example, Tepper et al. (1993) found that influence tactics required for high performance ratings and receipt of mentoring differed depending upon gender. Men were more likely to attain positive results when they used strong upward influence tactics while women were more successful when using weaker tactics. Miller and Karakowsky (2005) conducted a laboratory study to test gender effects on feedback seeking behavior. The authors examined gender composition of groups during a gender-biased negotiation exercise and found that men sought feedback more often when working in a male-dominated group and on a male-oriented task. Women sought feedback in male-dominated groups working on a male-oriented task but did not seek feedback as often in female-dominated groups working on a female-oriented task. More evidence is presented by Karakowsky and Elangovan (2001) who examined gender in relation to risk tolerance and decision making and found that men were more tolerant of risk than were women, and in mixed-gender groups it was the risk tolerance of men more than women that determined group decision making. These studies emphasize the impact and importance of gender and related behaviors in relation to work and the perceptions of others. Findings support the notion that many behaviors in which we engage are perceived and perhaps evaluated based in part on our gender and expected gender-related behavior. Gender enactment requires a focus on the individual and the situation or group. Thus we now change our focus from the individual to the group and discuss the gender orientation of the work group, or the micro-culture, within which a person works. The gender-based micro-culture Within every organization, there are many sub-cultures that might be unique from the rest of the organization in terms of culture or accepted behaviors. These sub-cultures are referred to in this paper as micro-cultures, the culture of one area in the organization that may differ from the rest of the organization. Employees must be attuned to the micro-cultures of the organization as they conduct business in different parts of the organization or consider job changes into other organizational areas. The micro-culture, as used in this paper, refers to two primary components: the level of tolerance for gender-related behavior that differs from the collective expectations (i.e.

Group Acceptance) and the collective gender orientation of the group (i.e. Group Orientation). Group acceptance. A group’s level of acceptance of gender-related behavior that is incongruent with the group’s collective expectations could impact the team’s performance and team member performance. Karakowsky et al. (2004) presented evidence that gender-diverse groups tended to be more open to nontraditional and traditional gender-related behavior. Acceptance is important to consider because feelings of acceptance within a group have been found to impact an individual’s sense of altruism and extra-role behavior (Tierney et al., 2002). Acceptance of incongruent gender-related behavior in the workplace could be assessed individually or combined with other group member scores to assess an overall group attitude. A stronger sense of person-group fit is likely to occur when individuals feel accepted, while lower levels of fit may be associated with a lack of acceptance. The higher the level of acceptance across all group members, perhaps the higher an individual’s perceived person-group fit. Group orientation. There is evidence that collective gender orientation can be identified among groups and when team members hold similar gender schemas, team member performance increases (Scherer and Petrick, 2001). Scherer and Petrick (2001) investigated gender and team schema in relation to team outcomes. The researchers examined teams from two perspectives: gender composition of the group and gender-orientation in terms of traditional or non-traditional, masculine or feminine, orientation. They found that teams in which members held similar schemas performed better than teams with differing schemas and that teams comprised of predominantly one sex supported enactment of behaviors congruent with the dominant sex-type of the group. This study supports the idea that a group gender orientation exists and provides evidence that gender and group gender-orientation differences impact work outcomes. Karakowsky et al. (2004) investigated communication behavior and gender in groups, and presented evidence that men in male dominated groups engaged in more frequent verbal interruptions that did women in female dominated groups. In addition, they found that men in male dominated groups interrupted others more than men in mixed-gender groups, and that the greater the proportion of women to men in a group, the fewer the number of total interruptions. The authors also manipulated gender-orientation of the task finding that women, more than men, decreased interruptive behavior when working on a male-oriented task. The authors speculated that communication behaviors are indicative of perceived power and status, and thus may influence perceived leadership ability and may change depending upon the gender composition of the group. More evidence of the impact of group gender orientation can be found in a study by Sywensky and Madden (1996) who examined sex, psychological gender and perceived leadership ability among residence hall students at a university. The researchers expected a leader to emerge within each residence hall but found that small groups of leaders emerged, and leader-groups were typically of mixed-gender composition. However, even among the women who emerged as leaders, there was a predominant male gender orientation. For the purpose of this paper, it is proposed that the gender-based micro-culture represents the collective gender orientation of the group. When an individual’s gender

Gender enactment

175

JMP 22,2

orientation is congruent with the collective gender orientation, there is likely to be a stronger perceived person-group fit.

176

Gender enactment and person-group fit Fit has been viewed from many perspectives including job, organization, and group (Kristof, 1996). Much of the person-group fit research deals with personality traits of team members and outcomes such as team performance or effectiveness (Kristof-Brown et al., 2005). Few researchers focus on gender and related issues in terms of person-group or person-organization fit (Scherer and Petrick, 2001). Person-group fit Person-group fit is the extent to which an individual perceives his or her characteristics, behaviors, and interaction to be compatible with other group or team members (Chuang and Sackett, 2005; Kristof, 1996). Opinions, ideas, attitudes, and beliefs can all, to some extent, be hidden. Gender, however, is unequivocally visible. Moreover, norms of behaviors for men and women are learned, practiced, and manifest in most of our activities. When actual behavior deviates from expected norms, it is likely that such behavior will influence perceptions of other group members, and the person whose gender-related behavior does not mesh with the group’s expectations may be seen as a poor fit. Thus, it is not so much a matter of biological sex or gender-related behavior consistent with one’s biological sex, but the gender-related behavior engaged in by a man or a woman within the context of a work group. For example, if a man engages in overtly feminine behaviors, but many of the peers in his work group behave in much the same manner, he is likely to be perceived as a good fit in that particular environment. Likewise, if a woman engages in particularly masculine behaviors but works in a group that is predominantly female and those women, in a collective sense, engage in extremely feminine behaviors, that woman is likely to be seen as a poor fit. Combining the aspects of gender enactment and the micro-culture, we propose that individual awareness and gender orientation, and the level of group acceptance and orientation will influence person-group fit: P1.

When awareness for the need to adjust gender-related behavior is high, it is likely that a person will be perceived, by self and others, as a good fit in the group.

P1b. When a person’s gender orientation is congruent with the micro-culture’s gender orientation, it is likely that a person will be perceived, by self and others, as a good fit in the group. P1c. When a micro-culture has a high level of acceptance of incongruent gender-related behavior, it is likely that a person will be perceived, by self and others, as a good fit in the group, regardless of the person’s gender orientation or awareness. Skills and education There is evidence that men and women working in jobs that are predominantly held by the opposite sex experience different work outcomes (Scherer and Petrick, 2001). In addition, a person who is in a position typically held by a member of the opposite sex is

not always perceived as competent of befitting the position (cf. Heilman et al., 1995). As skills and education increase among women, more women will be filling traditionally male jobs. Further, Scherer and Petrick (2001) found that nontraditional groups reported higher group cohesion than more traditional groups indicating that perhaps acceptance of nontraditional behaviors will be of greater necessity as the world of work becomes more diverse. Extant research supports the idea that gender-related behavior influences the extent to which a person is seen as fitting or befitting a particular position or job. However, skills and education do much to improve the chances of career success, at least for women (Daley, 1996). Daley found that women’s career advancement was more highly dependent on objective factors such as education and skills than was men’s career development. It is unknown whether or not men working in predominantly female environments have the same reliance on skills and education. Given that research is limited, we therefore propose that skills and education will moderate the relationship between gender enactment and person-group fit: P2.

Gender enactment and the micro-culture will impact person-group fit but the relationship will be moderated by skill and education level.

Person-group fit, person-organization fit, and career decisions The extent to which people feel as though they fit into a work group is likely to result in emotional outcomes such as levels of perceived stress and self-efficacy, and the positive and negative emotions related to person-group fit are likely to play a part in career decision making. Possible career decisions such as seeking promotions, lateral moves, remaining in a position, or leaving the organization might also depend upon the general perceptions of the organization’s culture in a broader sense. That is, organizations are comprised of multiple micro-cultures, but there may be an overall perception held by individuals in terms of possibilities for advancement and opportunities. Some organizations communicate opportunities openly to employees increasing the chances that employees will seek positions and opportunities while other organizations require individuals to actively seek opportunities (Carless, 2005). We propose that, among the many factors that influence perceptions of person-organization fit, person-group fit will have an impact. The resulting career decision, however, will likely depend upon emotional factors related to perceived person-group fit and person-organization fit such as perceived stress and self-efficacy. Person-group fit and perceived stress When a person perceives a lack of ability to meet perceived or actual demands, an imbalance is created, that imbalance has been identified as perceived stress (Edwards and Cooper, 1990). When there is a balance between perceived control and demands, stress is minimized (Karasek, 1979). Among the many types of stressors at work, most researchers agree that relationships at work are a potential source of stress (Murphy, 2002). As person-group fit depends upon work relationships, it is proposed that person-group fit has an influence on perceived stress. Extant research provides evidence that men and women differ in the extent to which they perceive stress, depending upon the gender-orientation of the environment in which they work. Gardiner and Tiggermann (1999) examined differences in leadership style, job stress, and mental health among men and women in male and female

Gender enactment

177

JMP 22,2

178

dominated industries and found that leadership style differed depending upon industry. Women in male dominated industries were as interpersonally oriented as men but reported higher interpersonal orientation in female dominated industries. Women in male dominated industries also reported more stress than men due to tokenism, discrimination, and other stressors in the organization, but men and women reported similar scores in terms of mental health. Thus, gender, and perhaps gender-related behavior, may impact perceptions of fit and perceived stress. As individuals seek ways to cope with and improve person-group fit, they may seek to change some aspects of their behavior including gender-related behavior depending upon their gender orientation and awareness as well as the micro-culture. Specifically, a person’s gender orientation may support a natural fit within in the micro-culture and person-group fit will be high. In that case, very little adjustment of gender-related behavior is needed in order to fit into a group. However, when a person’s gender orientation and micro-culture are incongruent, a person’s awareness of the need to adjust gender-related behavior is low, or when the micro-culture is intolerant of incongruent gender-related behavior, there is likely to be a low perceived person-group fit. When perceived group fit is low, it is proposed that a person will perceive higher job-related stress. Person-group fit and self-efficacy Bandura (1977, 1986) presented the construct of self-efficacy as it relates to work behavior and defined self-efficacy as the belief individuals hold about their abilities to perform a specific task. Self-efficacy has been linked to work performance and other organizational behaviors and has become a commonly accepted construct in understanding training and performance issues (cf. Eylon and Bamberger, 2000). Self-efficacy, in the most straightforward sense, is based on a person’s belief about his or her ability to achieve a specific goal or complete a specific task, but perceived ability is not the only influence on self-efficacy (Bandura, 1986). The way that people react to us, or how we perceive they act toward us, influences our self-perceptions and feelings of self-efficacy and confidence about job-related performance (Helms, 1990; Kane and Montgomery, 1998). Evidence of the influence of groups on self-perception can be found in the work of Helms (1990) who explained that individual intra- and interpersonal characteristics are the basic components of group climate. The author contended that very subtle influences on group perceptions and behaviors could occur even by changing group demographic composition. Even differences in attitudes could be impacted based on the group or social context in which one functions, and in particular, when an individual perceives himself in a minority relative to the group norm (Davis and Cheng, 1996; Helms, 1990). Wells-Parker et al. (1990) created and tested role specific self-efficacy scales and linked self-efficacy to depression and coping orientation in a study of women in four distinct roles: occupation, parenting, marriage, and economic management. The authors acknowledged that findings must be validated for men, but the study provided support for the importance of self-efficacy in terms of emotional outcomes related to working in an environment dominated by people of the opposite sex. DeJong et al. (1999) investigated team self-efficacy and personality factors in relation to team effectiveness finding that extraversion, conscientiousness, and self-efficacy for

teamwork were significantly and positively correlated with attraction to a team and team members’ ratings of team effectiveness. Therefore, self-efficacy, an important consideration in career decision-making, can be influenced by many aspects of group interaction. It is proposed that job-related self-efficacy will be impacted by person-group fit and will be an influence in career decision making. Specifically, when person-group fit is low, job-related self-efficacy will diminish while a high level of perceived person-group fit will increase job-related self-efficacy. Therefore, considering both stress and self-efficacy, we propose that the level of perceived person-group fit will impact perceived stress, self-efficacy, and person-organization fit: P3.

The level of person-group fit will influence levels of perceived stress and self-efficacy.

Person-organization fit Person-organization fit is generally viewed as compatibility between the person and the organization based primarily on congruent goals and values (Chatman, 1989; Vancouver and Schmitt, 1991). Fit occurs when some mutually beneficial arrangement occurs between a person and an organization that allows each to meet goals, referred to as complimentary fit, or when a similarity exists in terms of values, characteristics or culture, labeled supplementary fit (Carless, 2005; Kristof, 1996; Muchinsky and Monahan, 1987). Person-organization fit typically focuses on selection processes and is based on a perception that a job candidate fits with a job, organization, and team in terms of personality, skills, values, and the environment (Lievens et al., 2001; West and Cyr, 2004). There is evidence that person-organization fit influences selection decisions and recruiters often develop a sense of a candidate’s potential fit (Cable and Judge, 1997). Of course, candidates make an assessment about their potential fit in organizations and use that perception in making job choices (Cable and Judge, 1996). Chuang and Sackett (2005) examined person-organization fit from recruiters’ perspectives and found that, while person-job fit is a constant and important factor in hiring decisions, person-organization fit became more important to recruiters across interview stages. Recruiters placed the most importance on person-organization fit during the final stages of the interview process. From a candidate’s perspective, Carless (2005) found that person-organization fit was related to a candidate’s perception of organizational attractiveness, but not related to intentions to accept a job. We focus on the latter type of person-organization fit, the more subjective fit, based on an individual’s perception of how well they fit into an organization’s broader culture. Beyond the selection process, person-organization fit has been shown to influence work attitudes and career decisions (Bretz and Judge, 1994; Verquer et al., 2003). The extent to which person-organization fit changes over time was studied by Cooper-Thomas et al. (2004) who focused on perceived fit of newcomers at organizational entry and again after four months. The authors found that, although perceived fit was generally stable, newcomers who received more social support had higher levels of perceived fit. Kim et al. (2005) examined socialization and person-organization fit in an international context and considered the employee’s role in perceived fit. Specifically, the authors emphasized the influence employees have on perceived fit in terms of their acceptance of socialization, organizational messages, and their own behaviors. Perceived fit was higher in employees who had a positive attitude toward early socialization experiences.

Gender enactment

179

JMP 22,2

180

Person-organization fit can be seen from two perspectives, that of others and of ourselves. A person may perceive that he or she fits into an environment, and may behave in ways that support that belief. However, peers and others may feel differently. The difference between actual and perceived fit is important, but it is the perception of the employee that is likely to be the ultimate influence on long-term career decisions (Cooper-Thomas et al., 2004). Feedback from others is supplemented by organizational communication. Organizations seek to communicate values through formal and informal means, thus providing additional information to individuals about the extent to which they fit into the broader organizational environment (Carless, 2005). These messages, formal and informal, from peers or others in various parts of the organization combine and provide us with a general perception of the organization’s culture. Based on an evaluation of information, individuals can determine the extent to which they perceive themselves as a fit in the organization. Our self-perceived fit is important because it is a strong driving force in career decision making. Whether accurate or not, our self-perception of person-organization fit might lead to decisions regarding seeking promotions, making lateral moves to different jobs or areas in the organization, leaving the organization, or remaining in a position and work group. Specific career decisions to remain in a job, leave a job through a lateral move or promotion but remain in the organization, or to leave the organization altogether depend upon the opportunities perceived to exist within the organization. Thus the more general perceptions of the organization’s culture, the macro-culture, are proposed to play an important role in career decision making. Gender-based macro-culture The term macro-culture is used here to define the broader and more general perceptions employees hold about the culture in other parts of the organization. It is a more generalized perspective of the organization’s culture and could be quite different from any micro-culture in the organization. Specifically, this paper focuses on an individual’s perception of acceptance of diverse gender-related behavior in the organization (i.e. macro-culture acceptance) and the gender orientation of the organization, at large (i.e. macro-culture orientation). Macro-culture acceptance. The macro-culture of an organization is proposed to influence the relationship between person-organization fit and career decision making because it is a person’s perception of opportunities that await them in the organization that plays an important role in actual and perceived career options (Carless, 2005). Again, we focus on the individual and his or her perception of the organization’s culture. If a person perceives a low level of person-organization fit in a current job, any decision about moving to another area in the organization or leaving the organization would depend, in part, on the perceived macro-culture. A macro-culture that is intolerant of diverse gender-reproduced behaviors leaves little opportunity for someone wishing to change work groups or jobs. A macro-culture that is perceived to be accepting of incongruent gender-related behavior may create a perception of opportunity and openness so that people may consider seeking opportunities within the organization. Macro-culture orientation. Just as the team or group can form a collective gender orientation, we propose that, from an employee’s perspective, the organization itself can be perceived as having a collective gender orientation. Of course, among the many

subsets of cultures within the organization, each may be different from others, but generally speaking, individuals may form perceptions about the organization’s collective culture (Carless, 2005; Muchinsky and Monahan, 1987; West and Cyr, 2004). Different from acceptance, orientation pertains to the actual gender orientation of the majority of members of the organization, or the predominant gender-based behavior of the collective. We are proposing that a collective organizational gender orientation may be perceived by individuals and therefore impact the decision to remain with or exit an organization. It is proposed that in the low fit, high stress, low self-efficacy condition in which the macro-culture is accepting of gender-related behavior incongruent with the collective, a person will be more likely to seek a promotion or lateral move within the company. In a situation of low fit, high stress, low self-efficacy and a macro-culture that is not tolerant of incongruent gender-related behavior, it is proposed that a person will be likely to leave the organization. In a high fit situation where there is an intolerant macro-culture, a person is likely to remain in a current position or to leave the organization. Finally, in a high fit condition with a tolerant macro-culture, employees might decide to remain in a position or seek promotion or lateral moves elsewhere in the organization when an appropriate time comes. P4a. Person-group fit will impact person-organization fit, but the relationship will be moderated by a person’s perception of the acceptance of incongruent gender-related behavior and the gender orientation of the organization. P4b. The extent to which a person perceives person-organization fit will impact career decisions, but the relationship will be moderated by perceived stress and self-efficacy. Directions for future research Empirical verification of the proposed relationships would yield valuable information about the importance of acceptance in the workplace and the impact of gender-related behavior on person-group fit, person-organization fit, and career decision-making. The first perspective in which the model could be tested is the perception of the respondent’s perception of himself or herself and the micro- and macro-cultures that are proposed to influence career decisions. The constructs presented in the model represent constructs measured at the individual level. For example, gender role orientation and awareness are presented as individual-level constructs. Group constructs such as group gender orientation and acceptance are suggested to represent individuals within a group and are suggested to be and the sum of individual assessments. Person-organization fit is the subjective assessment of an individual of his or her perceived fit in the broader organizational culture. Thus the model focuses primarily on individual perceptions of person-group and person-organization fit in relation to career decision-making. The model distinguishes among several of the constructs comprising gender enactment such as awareness, gender orientation of person and group, and acceptance of incongruent gender-related behavior. These constructs are measurable; however, some have been the focus of theoretical research but very little empirical verification. For example, awareness on the part of an individual has not been the focus of research and few studies have investigated group gender orientation (cf. Scherer and Petrick,

Gender enactment

181

JMP 22,2

182

2001) or acceptance in groups (c.f., Karakowsky et al., 2004). Because few studies have quantitatively addressed these issues, some measures would require development and validation. Focus groups might be useful to identify compelling issues and provide fruitful ideas to develop measures for gender enactment constructs, and for micro- and macro-culture constructs. In addition, some existing measures may be adapted and validated. For example, related to awareness of an individual for the need to change gender-related behavior, a self-monitoring scale (Snyder, 1987) or emotional intelligence measures (Mayer and Salovey, 1997; Mayer et al., 2000) may be used as a basis to begin developing scale items. Using focus groups and existing measures may provide researchers a useful means of tapping into awareness and other constructs. The constructs in this model could be assessed through more qualitative methods such as interviews with respondents and peers to delve into any emotional undertones of these constructs and proposed relationships. Some of the more emotional aspects of the model require in-depth investigation and may include emotions related to a person’s struggle to engage in gender appropriate behavior that does not feel natural or the discomfort associated with suppressing a desire to engage in gender-related behavior that is not accepted in a micro- or macro-culture. Likewise, peers and others in the organization may feel discomfort around group members who freely express incongruent gender-related behavior. These more emotional aspects of gender and fit may uncover some important considerations in career decision-making. This paper presents a model that identifies gender-related behavior and recognizes the individual’s ability to formulate the decision to stay or leave an organization. By understanding the individually intrinsic process that links awareness to gender enactment an organization can begin to truly modify its culture, if it so desires, so that a more equitable and open environment can evolve where individual differences can be respected, valued and accepted. Regardless of the approach or level of study used to investigate relationships proposed in the model, findings would be useful in understanding the impact of gender enactment in terms of person-group fit, person-organization fit, and career decisions. As organizations expand globally, the need to understand person-organization fit across micro-, macro-, and global-cultures is important, but very little has been done to assess gender and fit, or fit across cultures (Parkes et al., 2001). The primary focus of the model relates to issues of diversity and how to maximize the human capital in organizations. For example, organizations (domestic and global) have to deal with changing demographics and many authors have explored the changes brought about by issues of diversity (Allen and Montgomery, 2001; Friday and Friday, 2003; Miller, 1998; Pless and Maak, 2004). For example, Dass and Parker (1999) examined the linkage between executives’ priorities in managing workforce diversity, organizational conditions, and performance. To truly understand the effectiveness of an organization’s efforts with regards to person-organizational fit it is crucial that the elements of the model be recognized as important components in the person-organization fit equation. Faced with the reality of demographic changes taking place globally and the growth of global markets, organizations must be able to manage and maximized the diverse individuals that work for them. Understanding why an individual supports or discards an organizational should be as important as understanding how organizations achieve

objectives and goals. Human resource managers, supervisors, and anyone responsible for staffing or management must be made aware of the impact of gender-related behavior in order to justly and effectively develop policies and practices that nurture open and cohesive environments not only to avoid costly litigation, but to create a workplace that allows for each individual to have equal opportunity to be productive, to contribute, and to participate fully in the work environment. Supervisors, in particular, would benefit from an awareness of the impact of gender orientation and gender-related behavior within work groups to ensure that a cohesive and productive workgroup is created and to avoid unhealthy conflict among group members. Unhealthy behaviors indicative of intolerance have been found to negatively impact perceptions of personal well-being, lower self-efficacy (Eylon and Bamberger, 2000), lower productivity, and can lead to litigation (Laband and Lentz, 1998). Training managers too should be mindful of gender enactment and include it in sensitivity or diversity training programs. Diversity training that includes more subtle aspects of human behavior such as gender enactment at work could promote healthy relationships and behaviors at work. Finally, the employee himself or herself, who works in an environment in which the micro- or macro-culture is not congruent or comfortable might be helped by learning more about gender issues in the workplace and their potential emotional and tangible outcomes for personal career development. In terms of fitting into a work-group or organization, we must examine the extent to which gender-related behavior impacts our career decision-making. Conclusion The purpose of this paper is to explore gender-related behavior at work and to develop a model which can be used to guide future research and to learn more about gender-related behavior, person-group and person-organization fit, and its impact on career decision making. A model was developed and introduced the interactive relationship between a person’s awareness of the need to adjust gender-related behavior to situations or groups, gender orientation of individuals, gender orientation of groups, and the tolerance of groups for incongruent gender-related behavior. Gender-related behavior was proposed to impact person-group fit, person-organization, and ultimately career decisions. References Allen, R. and Montgomery, K. (2001), “Applying an organizational development approach to creating diversity”, Organizational Dynamics, Vol. 30, pp. 149-61. Anderson, L.R. and Randlet, L. (1993), “Self-monitoring and life satisfaction of individuals with traditional and nontraditional sexual orientations”, Basic and Applied Social Psychology, Vol. 14, pp. 345-61. Around-Thomas, M. (2004), “Understanding emotional intelligence can help after problem behavior”, Physician Executive, Vol. 30, pp. 36-9. Bandura, A. (1977), Social Learning Theory, Prentice-Hall, Upper Saddle River, NJ. Bandura, A. (1986), Social Foundations of Thought and Action: A Social Cognitive Theory, Prentice-Hall, Upper Saddle River, NJ. Bem, S. (1974), “The measurement of psychological androgyny”, Journal of Consulting and Clinical Psychology, Vol. 42, pp. 155-62.

Gender enactment

183

JMP 22,2

184

Blakely, G.L., Andrews, M.C. and Fuller, J. (2003), “Are chameleons good citizens? A longitudinal study of the relationship between self-monitoring and organizational citizenship behavior”, Journal of Business and Psychology, Vol. 18, pp. 131-41. Bretz, R.D. and Judge, T.A. (1994), “Person-organization fit and the theory of work adjustment: implications for satisfaction, tenure, and career success”, Journal of Vocational Behavior, Vol. 44, pp. 32-54. Cable, D.M. and Judge, T.A. (1996), “Person-organization fit, job choice decisions, and organizational entry”, Organizational Behavior and Human Decision Processes, Vol. 67, pp. 294-311. Cable, D.M. and Judge, T.A. (1997), “Interviewers’ perceptions of person-organization fit and organizational selection decisions”, Journal of Applied Psychology, Vol. 82, pp. 546-61. Carless, S.A. (2005), “Person-job fit versus person-organization fit as predictors of organizational attraction and job acceptance intentions: a longitudinal study”, Journal of Occupational and Organizational Psychology, Vol. 78, pp. 411-29. Chatman, J.A. (1989), “Improving interactional organizational research: a model of person-organization fit”, Academy of Management Review, Vol. 14, pp. 333-49. Chuang, A. and Sackett, P.R. (2005), “The perceived importance of person-job fit and person-organization fit between and within interview stages”, Social Behavior and Personality, Vol. 33, pp. 209-26. Cooper-Thomas, H.D., van Vianen, A. and Anderson, N. (2004), “Changes in person-organization fit: the impact of socialization tactics on perceived and actual P-O fit”, European Journal of Work and Organizational Psychology, Vol. 13, pp. 52-78. Daley, D.M. (1996), “Paths of glory and the glass ceiling: differing patterns of career advancement among women and minority federal employees”, Public Administration Quarterly, Summer, pp. 143-62. Dass, P. and Parker, B. (1999), “Strategies for managing human resource diversity: from resistance to learning”, The Academy of Management Executive, Vol. 13, pp. 68-80. Davis, L.E. and Cheng, L.C. (1996), “Differential effects of racial composition on male and female groups: implications for group work practice”, Social Work Research, Vol. 20, pp. 157-68. DeJong, R.D., Bouhuys, S.A. and Barnhoorn, J.C. (1999), “Personality, self-efficacy and functioning in management teams: a contribution to validation”, International Journal of Selection and Assessment, Vol. 7, pp. 46-9. Eagly, A.H. and Johannesen-Schmidt, M.C. (2001), “The leadership style of women and men”, Journal of Social Issues, Vol. 57, pp. 781-97. Eagly, A.H., Johannesen-Schmidt, M.C. and van Engen, M.L. (2003), “Tranformational, transactional, and laissez-faire leadership styles: a meta-analysis comparing women and men”, Psychological Bulletin, Vol. 129, pp. 569-91. Edwards, J.R. and Cooper, C.L. (1990), “The person-environment fit approach to stress: recurring problems and some suggested solutions”, Journal of Organizational Behavior, Vol. 11, pp. 293-307. England, D. (2002), “Inner leadership-personal transformation”, Industrial and Commercial Training, Vol. 34, pp. 21-7. Eylon, D. and Bamberger, P. (2000), “Empowerment cognitions and empowerment acts”, Group and Organization Management, Vol. 25, pp. 354-72. Fox, M.L. and Dwyer, D.J. (1995), “Stressful job demands and worker health: an investigation of the effects of self-monitoring”, Journal of Applied Social Psychology, Vol. 25, pp. 1973-95.

Friday, E. and Friday, S. (2003), “Managing diversity using a strategic planned change approach”, The Journal of Management Development, Vol. 22, pp. 863-80. Gardiner, M. and Tiggermann, M. (1999), “Gender differences in leadership style, job stress and mental health in male- and female-dominated industries”, Journal of Occupational and Organizational Psychology, Vol. 72, pp. 301-15. Goffman, E. (1977), “The arrangement between the sexes”, Theory and Society, Vol. 4, pp. 301-31. Goleman, D. (1998), Working with Emotional Intelligence, Bantam Books, New York, NY. Goleman, D. (2004), “What makes a leader?”, Harvard Business Review, Vol. 82, pp. 82-90. Heilman, M.E., Block, C.J. and Martell, R.F. (1995), “Sex stereotypes: do they influence perceptions of managers?”, Journal of Social Behavior and Personality, Vol. 10, pp. 237-52. Helms, J.E. (1990), “Generalizing racial identity interaction theory to groups”, in Helms, J.E. (Ed.), Black and White Racial Identity: Theory, Research, and Practice, Praeger, Westport, CT, pp. 187-204. Johnson, C. (1994), “Gender, legitimate authority, and leader-subordinate conversations”, American Sociological Review, Vol. 59, pp. 122-35. Kane, K. and Montgomery, K. (1998), “A framework for understanding disempowerment in organizations”, Human Resource Management, Vol. 37, pp. 263-75. Karakowsky, L. and Elangovan, A.R. (2001), “Risky decision making in mixed-gender teams: whose risk tolerance matters most?”, Small Group Research, Vol. 32, pp. 94-111. Karakowsky, L., McBey, K. and Miller, D.L. (2004), “Gender, perceived competence, and power displays: examining verbal interruptions in a group context”, Small Group Research, Vol. 35, pp. 407-39. Karasek, R.A. (1979), “Job demands, job decision latitude, and mental strain: implications for job redesign”, Administrative Science Quarterly, Vol. 24, pp. 285-310. Kim, T., Cable, D.M. and Kim, S. (2005), “Socialization tactics, employee proactivity and person-organization fit”, Journal of Applied Psychology, Vol. 2, pp. 232-41. Kristof, A.L. (1996), “Person-organization fit: an integrative review of its conceptualizations, measurement, and implications”, Personnel Psychology, Vol. 49, pp. 1-49. Kristof-Brown, A., Barrick, M.R. and Kay Stevens, C. (2005), “When opposites attract: a multi-sample demonstration of complementary person-team fit on extraversion”, Journal of Personality, Vol. 73, pp. 935-58. Laband, D.N. and Lentz, B.F. (1998), “The effects of sexual harassment on job satisfaction, earnings, and turnover among female lawyers”, Industrial and Labor Relations Review, Vol. 51, pp. 594-607. Lane, R., Quinlan, D., Schwartz, G., Walker, P. and Zeitlin, S. (1990), “The levels of emotional awareness scale: a cognitive-development measure of emotion”, Journal of Personality Assessment, Vol. 55, pp. 124-34. Laslett, B. and Brenner, J. (1989), “Gender and social reproduction: historical perspectives”, Annual Review of Sociology, Vol. 15, pp. 381-404. Lennox, R. and Wolfe, R. (1984), “Revision of the self-monitoring scale”, Journal of Personality and Social Psychology, Vol. 46, pp. 1349-64. Lievens, F., Decaesteker, C., Coetsier, P. and Geirnaert, J. (2001), “Organizational attractiveness for prospective applicants: a person-organization fit perspective”, Applied Psychology: An International Review, Vol. 50, pp. 30-51.

Gender enactment

185

JMP 22,2

Lilly, J.D., Duffy, J.A., Goswami, R.M. and Virick, M. (2005), “Work and family conflict: a needs theory perspective”, paper presented at the annual meeting of the Institute for Behavioral and Applied Management, Scottsdale, AZ. Mayer, J. and Salovey, P. (1997), “What is emotional intelligence?”, in Salovey, P. and Sluyter, D. (Eds), Emotional Development and Emotional Intelligence: Educational Implications, Basic Books, New York, NY, pp. 3-24.

186

Mayer, J., Salovey, P. and Caruso, D. (2000), “Models of emotional intelligence”, in Sternberg, R.J. (Ed.), Handbook of Intelligence, Cambridge University Press, Cambridge, pp. 396-420. Miller, D.L. and Karakowsky, L. (2005), “Gender influences as an impediment to knowledge sharing: when men and women fail to seek peer feedback”, The Journal of Psychology, Vol. 139, pp. 101-18. Miller, F. (1998), “Strategic culture change: the door to achieving high performance and inclusion”, Public Personnel Management, Vol. 27, pp. 151-60. Muchinsky, P.M. and Monahan, C.J. (1987), “What is person-environment congruence? Supplementary versus complementary models of fit”, Journal of Vocational Behavior, Vol. 31, pp. 268-77. Murphy, L.R. (2002), “Job stress at NIOSH: 1972-2002”, in Perrewe´, P.L. and Ganster, D.C. (Eds), Research in Occupational Stress and Well Being: Historical and Current Perspectives on Stress and Health, Vol. 2, JAI, New York, NY, pp. 1-56. Parkes, L.P., Bochner, S. and Schneider, S.K. (2001), “Person-organisation fit across cultures: an empirical investigation of individualism and collectivism”, Applied Psychology: An International Review, Vol. 50, pp. 81-108. Pless, N. and Maak, T. (2004), “Building an inclusive diversity culture: principles, processes and practice”, Journal of Business Ethics, Vol. 54, pp. 129-40. Ridgeway, C.L. and Smith-Lovin, L. (1999), “The gender system and interaction”, Annual Review of Sociology, Vol. 25, pp. 191-216. Scherer, R.F. and Petrick, J.A. (2001), “The effects of gender role orientation on team schema: a multivariate analysis of indicators in a US federal health care organization”, The Journal of Social Psychology, Vol. 14, pp. 7-22. Shipper, F., Kincaid, J., Rotondo, D. and Hoffman, R. (2003), “A cross-cultural exploratory study of the linkage between emotional intelligence and managerial effectiveness”, International Journal of Organizational Analysis, Vol. 11, pp. 171-91. Smith, J.L., Berry, N.J. and Whitely, P. (1997), “The effect of interviewer guise upon gender self-report responses as a function of interviewee’s self-monitoring position”, European Journal of Social Psychology, Vol. 27, pp. 237-43. Snyder, M. (1974), “Self-monitoring of expressive behavior”, Journal of Personality and Social Psychology, Vol. 30, pp. 526-37. Snyder, M. (1987), Public Appearances/Private Realities: The Psychology of Self-monitoring, Freeman, New York, NY. Sywensky, J.M. and Madden, J.L. (1996), “Effects of gender and sex type on perceived leadership abilities”, Journal of Group Psychotherapy, Psychodrama and Sociometry, Vol. 49, pp. 76-85. Tepper, B.J., Brown, S.J. and Hunt, M.D. (1993), “Strength of subordinates’ upward influence tactics and gender congruency effects”, Journal of Applied Social Psychology, Vol. 23, pp. 1903-19.

Tierney, P., Bauer, T.N. and Potter, R.E. (2002), “Extra-role behavior among Mexican employees: the impact of LMX, group acceptance, and job attitudes”, International Journal of Selection and Assessment, Vol. 10, pp. 292-303. Vancouver, J.B. and Schmitt, N.W. (1991), “An exploratory examination of person-organization fit: organizational goal congruence”, Personnel Psychology, Vol. 44, pp. 333-52. Verquer, M.L., Beehr, T. and Wagner, S.H. (2003), “A meta-analysis of relations between person-organization fit and work attitudes”, Journal of Vocational Behavior, Vol. 63, pp. 473-89. Wells-Parker, E., Miller, D.I. and Topping, J.S. (1990), “Development of control-of-outcome scales and self-efficacy scales for women in four life roles”, Journal of Personality Assessment, Vol. 54, pp. 564-75. West, C. and Zimmerman, D.H. (1987), “Doing gender”, Gender and Society, Vol. 1, pp. 125-51. West, J.W. and Cyr, L.A. (2004), “An integrative analysis of person-organization fit theories”, International Journal of Selection and Assessment, Vol. 12, pp. 252-61. Young, A.M. and Hurlic, D. (2006), “Genderneering at work: the importance of gender and gender-related behavior to person-group fit and person-organizational fit on career decisions”, paper presented at the annual meeting of the Institute for Behavioral and Applied Management, Memphis, TN. Corresponding author Angela M. Young can be contacted at: [email protected]

To purchase reprints of this article please e-mail: [email protected] Or visit our web site for further details: www.emeraldinsight.com/reprints

Gender enactment

187

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

JMP 22,2

Work value fit and turnover intention: same-source or different-source fit

188

Annelies E.M. van Vianen, Irene E. De Pater and Floor Van Dijk University of Amsterdam, Amsterdam, The Netherlands Abstract Purpose – The purpose of this article is to examine the relationship between work value PO fit with fit being operationalized as a same-source or different-source measure. Design/methodology/approach – A sample of 94 employees from a company located in Curac¸ao completed a questionnaire in which they rated their personal work values, their perceptions of the work values of the organization, and their turnover intention. Findings – The different-source work value fit measure was significantly and more strongly related to turnover intention than the same-source work value fit measure. Research limitations/implications – This study underscores the need for a careful reflection on the content and operationalization of fit measures. Practical implications – Managers are able to manage the fit of their employees by creating positive group level perceptions of the work environment. Originality/value – This study examines the effects of different fit measures on individuals’ turnover intention. Moreover, work values fit was investigated. Keywords Employee turnover, Influence Paper type Research paper

Journal of Managerial Psychology Vol. 22 No. 2, 2007 pp. 188-202 q Emerald Group Publishing Limited 0268-3946 DOI 10.1108/02683940710726438

Work value fit and turnover intention: same-source or different-source fit It has long been recognized that human behavior is a function of the person and the environment (Lewin, 1935). In organizational psychology, the person-environment (PE) fit perspective has become increasingly important. PE fit is defined as the compatibility that occurs when personal and situational characteristics are well matched (Schneider, 2001, p. 142). Theories that have dominated PE fit research in organizational psychology such as the theory of work adjustment (TWA) (Dawis and Lofquist, 1984) and the attraction-selection-attrition (ASA) framework (Schneider et al., 1995), propose that the degree of fit between people and their environment is positively related to important individual outcomes. In the context of work, PE fit studies have distinguished between person-job fit, person-team fit and person-organization fit, with the latter being the most investigated one. People’s fit with the organization (PO fit) associates a person’s personality, goals and values with those of the organization. Most PO fit studies have used needs and values as attributes of comparison between persons and organizations (Kristof-Brown et al., 2005). Particularly, peoples’ values seem to be of importance as attributes on which to establish their fit. The authors would like to thank Kiki Van den Berg for collecting the data.

Researchers that examined people’s value fit have used a wide variety of conceptualizations. Some researchers have treated peoples’ preferences and perceptions of organizational cultures as being values (Chatman, 1991; O’Reilly et al., 1991; Vandenberghe, 1999). Others have used peoples’ general core values as being the focal construct in value fit research (Van Vianen et al., 2004). To our best knowledge, there are no studies that have included values that refer to the different outcomes that people can obtain through their work, that is, their instrumental, cognitive and affective work values (see Elizur et al., 1991). A central proposition of PO fit theories, with Schneider’s ASA framework being the most prominent one, is that people will leave the organization if their characteristics do not fit those of the organization. However, most fit studies have examined relationships between levels of fit and job satisfaction and commitment rather than people’s turnover intention. In the present study we, therefore, investigated the linkage between fit regarding work values and turnover intention. PO fit studies not only vary with respect to the content of the fit measures and outcomes they employ, they also differ with respect to the methods they use to establish PO fit. Most PO fit researchers have used subjective measures of fit, that is, fit indices that depend on people’s perceptions. Examples of subjective fit measures are asking people directly about their perceived fit, or combining people’s reported values (P) with their perceptions of organizational values (O) into a fit index. The latter measure has recently been referred to as a same-source fit index in order to distinguish this measure from a different-source fit index in which other referents than the focal person are involved to provide the O-component of the fit index (Van Vianen, 2005). The different-source fit index has been often conceptualized as actual fit. This label, however, misleadingly suggests that the O-component has been established in an objective way, which is seldom the case[1]. In contrast to same-source fit measures, different-source measures of fit do not suffer from common method variance. Furthermore, they also reflect the phenomenon that the perceptions of others are significant sources of information for individual employees. For, employees may not only compare their values to their own perceptions of organizational values, but also to those of others. Theory and research on social information processing have convincingly shown that employees’ perceptions are affected by those of their co-workers (Salancik and Pfeffer, 1978). This is particularly the case when employees work in teams in which the team members perform tasks that are highly interdependent. PO fit studies that used same-source fit measures have been compared to those that used different-source fit measures. Generally, it was found that same-source methods yielded stronger results. However, to date, there are hardly any studies that compared these two methods within one and the same sample. In the current study we fill this void by assessing value fit in three different ways. Employees’ work values were compared with the work values of the organization as perceived by themselves, their team members, and organizational citizens in general. The first method concerns a same-source fit measure, whereas the second and third ones concern different-source measures. Below, we first address our choice for the specific content of the value-fit measure. Thereafter, we discuss issues that are related to the operationalization of fit measures.

Work value fit

189

JMP 22,2

Value fit Many PO fit studies have examined the match between people’s values and those of the organization, because values are conceived of as fundamental and relatively enduring. Values represent conscious desires held by the person and encompass preferences, interests, motives and goals (Edwards, 1996). They are defined as:

190

Desirable states, objects, goals, or behaviors, transcending specific situations and applied as normative standards to judge and to choose among alternative modes of behavior (Sagie and Elizur, 1996, p. 573).

Sagie and Elizur showed that individuals distinguish values based on two criteria: the life area and the modality. Life area refers to the area or domain of life to which the values are relevant, such as life in general, work, culture, sports, and religion. The modality refers to whether a value is instrumental, affective, or cognitive. Within each life area, instrumental, affective, and cognitive values are distinguished. Within the work context, values have been labeled as work values. Individuals differ regarding the importance they give to certain outcomes that they can obtain in the work setting (Sagie et al., 1996). Instrumental work values are work values related to obtaining desired ends, and refer to work benefits, work security, and success at work. Cognitive work values are those values that relate to a belief system about appropriate behaviors, and refer to broadening one’s horizons, contributing to society, and having meaningful work. Affective work values are related to feelings and emotions, and refer to happiness, good human relationships, and friendships at work. Although the structure of work (and life) values has been investigated (Elizur and Sagie, 1999; Sagie and Elizur, 1996), research has not yet addressed relationships between the specific types of work values (i.e. instrumental, affective, and cognitive) and organizational attitudes and behaviors. Also, research has not examined correlates of (in)congruence in cognitive, instrumental and affective work values. Rather, the research literature has treated values in work domains as an equivalent of organizational culture, and value congruence and person-organization fit have been perceived of as being identical (Kristof, 1996). Contrary to the general conceptualization of values that has been applied in the person-organization fit literature, work values refer to what people specifically strive for in work and they may, therefore, be more directly related to decisions about staying or leaving the job. Fit measures PO fit studies have used a large variety of operationalizations of fit. Table I presents an overview of the methods to establish fit that have been employed in the research literature. Researchers should make two types of decisions before choosing a fit measure. First, they need to decide who will provide information about the O-component of the fit measure; should the P and O component be derived from same Table I. Methods of PO fit assessment depending on the source of the O-component of the fit measure and the type of comparison between P and O

O-source

Hidden

Type of PO comparison Relative Absolute

Same as P

Direct fit measure

Profile comparison

Different from P

–

Profile comparison

(Absolute) discrepancy polynomial regression O-source is a constant

or different sources? Secondly, they need to make a decision about the type of comparison they want to make between the P and O component; are they interested in assessing absolute or relative fit between P and O? A simple method to assess fit is a direct one, namely asking respondents themselves to estimate the congruence between P and O, for example “my personal values match my organizations’ values and culture” (cf. Cable and DeRue, 2002). The problem with this measure is that it provides no information about the specific content of people’s fit perception. What do people have in mind when they, for example, say that they do not fit? Moreover, what is the direction of this misfit: do they experience excess (O . P) or deficiency (O , P)? These questions can be answered if respondents are asked to report the P and O component separately. They, for example, first reflect on their own values (P) and then on what they perceive as being the values of the organization (O). In this case, researchers subsequently need to decide how to deal with the P and O ratings in order to converge them into a measure of fit. They may compare the P and O ratings with each other in an absolute or relative way. When they are concerned with establishing the actual size of the differences between P and O, they estimate (absolute) discrepancy scores or use polynomial regression to assess the direction and shape of fit (Edwards, 1994). When the fit measure comprises several dimensions on which P and O differences are assessed, researchers may take the sum of the separate discrepancy scores rather than employing polynomial regression techniques. Many researchers have adopted a profile comparison approach for assessing fit, which is a relative and holistic comparison of persons and situations across multiple ratings. The P-ratings are correlated with the (commensurate) O-ratings (e.g. O’Reilly et al., 1991). These researchers are typically interested in similarity (i.e. fit) in the rank ordering of P and O ratings rather than absolute differences between P and O. Edwards (1993) noted that profile similarity indices rely on the assumption that each dimension of fit contributes equally to outcome measures. This may indeed be a problem, but particularly so if the P and O components concern a broad construct including a large diversity of dimensions, such as sub-dimensions of organizational culture preferences and perceptions. A profile comparison approach seems, however, less problematic if the construct is narrower defined. Another method to establish PO fit is to involve other persons than the focal person to provide the ratings for the O-component of the fit measure. This fit measure is based on different sources, that is P and O are measured independently. This approach is warranted if the P sources are not able to provide the O ratings, such as applicants or new entrants who are not yet aware of the characteristics of their (future) organization. Whether same-source or different-source measures should be used to assess fit mainly depends on the aim of the study. If researchers want to investigate how pre-entry fit (e.g. applicant fit) correlates with post-entry outcomes they will probably rely on different-source measures. Researchers have, however, also chosen for utilizing different-source measures when same-source measures were potentially available. These researchers were exclusively interested in assessing actual fit. They used aggregated O ratings, thus the shared perceptions of people, as reflecting the “true” organizational environment as a substitute for real objective environmental indicators that are generally not obtainable. Shared organizational perceptions are, however, not always found (West and Anderson, 1996). Should researchers conclude then that these O-ratings are unreliable

Work value fit

191

JMP 22,2

192

and, therefore, not suitable for calculating fit? The answer to this question again depends on the purpose of the particular study. If others’ perceptions are indeed used as a substitute for an objective assessment of environmental characteristics, the answer will be “yes”. If the perceptions of others are perceived of as an important source that influences the attitudes of individual employees, the answer might be “no”. This latter point-of-view can be best elucidated with theory that serves as a base for multilevel research (Lindell and Brandt, 2000). People’s attitudes are not only affected by their own perceptions but also by those of others. Social information processing (SIP) theory argues that individual attitudes are likely to be influenced by the opinion and information of others, because the perceptions of others are used to make sense of their environment (Salancik and Pfeffer, 1978; Thomas and Griffin, 1989). Multilevel research corresponds with SIP theory, for it was shown that, when people are “nested” in groups, researchers should include group level variables in addition to individual factors (Klein and Kozlowski, 2000). The primary sources of information in the work setting are those with whom one communicates frequently, such as people’s direct co-workers. Thus, individuals in groups may use their co-workers’ perceptions as sources of information about environmental conditions, even when their own perceptions are different from those of their peers. It has been argued that the organizational environment, for example organizational culture, can be described in terms of two parameters: strength and level (e.g. Zohar and Luria, 2004). Strength refers to the consensus among group members concerning their perceptions, whereas level refers to group members’ average perception, which also has been called the “quality” of environmental perceptions. Lindell and Brandt (2000) have shown that strength (consensus) and quality are constructs that are differently related to individual level and group level outcomes. Particularly, the quality of environmental perceptions was significantly associated with individual level outcomes, whereas consensus in environmental perceptions could not explain additional variance in individual outcomes once the quality of environmental perceptions was involved in the analyses. Because agreement of perceptions (i.e. consensus) can vary among groups, within group agreement statistics merely reflect the strength of perceptions. These statistics should not be considered as an aggregation criterion for estimating the quality of environmental perceptions. Although individuals will be affected by the perceptions of their peers, same-source measures of fit seem to have stronger relationships with individual outcomes such as turnover intention than different-source measures of fit (Kristof-Brown et al., 2005). This conclusion was, however, drawn from a variety of studies with most studies utilizing a single measurement method. To date, there are only few studies that compared the effects of different fit measurements within one and the same sample. In the present study, we examined relationships between work values fit and turnover intention with employing three different fit measures, one “same-source” measure of fit and two “different-source” measures of fit. More specifically, employees’ work values were linked to their own perceptions of the work values of the organization (Pi-Oi fit), the aggregated perceptions of the team members (Pi-Oteam fit)[2], and the aggregated perceptions of a larger group of organizational members (Pi-Oorg fit).

We have argued that people are not only influenced by their own perceptions of the work environment but also by those of their co-workers. We, therefore, expected significant relationships between both same-source and different-source value fit measures and turnover intention (H1). Based on the results of earlier PO fit research as presented above, we furthermore expected that the same-source fit measure would show stronger relationships with turnover intentions than the different-source fit measures (H2). Concerning the two different-source fit measures, we expected that a different-source fit measure that includes perceptions at the team level (as the O-component) would show stronger relationships with turnover intention than a different-source fit measure that includes perceptions at the organizational level (H3). Employees’ direct co-workers, thus the people with whom they work on a daily base, will have a stronger effect on individuals’ perceptions than the larger group of organizational members as a whole. Method Sample and procedure Respondents were employees from a liquor trading company in the Netherlands Antilles. A total of 105 employees were involved, grouped into sixteen teams, each consisting of three to sixteen team members, with a mean of nine employees per team. Team supervisors informed their employees about the study, asked them to participate, and handed out the questionnaires. A total of 94 respondents (73 percent were male) returned their questionnaire, which constituted a total return rate of 89.5 percent. Mean age was 39 (SD ¼ 10:01), and mean tenure at the company was 10.81 years (SD ¼ 8:89). Most respondents had received lower levels of education: 51 percent had received education lower than high school, and another 28 percent had followed the first years and lower levels in high school. Measures The survey included measures of employees’ own values, their perceived organizational values, turnover intention, and demographic characteristics. Work values were measured with 21 items derived from Elizur et al. (1991). The list of work values was asked twice: first as preferred work values, and secondly as perceived work values of the organization. Preferred work values were measured by asking respondents to indicate how important a value is for them, using a five-point scale ranging from “very unimportant” (1) to “very important” (5). An example of a cognitive work value is: “In my work, I find independence . . . ”. An example of an affective work value is: “In my work, I find interaction . . . ”. An example of an instrumental work value is: “In my work, I find work conditions . . . ”. Alpha coefficients were 0.87, 0.72 and 0.65 for cognitive, affective, and instrumental work values, respectively. Perceived work values were measured by asking respondents to indicate how important a value is for their organization (“This organization finds (work value) . . . ”). Alpha coefficients were 0.88, 0.79, and 0.78 for cognitive, affective, and instrumental perceived work values, respectively. Within-group agreement in perceived work values as estimated with rwg(j) (Lindell and Brandt, 2000) varied among the teams. Seven of the 16 teams had a minimum rwg(j) of 0.70 or higher. Mean rwg(j)’s were 0.60, 0.63, and 0.56 for cognitive, affective, and instrumental perceived work values, respectively.

Work value fit

193

JMP 22,2

194

Person-organization fit was operationalized in three ways, as: same-source fit (Pi-Oi fit), team-source fit (Pi-Oteam fit), and organization-source fit (Pi-Oorg fit). In order to be able to compare the three fit indices with each other, we used a profile comparison method. Hence, fit was calculated as the correlation between individuals’ work values and their and others’ perceived work values. Turnover intention was measured with three items derived from Van Vianen (2000). Respondents were asked to indicate their agreement with the items on a five-point scale ranging from “strongly disagree” (1) to “strongly agree” (5). One of the items was: “When I have the opportunity to work in another organization, I will take that opportunity.” Alpha coefficient was 0.68. Results Table II displays the means, standard deviations and correlation coefficients among the variables of this study. According to French et al. (1982) a moderate level of correlation between the P and O components of the fit measure will occur because of the tendencies for individuals to seek out jobs and organizations that meet their preferences. Therefore, correlations of 0.40 and lower are acceptable. Our correlations between the commensurate personal and perceived work values were 0.36, 0.40, and 0.25, for affective, cognitive, and instrumental work values, respectively. Small to modest correlations were found between personal work values on the one hand and fit measures and turnover intention on the other hand. Personal affective work values were significantly related to Pi-Oorg fit (r ¼ 0:33, p , 0.01) and turnover intention (r ¼ 20:25, p , 0.05). Employees that adhere relatively more to these values showed higher levels of Pi-Oorg fit and lower levels of turnover intention. Also, personal cognitive values were significantly related to Pi-Oteam fit (r ¼ 20:24, p , 0.05) and Pi-Oorg fit (r ¼ 20:28, p , 0.01). Employees with relatively high cognitive values showed lower levels of fit. As could be expected, perceived work values were negatively related to turnover intention (r ¼ 20:27, p , 0.01, r ¼ 20:35, p , 0.01, r ¼ 20:32, p , 0.01 for perceived affective, cognitive, and instrumental work values, respectively). Significant positive relationships between perceived work values and fit measures were only found for instrumental values (r ¼ 0:28, p , 0.01, r ¼ 0:22, p , 0.05 for Pi-Oi fit and Pi-Oteam fit, respectively). Correlations among the three fit measures ranged from 0.07 to 0.55, whereby Pi-Oi fit and Pi-Oorg fit were not significantly correlated with each other, but both these fit measures correlated with the Pi-Oteam fit measure. Additionally, Pi-Oteam fit and Pi-Oorg fit were significantly related to turnover intention (r ¼ 20:35, p , 0.01 and r ¼ 0:22, p , 0.05, respectively). The correlation between Pi-Oi fit and turnover intention was not significant (r ¼ 20:18). H1 to H3 were tested with hierarchical regression analysis. We specifically tested whether the three fit measures could explain additional variance in turnover intention above and beyond personal work values (P) and perceived work values (O). In the first step we entered the three P work values scales (affective, cognitive, and instrumental) together with the three O work values scales. In the second step we added the three fit measures: Pi-Oi fit, Pi-Oteam fit, and Pi-Oorg fit. As can be seen in Table III, personal work values were not significantly related to turnover intention. A significant relationship was only found for perceptions of organization’s cognitive work values

0.29 * * 0.13 0.23 * 0.24 *

0.70 2 0.22 * 0.33 2 0.19 0.26 2 0.18 0.25 0.02

0.85

4.15 0.26 0.23 0.26

2.01

0.05

0.24 *

0.36 * *

0.50 * *

0.65 * *

–

4

0.40 * *

0.33 * *

0.61 * *

–

5

0.24 *

0.18

–

6

2 0.39 * * 2 0.18

2 0.25 * 20.20

2 0.20

0.29 * * 0.24 * 0.24 * 0.25 * 0.23 * 2 0.09 20.16 2 0.08 0.23 * 0.12 20.24 * 2 0.09 0.15 0.33 * * 20.28 * * 0.10

0.25 *

0.30 * *

0.60 2 0.10

4.00

0.18

0.29 * *

0.62 2 0.18

4.19

0.19

0.18

0.23 *

0.49

4.59

0.08

0.51 2 0.01

4.32

0.17

0.12

0.26 *

0.50 2 0.11

4.52

3

–

2 – 0.57 * *

1

0.27 0.44 2 39.10 10.01 2 0.06 10.81 8.89 2 0.19

SD

Note: *p , 0.05, * *p , 0.01; * * *Gender 0 ¼ male, 1 ¼ female

Gender * * * Age Tenure Personal affective values Personal cognitive values Personal instrumental values Perceived affective values Perceived cognitive values Perceived instrumental values Pi-Oi fit Pi-Oteam fit Pi-Oorg fit Turnover intention

M

–

8

– 0.28 * 0.22 * 0.14

9

– 0.41 * * 0.07

10

2 0.27 * * 20.35 * * 2 0.32 * * 2 0.18

0.75 * * 0.76 * * 0.10 0.10 0.14 20.01 0.12 20.03

0.81 * *

–

7

–

12

13

2 0.35 * * 2 0.22 * –

– 0.55 * *

11

Work value fit

195

Table II. Means, standard deviations and correlations of the study variables (n ¼ 94)

JMP 22,2

196

Table III. Regression of turnover intention on individual work values, perceived work values, and fit measures

Step 1 Personal work values: Affective Cognitive Instrumental Perceived work values: Affective Cognitive Instrumental Step 2 Pi-Oi fit Pi-Oteam fit Pi-Oorg fit

b

b

20.15 0.16 20.02

0.02 2 0.07 2 0.01

0.19 20.40 * 20.19

0.21 2 0.49 * 2 0.05 2 0.02 2 0.32 * 2 0.09

Notes: Step 1: R 2 ¼ 0:18, Fð6; 75Þ ¼ 2:70, p ¼ 0:02; Step 2: R 2 ¼ 0:28, Fð9; 72Þ ¼ 3:08, p ¼ 0:003, DR 2 ¼ 0:10, Fð3; 72Þ ¼ 3:34, p ¼ 0:024; *p , 0.05; * *p , 0.01

(b ¼ 20:40 in the first model and b ¼ 20:49 in the second model). Additional variance in turnover intention was explained by Pi-Oteam fit (b ¼ 20:032). Relationships between turnover intention and the other two fit measures were nonsignificant. H1 was partly supported, since only Pi-Oteam fit was significantly related to turnover intention. Contrary to our expectation, the same-source fit measure was not related to turnover intention. Hence, H2 was rejected. However, as expected, the different-source fit measure that included perceptions at the team level (Pi-Oteam fit) showed stronger relationships with turnover intention than a different-source measure including perceptions at the organizational level (Pi-Oorg fit). We have argued that employees may use their co-workers as referents to establish their fit. The results seem to support this view, but the data did not allow us to test this in a direct way. There was, however, an opportunity to provide some evidence in a more indirect way. For example, by showing that group level perceptions indeed affected employees’ attitudes. We have already found that particularly employees’ perceived cognitive work values were associated with turnover intention. This may explain why a fit measure that is based on these perceptions (i.e. Pi-Oi fit) could not explain additional variance above and beyond perceived work values (Oi), a result that is often found in the PO fit literature (Van Vianen, 2005). In an additional analysis, we tested whether group level perceptions of organization’s cognitive, affective, and instrumental work values could explain variance in turnover intention above and beyond individual level perceptions of these values. Furthermore, we examined whether Pi-Oteam fit could contribute to explaining additional variance in turnover intention after the inclusion of group level variables. We used a mixed model of analysis. This is a regression model for hierarchically structured data that takes into account within-group variability as well as between-group variability. As a first step, we estimated a null model in order to test for possible variance in intercepts (groups varying regarding turnover intention). A fixed intercept model was tested against a random intercept model. The random intercept model fitted the data significantly better than the fixed intercept model (2 2

Log Likelihood AIC ¼ 200:286 and 205.286, respectively, with df ¼ 1). Moreover, results show an intraclass correlation of 0.21. Hence, there were differences between units with respect to turnover intention, which indicates that multilevel regression analyses rather than ordinary regression analyses should be performed. In the following steps, we estimated different models for each of the perceived work values scales, with the individual level variables defined as fixed or random. Through this we tested whether the slopes varied among the groups (i.e. groups differ randomly in the regression coefficients of individual level variables). An overview of the results is presented in Table IV. Models 2 and 3 show that fit with the data was found when individual level variables were fixed, meaning that the slopes did not differ among the groups. The intercept remained random for the analyses involving affective and instrumental work values when the group level variable was entered into the regression. However, a similar model was not applicable to the analyses involving cognitive work values. Instead, a model in which the intercept was estimated as fixed provided a better fit to the data. Hence, when group perceptions of cognitive work values were included in the regression, differences in intercepts among the groups did not exist any longer. This suggests that group perceptions of cognitive work values were strongly related to turnover intention. Pi-Oteam fit was added to the regression equation in model 4. Significant coefficients were found for individual perceptions of cognitive and instrumental work values, group perceptions of cognitive work values, and Pi-Oteam fit. We conducted a final regression analysis including these independent variables. Table V shows that group perceptions of cognitive work values (b ¼ 20:87, p , 05) and Pi-Oteam fit (b ¼ 20:97, p , 0.01) could significantly explain variance in turnover intention.

Affective work values Intercept Slope

Cognitive work values Intercept Slope

Instrumental work values Intercept Slope

Model 1 Empty model

Random

–

Random

–

Random

–

Model 2 Individual level work value

Random

Fixed *

Random

Fixed * *

Random

Fixed * *

Random

Fixed * Fixed *

Random

Fixed * Fixed Fixed *

Model 3 Individual level work value Group level work value Model 4 Individual level work value Group level work value PO fit Notes: *p , 0.05; * *p , 0.01

Random

Fixed Fixed *

Fixed

Fixed * Fixed * *

Random

Fixed Fixed Fixed *

Fixed

Fixed * Fixed * Fixed * *

Work value fit

197

Table IV. Testing different hierarchical multilevel models with turnover intention as the dependent variable and individual level and group level work values perceptions as independent variables

JMP 22,2

198

Table V. Final hierarchical multilevel analyses with turnover intention as the dependent variable and individual level and group level work values perceptions, and Pi-Oteam fit as independent variablesa

Discussion This study examined person-organization fit, operationalized as work value fit, and its relationship with turnover intention. In contrast to earlier value fit studies, we used three different methods to estimate individuals’ work value fit with the organization. Fit researchers generally assume that people take their own environmental perceptions as a frame of reference to which they compare their personal characteristics, such as their preferences, needs and values. Based on Social Information Processing theory, we argued that employees might also take the environmental perceptions of their coworkers into account. As opposed to other PO fit researchers who used the perceptions of organizational citizens as a way to assess the objective environment, we considered these perceptions as potential sources of information. If researchers use the aggregated perceptions of organizational citizens as a proxy for the objective environment, high interrater agreement among individuals is required. However, consensus is not a prerequisite if the quality of perceptions is the focal construct, as was the case in the present study. We expected that employees would use the perceptions of others as a frame of reference, but we also assumed that employees would mainly rely on their own perceptions. Based on the results of previous fit studies, we hypothesized stronger fit relationships for the fit measure including employees’ own perceptions (Pi-Oi fit) as compared to fit measures including the perceptions of others (Pi-Oteam and Pi-Oorg fit). Unexpectedly, we found substantial relationships between employees’ turnover intention and the Pi-Oteam and Pi-Oorg fit indices, whereas no significant relationship was found with the Pi-Oi fit measure. Specifically, group level perceptions of cognitive work values and Pi-Oteam fit were significantly and most strongly related to employees’ turnover intention. This study involved employees who were already in their company for a certain amount of time. Typically, this restricts the possibility of finding strong fit effects, because employees who have stayed in the company are likely the ones that already fit their environment to a reasonable extent (e.g. Simmering et al., 2003). Our findings, therefore, can be conceived of as being the product of a conservative test rather than chance. Most important then is to come up with a reasonable explanation of why we found these relatively strong group level effects. The opinions of coworkers are a very salient component of an employee’s environment. Our results showed that coworkers’ perceptions of cognitive work values affected individuals’ turnover intention. This corroborates earlier research that found Coefficientb

SE

Cognitive work values Individual level perceptions Group level perceptions

2 0.31 2 0.87 *

0.24 0.34

Instrumental work values Individual level perceptions Pi-Oteam fit

2 0.07 2 0.97 * *

0.18 0.32

Notes: aIntercept and slopes are fixed; bThe coefficients are unstandardized regression coefficients; *p , 0.05; * *p , 0.01

that the quality of environmental perceptions were linked to individual level outcomes (Lindell and Brandt, 2000). Additionally, our results suggest that individuals apparently compare their personal values to coworkers’ perceptions of organizational values rather than their own environmental perceptions. Social comparison theory is another well-established model of how individuals’ behaviors and attitudes depend on the relative comparisons with other people (e.g. Buunk and Mussweiler, 2001). People do not only compare themselves (i.e. their characteristics, situation, outcomes) to other people but they are also interested in others’ thoughts in similar circumstances. Particularly, when people are uncertain about their opinions they are interested in the opinions of others (Festinger, 1954). In this study, employees’ opinions about the values of their organization refer to what has been called “belief-type” opinions, which are potentially verifiable assertions about the true nature of an entity (Suls et al., 2000). In these situations, people use others’ opinions as a source of establishing the true nature of their environment. Furthermore, asking employees about the work values of their organization might be a query that is more ambiguous than asking them about their perception of organizational supplies. In the first instance they may lean on the general opinions of their coworkers, whereas in the second case they may mainly trust their own opinion. This might explain why previous fit studies showed stronger effects of same-source as compared to different-source fit measures, because most of these studies were concerned with perceptions of organizational characteristics rather than perceptions about the importance of values for the organization. Sweeney and McFarlin (2004) suggest that comparison processes might vary across borders and cultures. They argue that “a perspective on the self reflects a western emphasis on individualism, one in which the self is viewed as independent of others rather than interdependent” (p. 150). Our sample concerned employees from Curacao. Curacao is one of the Antillean islands (The Netherlands Antilles) with about 15,000 inhabitants. Antilleans seem to identify mainly with their own island and they resist the western management models of Dutch organizations. From the literature little is known about the cultural characteristics of Curacao, but as compared to western nations, such as the US and The Netherlands, Curacao is less individualistic (personal communication Gert Jan Hofstede). Hence, Curacao culture has a stronger in-group interdependence in which the relevance of others is more salient for individuals. Social comparison research has shown that referent comparisons explain more variance in attitudes for collectivists than individualists (Bordia and Blau, 2003). This study was limited since only one organization was involved with a restricted number of teams. We were not able to delve deeper into possible antecedents and moderators that could further explain our results. For example, although high consensus among team members is not necessary if one is interested in examining the quality of the work environment, the influence of group-level perceptions may depend on the degree to which perceptions are shared (Bommer et al., 2003). Social Information Processes theory argues that group-level perceptions are particularly powerful when there is relatively high consensus among group members. Moreover, research on social comparison of opinions show that people who share basic underlying values are preferred sources of information for evaluating an individual’s belief about the true environment. Both these theories suggest that group members may rely more on the

Work value fit

199

JMP 22,2

200

perceptions of others to the extent that there is consensus in personal values and perceptions. Future research should test this proposition by comparing intra-personal and inter-personal fit measures under different consensus conditions. Another potential limitation is the cross-sectional nature of the data and common method variance. We believe this effect to be marginal, because the zero-order correlations between the different variables were moderate. Furthermore, fit measures and group perceptions could explain additional variance in turnover intentions, which would have been difficult to achieve if our individual measures were strongly confounded. Our results have both scientific and practical implications. First, this study underscores the need of a careful reflection on the content and operationalization of fit measures. Particularly the O-component of this measure merits specific attention from researchers. If the content of the fit measure is ambiguous and less easy to observe, researchers should not only take individual perceptions but also group perceptions into account. Moreover, the cultural context and the level of interdependence among employees should also be considered as highly relevant for the assessment of the O-component. People’s environment is not only in the eyes of the beholder but also in those of beholder’s social group. Whether people fit their environment may, therefore, also depend on how this environment is “constructed” by salient others. Future research could, for example, compare people’s direct fit perceptions to same-source and different-source measures of fit. This would be the most direct way to examine whether people take themselves and/or salient others as a source for assessing the organizational environment. From a practical point-of-view one could argue that employees apparently are sensitive to the general perceptions of their work group. Social information processing research has convincingly shown that employees’ perceptions can be influenced through the communication and behaviors of their supervisors. Consequently, organizational leaders are not only able to manage employees’ experiences but also their fit with the organization. Notes 1. Kristof-Brown et al. (2005) define a fit measure that is calculated through the comparison of P and O variables as reported by different sources as “objective” fit. We find this label misleading because the organizational environment is established through the perceptions of people, which is by definition subjective. 2. Note that Pi-Oteam fit is not the same as person-group (PG) fit or person-team (PT) fit. See Kristof-Brown et al. (2005) for an overview. They define PG/PT fit as the interpersonal compatibility between individuals and their work groups, whereas our Pi-Oteam fit measure refers to team members’ aggregated perceptions of organization’s values. References Bommer, W.H., Miles, E.W. and Grover, S.L. (2003), “Does one good turn deserve another? Coworker influences on employee citizenship”, Journal of Organizational Behavior, Vol. 24, pp. 181-96. Bordia, P. and Blau, G. (2003), “Moderating effect of allocentrism on the pay referent comparison-pay level satisfaction relationship”, Applied Psychology: An International Review, Vol. 52, pp. 499-514.

Buunk, B.P. and Mussweiler, T. (2001), “New directions in social comparison research”, European Journal of Social Psychology, Vol. 31, pp. 467-75. Cable, D.M. and DeRue, D. (2002), “The convergent and discriminant validity of subjective fit perceptions”, Journal of Applied Psychology, Vol. 87, pp. 875-84. Chatman, J.A. (1991), “Matching people and organizations: selection and socialization in public accounting firms”, Administrative Science Quarterly, Vol. 36, pp. 459-84. Dawis, R.V. and Lofquist, L.H. (1984), A Psychological Theory of Work Adjustment, University of Minnesota Press, Minneapolis, MN. Edwards, J.R. (1993), “Problems with the use of profile similarity indices in the study of congruence in organizational research”, Personnel Psychology, Vol. 46, pp. 641-65. Edwards, J.R. (1994), “The study of congruence in organizational behavior research: critique and a proposed alternative”, Organizational Behavior and Human Decision Processes, Vol. 58, pp. 51-100. Edwards, J.R. (1996), “An examination of competing versions of the person-environment fit approach to stress”, Academy of Management Journal, Vol. 39, pp. 292-339. Elizur, D. and Sagie, A. (1999), “Facets of personal values: a structural analysis of life and work values”, Applied Psychology: An International Review, Vol. 48, pp. 73-87. Elizur, D., Borg, I., Hunt, R. and Beck, I.M. (1991), “The structure of work values: a cross-cultural comparison”, Journal of Organizational Behavior, Vol. 12, pp. 21-38. Festinger, L. (1954), “A theory of social comparison processes”, Human Relations, Vol. 7, pp. 117-40. French, J.R.P. Jr, Caplan, R.D. and Harrison, R.V. (1982), The Mechanisms of Job Stress and Strain, Wiley, London. Klein, K.J. and Kozlowski, S.W.J. (2000), Multilevel Theory, Research, and Methods in Organizations: Foundations, Extensions, and New Directions, Jossey-Bass, San Francisco, CA. Kristof, A.L. (1996), “Person-organization fit: an integrative review of its conceptualizations, measurement, and implications”, Personnel Psychology, Vol. 49, pp. 1-49. Kristof-Brown, A.L., Zimmerman, R. and Johnson, E. (2005), “Consequences of individuals’ fit at work: a meta-analysis of person-job, person-organization, person-group, and person-supervisor fit”, Personnel Psychology, Vol. 58, pp. 281-342. Lewin, K. (1935), Dynamic Theory of Personality, McGraw-Hill, New York, NY. Lindell, M.K. and Brandt, C.J. (2000), “Climate quality and climate consensus as mediators of the relationship between organizational antecedents and outcomes”, Journal of Applied Psychology, Vol. 85, pp. 331-48. O’Reilly, C.A., Chatman, J. and Caldwell, D.F. (1991), “People and organizational culture: a profile comparison approach to assessing person-organization fit”, Academy of Management Journal, Vol. 34, pp. 487-516. Sagie, A. and Elizur, D. (1996), “The structure of personal values: a conical representation of multiple life areas”, Journal of Organizational Behavior, Vol. 17, pp. 573-86. Sagie, A., Elizur, D. and Koslowsky, M. (1996), “Work values: a theoretical overview and a model of their effects”, Journal of Organizational Behavior, Vol. 17, pp. 503-14. Salancik, G.R. and Pfeffer, J. (1978), “A social information processing approach to job attitudes and task design”, Administrative Science Quarterly, Vol. 23, pp. 224-53. Schneider, B. (2001), “Fits about fit”, Applied Psychology: An International Review, Vol. 50, pp. 141-52.

Work value fit

201

JMP 22,2

202

Schneider, B., Goldstein, H.W. and Smith, D.B. (1995), “The ASA framework: an update”, Personnel Psychology, Vol. 48, pp. 747-73. Simmering, M.J., Colquitt, J.A., Noe, R.A. and Porter, C.O.L.H. (2003), “Conscientiousness, autonomy fit, and development: a longitudinal study”, Journal of Applied Psychology, Vol. 88, pp. 954-63. Suls, J., Martin, R. and Wheeler, L. (2000), “Three kinds of opinion comparison: the triadic model”, Personality and Social Psychology Review, Vol. 4, pp. 219-37. Sweeney, P.D. and McFarlin, D.B. (2004), “Social comparisons and income satisfaction: a cross-national examination”, Journal of Occupational and Organizational Psychology, Vol. 77, pp. 149-54. Thomas, J.G. and Griffin, R.W. (1989), “The power of social information in the workplace”, Organizational Dynamics, Vol. 18, pp. 63-75. Van Vianen, A.E.M. (2000), “Person-organization fit: the match between newcomers’ and recruiters’ preferences for organizational cultures”, Personnel Psychology, Vol. 53, pp. 113-49. Van Vianen, A.E.M. (2005), “A review of person-environment fit research: prospects for personnel selection”, in Evers, A., Anderson, N. and Voskuijl, O. (Eds), Handbook of Selection, Blackwell Publishing, Oxford, pp. 419-39. Van Vianen, A.E.M., De Pater, I., Kristof-Brown, A.L. and Johnson, E.C. (2004), “Fitting in: surface- and deep-level cultural differences and expatriates’ adjustment”, Academy of Management Journal, Vol. 47, pp. 687-709. Vandenberghe, C. (1999), “Organizational culture, person-culture fit, and turnover: a replication in the health care industry”, Journal of Organizational Behavior, Vol. 20, pp. 175-84. West, M.A. and Anderson, N.R. (1996), “Innovation in top management teams”, Journal of Applied Psychology, Vol. 81, pp. 680-93. Zohar, D. and Luria, G. (2004), “Climate as a social-cognitive construction of supervisory safety practices: scripts as proxy of behavior patterns”, Journal of Applied Psychology, Vol. 89, pp. 322-33. Corresponding author Annelies E.M. van Vianen can be contacted at: [email protected]

To purchase reprints of this article please e-mail: [email protected] Or visit our web site for further details: www.emeraldinsight.com/reprints

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

When person-organization (mis)fit and (dis)satisfaction lead to turnover

P-O misfit and turnover

The moderating role of perceived job mobility

203

Anthony R. Wheeler Bradley University, Peoria, Illinois, USA

Vickie Coleman Gallagher and Robyn L. Brouer Florida State University, Tallahassee, Florida, USA, and

Chris J. Sablynski California State University, Sacramento, California, USA Abstract Purpose – The present study examined the relationships between P-O fit, job satisfaction, perceived job mobility, and intent to turnover. It was hypothesized that job satisfaction mediated the P-O fit-intent to turnover relationship and that perceived job mobility moderated the job satisfaction-intent to turnover relationship such that the combined effect of high job dissatisfaction and high perceived job mobility predicted intent to turnover. Design/methodology/approach – Data were obtained utilizing a field survey from a sample of 205 full-time employed adults working in two geographic regions in the USA. Participants completed an HTML-based web survey that contained measures of the constructs of interest to this study. Findings – Mediated and moderated regression analyses revealed statistical support for the hypothesized relationships, which were interpreted as evidence that P-O misfit and job dissatisfaction do not necessarily lead to intent to turnover. Research limitations/implications – The potential for common method variability was present in the study, the impact of which could either attenuate or inflate estimated statistical relationships. Practical implications – While P-O fit researchers typically associate misfit with decreased job satisfaction and increased turnover, the present research suggests that intervening variables, such as job mobility, influence employee intentions to turnover. Originality/value – The phenomenon of misfit is understudied in larger context of P-O fit; thus this research represents one of the first studies in this area of research. Keywords Job mobility, Job satisfaction, Employee turnover Paper type Research paper

The empirical findings and the advice to organizations are clear. Increase person-organization (P-O) fit to increase job satisfaction and to decrease intent to turnover, with the converse of those relationships being true, as well (Kristof-Brown et al., 2005; Verquer et al., 2003). Perhaps no other construct studied in organizational research has led so many scholars to draw the same conclusion and offer the same advice. Indeed, from applying to the organization (Saks and Ashforth, 2002) to entering The authors would like to thank Kenneth J. Harris and Laurence G. Weinzimmer for their assistance and comments during the development of this manuscript.

Journal of Managerial Psychology Vol. 22 No. 2, 2007 pp. 203-219 q Emerald Group Publishing Limited 0268-3946 DOI 10.1108/02683940710726447

JMP 22,2

204

into the organization (Judge et al., 2000) to socializing employees within the organization (Cable and Parsons, 2001) to exiting the organization (Chatman, 1991), P-O fit has been used to explain the fundamental psychological process underlying the daily experiences of job applicants, employees, and employers. While nearly two decades of research consistently support the beneficial outcomes associated with increased P-O fit (c.f. Kristof, 1996; Verquer et al., 2003), thus creating the maxim stated above, it is assumed that poor P-O fit necessarily leads to decreased job satisfaction and thus leads to turnover (Wheeler et al., 2005). There is good reason to believe the poor P-O fit – dissatisfaction – turnover sequencing. Two recent meta-analyses provide support for both the P-O fit – job satisfaction and P-O fit – turnover relationships. Verquer et al. (2003) conducted a meta-analysis of 21 P-O fit studies, most of which measured both job satisfaction and intent to turnover. They reported modest mean correlations between P-O fit and job satisfaction and P-O fit and intent to turnover (r ¼ 0:25 and 2 0.18, respectively). Moreover, they also reported that P-O fit predicted minimal explained variability in job satisfaction and intent to turnover (mean R 2 ¼ 0:06 and 0.03, respectively). Kristof-Brown et al. (2005) conducted an expanded meta-analysis of the outcomes of P-O fit and found stronger relationships between P-O fit – job satisfaction and P-O fit – intent to turnover (mean r ¼ 0:44 and 2 0.35, respectively). However, they too reported minimal explained variability of job satisfaction and intent to turnover (mean R 2 ¼ 0:09 and 0.07, respectively). So, while P-O fit and job satisfaction share a strong positive relationship, the negative relationship between P-O fit and intent to turnover is much weaker. Three issues stand out with regards to these meta-analyses. First, P-O fit researchers do not concurrently examine the P-O fit – job satisfaction and P-O fit – intent to turnover relationships. That is, P-O fit researchers tend to examine the independent effects of P-O fit on each outcome. Second, given that meta-analytic summaries of the job satisfaction – intent to turnover report strong negative correlations (mean r ¼ 20:34 in Hellman, 1997; mean r ¼ 20:25 in Tett and Meyer, 1993), coupled with the strong positive correlations between P-O fit and job satisfaction, we find it surprising that the P-O fit – intent to turnover relationship is not stronger. Third, to this point in the study of P-O fit and its commonly associated outcomes, scant research addresses what individuals will do in the event of P-O misfit. These three issues, we feel, can be attributed to the lack of a clear theoretical framework that explains how P-O fit and job satisfaction trigger turnover intentions. Wheeler et al. (2005) delineated a multidimensional theory of fit, and they relied on Lee and Mitchell’s (1994) unfolding model of voluntary turnover to explain how the assessment of P-O fit could lead to job satisfaction and subsequent turnover intentions. Lee and Mitchell (1994) theorized that job dissatisfaction could indeed result in turnover but that it was more likely that other factors would act as precursors to the job dissatisfaction. Wheeler et al. (2005) suggested that the combination of job dissatisfaction and P-O misfit would lead to turnover in so far as the individual perceived viable job alternatives. That is, P-O misfit might indeed lead to job dissatisfaction but unless a poor-fitting, dissatisfied individual believes that other work opportunities exist, that individual will not leave his or her current position. Empirically, however, the topic of P-O misfit has yet to receive examination, which is the exact purpose of the present research.

The present research has four goals. First, we review the literature on the P-O fit[1] – job satisfaction – intent to turnover relationship and apply Lee and Mitchell’s (1994) unfolding model of voluntary turnover to explain the combined effects of P-O fit and job satisfaction on intent to turnover. Second, drawing upon Wheeler et al.’s (2005) theory of multidimensional fit, we examine the moderating influence on perceived job mobility on the P-O fit – job satisfaction – intent to turnover relationship. Third, we empirically test our hypothesized relationships between these variables, which we present in Figure 1. Finally, we discuss the implications of our model and our results.

P-O misfit and turnover

205

Theoretical background, review of literature, and hypotheses development Unfolding model of voluntary turnover Lee and Mitchell (1994) proposed a novel job search theory called the unfolding model of voluntary turnover (UMVT). Historically, turnover researchers viewed two variables as key to understanding why employees voluntarily leave organizations: job satisfaction and perceived job alternatives (Hulin et al., 1985). Mobley (1977) proposed that job dissatisfaction led to a linear series of cognitive evaluations, starting with initial thoughts of leaving the job followed by the comparison between the current job and possible job alternatives, and ending with intentions to leave the organization. Lee and Mitchell (1994) argued that while this linear decision-making process intuitively appeals to many researchers, the equivocal empirical support for these types of models suggests that voluntary turnover was more complex than previously thought. However, the UMVT does not nullify the traditional models of turnover as much as it incorporates and expands these models. Lee et al. (1996) summarize that “factors other than affect can initiate the turnover process, employees may or may not compare a current job with alternatives, and a compatibility judgment . . . may be used” (p. 6). The UMVT is theoretically grounded in Beach’s (1990) image theory. Image theory describes the process of how individuals process information during decision-making.

Figure 1. Hypothesized model of relationships examined in study

JMP 22,2

206

Beach argues that individuals seldom have the cognitive resources to systematically evaluate all incoming information, so individuals instead simply and quickly compare incoming information to more enduring heuristic-type decision-making alternatives. Beach considered this comparative process a default or status quo decision-making process. Lee and Mitchell (1994) reasoned that expected or unexpected dramatic events, or shocks, to an individual’s status quo decision-making process would lead to a series of possible job search paths. Image theory asserts that these existing status quo decision-making processes are context bound. That is, individuals possess mental images that represent specific domains of their lives (e.g. work, family, friends, etc.) that act as behavioral guides for specific environments (Mitchell and Beach, 1990). In decision-making situations, where individuals will scan the environment for information, these idiosyncratic images, which are akin to heuristics, are the default behavioral guides to which all other alternative information is compared (Beach, 1990). The key to understanding the UMVT centers on the fallout from experiencing a shock. In essence, a shock can cause individuals to reassess existing idiosyncratic images, which in some instances will cause image violation (Lee and Mitchell, 1994). That is, in some cases, individuals can receive information that shocks the decision-making process into abandoning existing images for newly created images. Lee and Mitchell (1994) assert that job search is typically a function of shocks, which cause individuals to scrap status quo reasoning (i.e. remaining with the current organization) in lieu of alternative decision-making processes. Lee and Mitchell (1994) identified four alternative decision-making paths by which individuals can travel in the job search process. Path 1 begins with a shock, which causes individuals to scan previous experiences for similarities to the present shock. Should the present shock match a past decision-making event and should the outcome of that previous experience be judged in retrospect as being the correct decision to make, individuals will simply follow the same decision-making process that was successful in the past event. Path 2 describes how a shock leads individuals to reassess their commitment to the organization, and the shocks that trigger the turnover decision-making process cause individuals to assess fit with the organization, even if no other job alternatives are present. Should the shock lead to an assessment of misfit, individuals likely increase intentions to turnover. The shocks resulting in path 3 cause individuals to assess whether or not their commitment could be associated with a different organization. The decision-making process in path 3 requires explicit comparisons between an individual’s current organization and at least one possible alternative. In path 3, the shock-induced assessment of misfit directly leads to increases in job dissatisfaction, which then results in the scanning for possible job alternatives. If the individual believes that a job alternative will not provide better fit than the current job, that individual will remain with their current organization (e.g. stay to avoid image violation). On the other hand, if the individual believes that better fit will be achieved by working for another organization, that individual will likely decide to leave the current organization. Finally, path 4 describes how individuals simply change over time and reassess commitment to an organization. That is, no shock occurs to stimulate job search; however, affective responses to daily organizational life (e.g. commitment and job satisfaction) over time can cause individuals to turnover even without suitable job alternatives. Path 4 most closely resembles the traditionally sequential models of

turnover, with job dissatisfaction leading to search of job alternatives and subsequent intention to turnover. In the present research, we assert that the UMVT, specifically paths 3 and 4, directly relate to assessments of P-O misfit and explain how and why P-O misfit will lead some to leave the organization while others will remain. P-O fit – job satisfaction – intent to turnover relationship Typically, P-O fit researchers theoretically ground their research in terms of Schneider’s (1987) attraction-selection-attrition (ASA) framework. The ASA framework describes how individuals and organizations are differentially attracted to each other. Should individuals and organizations share mutual attraction, they will select each other (in the form of a job offer and an offer acceptance). So long as the organization and the employee remain mutually attracted, the employee will remain with the organization. If either the organization or the employee at some point feels that the employee no longer fits with organization, the employee will leave the organization through involuntary or voluntary means. While much of the attraction literature related to P-O fit supports the ASA framework (Van Vianen, 2000; Wheeler et al., 2005), the ASA framework ultimately predicts that misfit will necessarily lead to turnover (Schneider, 1987; Schneider et al., 1995). In terms of the relationship between P-O fit and job satisfaction, P-O fit researchers theorize that the degree to which an individual’s and organization’s values overlap, termed value-goal congruence (Chatman, 1991), the more satisfied the employee will be in his or her job (Kristof, 1996). In turn, this satisfaction with the job, in continuation with the ASA framework, results in employee retention (Chatman, 1991). On the reverse side, lack of value-goal congruence reduces employee job satisfaction, most likely through violation of employee expectations, which in turn causes employee turnover (Kristof-Brown et al., 2005; Verquer et al., 2003). While this seemingly linear relationship would appear to support the traditional models of turnover (e.g. Mobley, 1977), P-O fit researchers test the P-O fit – job satisfaction and P-O fit – intent to turnover relationships independently instead of concurrently. That is, to support sequential turnover decision-making, P-O fit researchers need to simultaneously examine the combined effects of P-O fit and job satisfaction on intentions to turnover. We believe that the lack of theoretical grounding has prevented empirical tests of these relationships. We assert that the UMVT more comprehensively and parsimoniously predicts the P-O fit – job satisfaction – intent to turnover relationship. In the two empirical tests of the UMVT, Lee et al. (1996) and Lee et al. (1999) reported that of the four paths theorized by the UMVT paths 3 and 4 were the most frequently observed. Specifically, Lee et al. (1996) reported that 32 percent of nurses in their sample used path 3 (shock, assessment of misfit, job dissatisfaction, comparison to alternative) when making the decision to leave their current job, while 41 percent of nurses in their sample followed path 4 (no shock, job dissatisfaction, intent to leave) when making the decision to leave their current job. That is, almost 75 percent of their sample left due to value-goal misfit or general job dissatisfaction. Lee et al. (1999) reported similar findings, with 64 percent of their sample of public accountants following path 3 in the turnover decision-making process and 30 percent following path 4 in the turnover decision-making process. Moreover, Lee et al. (1999) found that the shocks associated with Path 3 decision-making resulted from changes made in the organization, which led to

P-O misfit and turnover

207

JMP 22,2

208

value-goal image violation. They also found the strongest relationship between path 3 and job satisfaction, meaning that organization shocks leading to value-goal image violation strongly predicted job satisfaction. Interpreting the two major meta-analyses on P-O fit in light of the UMVT, the strength of the P-O fit – job satisfaction – relationship becomes clear. The assessment of fit with the organization does strongly predict job satisfaction (Kristof-Brown et al., 2005; Lee et al., 1999; Verquer et al., 2003) because the shocks leading to the assessment of fit are organizationally bound. Moreover, the weaker meta-analytic relationships between P-O fit and intent to turnover (Kristof-Brown et al., 2005; Verquer et al., 2003) are due to the intermediary role of job satisfaction. That is, organization-induced shocks which lead to the assessment of fit proximally influence job satisfaction; and it is the combined effect of P-O fit on job satisfaction that will ultimately predict employee intent to turnover. Thus, we make the following three hypotheses. H1. P-O fit and job satisfaction are positively related, such that participants reporting high levels of P-O fit will also report high levels of job satisfaction. H2. Job satisfaction and intent to turnover are negatively related, such that participants reporting high levels of job satisfaction will also report low levels of intent to turnover. H3. Job satisfaction mediates the relationship between P-O fit and intent to turnover, such that high levels of P-O fit will decrease participant intent to turnover in so far as levels of participant job satisfaction are also high. Job satisfaction – job mobility – intent to turnover relationship P-O fit research assumes a linear relationship between P-O fit and intent to turnover (Wheeler et al., 2005); however, empirical studies show a weak relationship between the two constructs (Kristof-Brown et al., 2005; Verquer et al., 2003). Indeed, the weak relationship suggests that more employees remain in organization despite the lack of P-O fit and the resulting job dissatisfaction associated with P-O misfit. The question becomes, then, why do poor-fitting and dissatisfied employees remain with the organization? Consistent with both traditional sequential models of turnover (e.g. Mobley, 1977) and the UMVT, Wheeler et al. (2005) proposed a model of multidimensional fit that included possible explanations of how employees will behave in the event of misfit. Wheeler et al.’s model proposes that turnover is one of many options available for employees experiencing P-O misfit. Specifically, Wheeler et al. (2005) include perceived job mobility, which is defined as an individual’s perception of available alternative job opportunities, as a key moderating variable between causes of misfit and the decision to turnover. That is, Wheeler et al. incorporates the assessment of fit related organizational shocks found in path 3 of the UMVT and the key component of job satisfaction in paths 3 and 4 of the UMVT; moreover, Wheeler et al. include the traditional turnover variable of job mobility as moderating the relationship between job dissatisfaction and intent to turnover. Should an employee experience job dissatisfaction, either through an organizationally induced shock that causes assessment of P-O fit or through gradual affective decreases, the likelihood of an employee leaving the organization depends on that employee’s perceptions of available job alternatives.

In path 3 of the UMVT, P-O misfit leads to job dissatisfaction, which causes individuals to scan the environment for possible job alternatives. If no suitable job alternatives exist, the individual will remain with the organization. Thus, we expect perceived job mobility to moderate the relationship between job satisfaction and intent to turnover. Moreover, we expect this interaction to explain how high levels of job dissatisfaction coupled with low perceptions of job mobility lead to reduced intent to turnover compared with high levels of both job dissatisfaction and job mobility. Not only is this consistent with path 3 of the UMVT, the most frequently engaged turnover decision-making path, but is also consistent with Path 4 of the UMVT. Non-shock induced job dissatisfaction, as Lee and Mitchell (1994) reasoned, is most similar to the traditional models of turnover in that job dissatisfaction triggers turnover. The added component of job mobility will either ease or limit a dissatisfied employee’s intentions to turnover. Thus we make the following hypothesis. H4. Participant perceived job mobility moderates the relationship between job satisfaction and intent to turnover, such that participants who perceive high levels of job mobility but report low levels of job satisfaction will also report higher levels of intent to turnover compared to participants who perceive low job mobility and low job satisfaction. Method Sample and data collection A total of 205 fulltime employed participants completed a HTML-based web survey (e.g. Websurveyor); however we excluded data collected from seven of the participants due to incomplete responses (n ¼ 5) or responses from participants that were deemed as outliers on several measured variables (n ¼ 2). The criterion used to determine outliers was based on guidelines developed by Weinzimmer et al. (1994), where participant scores on measures greater than three standard deviations from the mean are considered in violation of the statistical assumption of normal distribution of scores. The final sample of participants (n ¼ 198) had a mean age of 39.12 (13.04 years SD), was 52.5 percent female, averaged 7.76 years of organizational tenure (8.35 years SD), and 50.7 percent held at least a four-year undergraduate degree. In terms of racial composition, the sample was predominantly white (72.7 percent), with some representation from participation from individuals of Asian (10.6 percent), African American (6.5 percent), and Hispanic (6.6 percent) decent. Some study participants did not report racial demographics (3.5 percent). We collected data over a two-week period by utilizing students enrolled in three human resources management classes at two universities, one a small private university located in the Midwest region of the US and the other a large public university located in the western region of the US. Utilizing a snowball sampling technique increasingly relied upon by field survey researchers (cf. Ferris et al., 2005; Kolodinsky et al., 2004) students in each class recruited study participants by distributing a survey web-link to fulltime employed adults in exchange for extra credit toward their final grade. To verify that the study participants completed the survey instrument instead of the students, we required study participants to include an email address at where they could be contacted. We randomly sampled 20 percent of all participants and emailed them to ask questions about the survey (e.g. position and organization tenure, total compensation, etc.). If participant responses to these

P-O misfit and turnover

209

JMP 22,2

210

questions matched their responses given on the survey, we considered the data reliable. Based upon this verification process, we found no evidence of fabricated responses. Measures As mentioned, all participants completed our survey instrument on-line. P-O fit researchers have utilized on-line surveys for data collection and have reported no systematic response biases inherent in collection strategy (c.f. Dineen et al., 2002). For ease of completion of the web-based survey, all attitudinal and behavioral items (unless otherwise noted) were anchored on a five-point scale ranging from 1 ¼ “totally disagree” to 5 ¼ “totally agree.” Midpoints (values of 3) were anchored with the word “neutral.” The items in each scale were summed and then averaged to arrive at an overall value for the scale. Higher scores represent higher levels of each of the constructs. P-O fit. Participants reported their subjective perceptions of P-O fit[2] utilizing a total of three items given by Lauver and Kristof-Brown (2001), which they adapted from Cable and Judge (1996). Subjective measures of P-O fit have been found to be the best measures of P-O fit with regards to their predictive validity of fit-related outcomes (Verquer et al., 2003). Items included “My values match or fit the values of the organization,” “I am able to maintain my values at this company,” and “My values prevent me from fitting in at this company because they are different from the company’s values” (reverse scored). Job satisfaction. We measured job satisfaction using three items developed by Cammann et al. (1979). A sample item is “All in all, I am satisfied with my job.” Perceived job mobility. We measured perceived job mobility via two items developed by McAllister (1995). A sample item is “If I were to quit my job, I could find another job that is just as good.” Intent to turnover. We measured participant intent to turnover via Seashore et al.’s (1982) three-item scale. A sample item is “I will probably look for a new job in the next year.” Results Table I presents means, standard deviations, correlations, and internal reliability (Cronbach’s alpha) estimates for variables collect in the present study. To test H1-H3, we conducted hierarchical mediated regression as outlined by Baron and Kenny (1986),

Table I. Means, standard deviations, correlations, and internal reliability estimates for variable included in study

Variable

Mean

SD

1

2

3

4

5

6

7

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

7.76 NA 39.12 12.76 12.41 7.03 6.55

8.34 NA 13.04 1.76 2.36 2.00 3.45

(–) 0.07 0.54 * 2 0.02 0.07 2 0.25 * 2 0.22 *

( –) 0.05 0.08 0.12 20.04 20.05

(–) 0.05 0.13 2 0.19 * 2 0.29 *

(0.76) 0.55 * 0.05 20.46 *

(0.90) 2 0.09 2 0.66 *

(0.83) 0.26 *

(0.90)

Org. tenure Gender Age P-O fit Satisfaction Job mobility Int. to turnover

Notes: n ¼ 198; *Significant p-value at 0.01; Internal reliability (Cronbach’s alpha) appear in parentheses along diagonal; Gender coded 1 ¼ female, 0 ¼ male

and we present the results from this analysis in Table II. In step 1 of the analysis, P-O fit was regressed onto job satisfaction. Consistent with H1, we found a statistically significant relationship between P-O fit and job satisfaction such that as levels of P-O fit increased so did levels of satisfaction. In step 2 of the analysis, P-O fit was regressed onto participant intent to turnover, which also yielded a statistically significant negative relationship between P-O fit and intent to turnover. That is, as participant levels of P-O fit increased, participant intent to turnover decreased. To complete the mediated regression, in step 3 we then regressed job satisfaction onto intent to turnover. In order to find full-mediated regression, the beta weight for the P-O fit – intent to turnover relationship should become non-significant as the beta weight for the job satisfaction – intent to turnover relationship becomes statistically significant. As seen in Table II, we observed this pattern; thus H2 and H3 were supported. As participant job satisfaction increased, participant intent to turnover decreased. Moreover, interpreting the mediated regression results, P-O fit decreases participant intent to turnover in so far as levels of participant job satisfaction remain sufficiently high. That is, the relationship between P-O fit and intent to turnover is not direct. P-O fit must first increase levels of job satisfaction in order to then decrease intent to turnover. To test H4, the moderating influence of perceived job mobility on the job satisfaction – intent to turnover relationship, we conducted a hierarchical moderated regression as outlined by Baron and Kenny (1986). We present the results of this analysis in Table III. Due to significant correlations between participant age and organizational tenure with intent to turnover, which are implied in Schneider’s (1987) ASA framework and empirically supported in Schneider et al. (1995), we included these two demographic variables as control variables in this regression. Therefore, in step 1 of this analysis, we regressed participant age and organizational tenure onto participant intent to turnover. Holding this relationship constant, we then regressed the main effects of participant job satisfaction and perceived job mobility onto intent to turnover in steps 2 and 3 of the analysis, respectively. Finally, in order to test the interaction between job satisfaction and perceived job mobility, we entered the interaction term at step 4 of the analysis. As seen in Table III, we found statistical support for the model at each step of the analysis, including statistically significant changes in the R 2 values. To interpret the direction of the interaction, we graphed the interaction utilizing Aiken and West’s (1991) guidelines for interpreting interactions (e.g. graphing one standard deviation above and below the median score of each

Dependent variable ! Predictors P-O fit Job satisfaction R-squared Adj. R-squared F

Model 1 Satisfaction

Model 2 Int to Turn

Model 3 Int to Turn

0.55 * *

20.44 * *

2 0.12 2 0.60 * *

0.2 0.19 48.7 * *

0.45 0.44 79.1 * *

0.3 0.3 86.6 * *

Notes: Standardized beta coefficients are provided; n ¼ 198; *p , 0.05, * *p , 0.01

P-O misfit and turnover

211

Table II. Mediated regression results for H1-H3

JMP 22,2 Control variables Age Organization tenure

212

Model 1

DV ¼ intent to turnover Model 2 Model 3

0.24 * * 20.10

20.15 * * 20.10

20.14 * 20.06

2 0.15 * 2 0.05

20.63

20.62 * *

2 0.37 *

0.17 * *

0.57 *

Independent variable Job satisfaction Moderating variable Job mobility Interaction term Satisfaction £ job mobility

Table III. Moderated regression results for H4

R-squared Adj. R-squared Change R-squared F df

Model 4

20.44 * 0.09 0.08 0.09 * * 9.5 * * 2,196

0.49 0.47 0.39 * * 145.4 * * 1,195

0.51 0.50 0.03 * 9.71 * 1,194

0.53 0.52 0.01 * 2.96 * 1,193

Notes: Standardized beta coefficients are provided; n ¼ 198; *p , 0.05; * *p , 0.01

predictor on the dependent variable). As Figure 2 depicts, participants perceiving high job mobility that also report low levels of satisfaction are more likely to report increased intent to turnover, especially compared to participants perceiving low job mobility in low job satisfaction situations. Moreover, upon examining the beta weights for each main effect in the presence of the significant interaction, these main effects remained statistically significant. These findings provide strong support H4. Combining the results from H1-H4, we offer the following interpretation. While increases in P-O fit does foster increases in participant job satisfaction, which would typically lead to decreased intent to turnover, this relationship is contingent upon perceived job mobility. Should a participant report low levels of P-O fit and experience subsequent reductions in job satisfaction, that participant is more likely to leave the organization only if that participant believes that he or she has viable alternate employment options. This is consistent with Lee and Mitchell’s (1994) unfolding model of turnover. Moreover, these findings are consistent with Wheeler et al.’s (2005) assertions that misfit in terms of P-O fit does not necessarily lead to employee turnover. Discussion The present study examined the relationship between P-O fit, job satisfaction, perceived job mobility, and intent to turnover. Specifically, the present study explored these variables in the context of misfit. That is, this is one of the first empirical studies to examine what occurs to individuals when they do not fit with an organization. We found that decreases in P-O fit, which led to decreases in job satisfaction, were more likely to result in increases in intent to turnover if the individual also perceived alternative job opportunities. Stated differently, an individual who feels misfit with an

P-O misfit and turnover

213

Figure 2. Interaction between satisfaction and perceived job mobility on intent to turnover

organization will only leave the organization if he or she believes that alterative job opportunities exist. The present study provides empirical support for Lee and Mitchell’s (1994) UMVT and Wheeler et al.’s (2005) comprehensive conceptual research on the outcomes of fit and misfit. The present research makes other notable contributions to the literature. From a theoretical perspective, the use of the UMVT research to account for two decades worth of studies relating P-O fit to both job satisfaction and intent to turnover is notable. Two major meta-analyses of P-O fit yielded interesting results related to the predictive validity of P-O fit on job satisfaction and intent to turnover, namely that P-O fit better predicts job satisfaction than it does intent to turnover (Kristof-Brown et al., 2005; Verquer et al., 2003). We attributed these findings to two causes. First, P-O fit researchers do not have a suitable theoretical framework to study the combined effects of P-O fit and job satisfaction on intent to turnover. Second, because of these theoretical shortcomings, P-O fit researchers test P-O fit – job satisfaction and P-O fit – intent to turnover relationships as independent effects. The UMVT provides that theoretical grounding; moreover, our results confirmed that P-O fit acts to reduce intent to turnover is so far as P-O fit increases job satisfaction. The second large contribution that the present research makes to the literature is the finding that perceived job mobility moderates the relationship between job satisfaction and intent to turnover. In conjunction with our results regarding how P-O fit indirectly

JMP 22,2

214

predicted intent to turnover through job satisfaction, our research showed that the previously held maxim that P-O fit leads to turnover was only true for those employees who perceive alternative job opportunities. Thus, the present research bolsters not only Lee and Mitchell’s (1994) UMVT but it also provides initial support for Wheeler et al.’s (2005) model of misfit. That is, we provide evidence that many employees will choose to stay with an organization in spite of poor P-O fit or job satisfaction because they do not perceive better options. We find this point both interesting and important. Job search researchers increasingly view job search as a function job availability and applicant marketability (Trevor, 2001). In the event of high job demand environments (e.g. low unemployment and high job availability), less marketable employees are likely to search for new jobs as a means of improving status or salary; whereas, in low job demand environments, these less marketable employees are likely to remain with their current jobs (Trevor, 2001). However, highly marketable employees will begin job search regardless of prevailing job availability conditions because their high marketability inoculates them from prevailing economic conditions. In terms of our study, it could be argued that highly marketable employees are most susceptible to the P-O misfit – job dissatisfaction – intent to turnover path because it is these highly skilled employees who will always perceive greater job mobility. Our findings potentially leave organizations without practical advice relating to the topic of P-O fit. Although almost two decades of research consistently report the beneficial outcomes of P-O fit, meta-analytic findings provide evidence that P-O fit minimally accounts for predicted levels of either job satisfaction or intent to turnover. Moreover, in the present research, we found that poor fitting and dissatisfied participants would choose to remain in the organization if they had no other opportunities. What, then, does this mean about the construct of P-O fit if it does not account for explained variability of important outcomes and that it does not explain how poor fitting employees will behave in the face of misfit? We believe that P-O fit should be viewed in the context of a larger psychological construct called job embeddedness (JE) (Mitchell et al., 2001). JE describes how employees become enmeshed not just with organizations but also with the communities in which employees live; furthermore, JE focuses why people stay in organizations as opposed to why people leave (Mitchell et al., 2001). Mitchell et al. (2001) theorized that JE had two components, organizational JE and community JE; and each dimension consists of three factors: fit, links, and sacrifice. Organizational JE fit describes the congruence between an individual’s “personal values, career goals, and plans for the future . . . with the larger corporate culture and the demands for his or her immediate job” (Mitchell et al., 2001, p. 1104), with community JE fit describing the individual’s congruence with the community. Links “are characterized as formal or informal connections between a person and institutions or other people” (Mitchell et al., 2001, p. 1104), which includes links with the community. Sacrifice describes the “perceived cost of material or psychological benefits that may be forfeited by leaving the job . . . or the community” (Mitchell et al., 2001, p. 1105). While P-O fit, in terms of value congruence, is included in the construct of JE, the role that P-O fit plays in retention represents just one aspect of something as complex as employee retention. Early empirical tests of JE are positive (Mitchell et al., 2001; Lee et al., 2004). Importantly, Lee et al. (2004) found that organizational JE, the combined effect of fit,

links, and sacrifice, predicted employee performance; however, they found that community JE predicted employee retention better than organizational JE. This suggests that while P-O fit plays an important role in performance, organizations concerned with retention should focus resources on programs such as employee assistance programs (EAPs), which are “formal programs that assist employees with personal problems that may be affecting their work-related behaviors” (Spell and Blum, 2005, p. 1125). Furthermore, Wayne et al. (1997) found that perceived organizational social support, such as EAPs, did in fact reduce employee turnover intentions. While organizational scholars sometimes prefer to recommend job-based solutions to organizational problems like turnover and job satisfaction, we are now beginning to better understand that these job-based solutions affect performance. To decrease employee turnover, organizations should maybe focus on off the job solutions for on the job problems. This might explain why P-O misfit might not always lead to turnover, as turnover might have little to do with the organization itself. Future directions The present research has numerous research implications. First, as stated above, future research needs to investigate the relationship between employee marketability and perceived job mobility in the context of P-O fit, job satisfaction, and intent to turnover. Wilk and Sackett (1996) found that highly skilled employees do in fact perceive more job opportunities, so it is important to understand how P-O fit influences these highly skilled employees, especially since it is these highly skilled employees that organizations cannot afford to lose. Second, Wheeler et al. (2005) proposed numerous outcomes related to employee misfit, and they proposed that misfit is not just limited to P-O fit. More and more fit researchers are examining the combined effects of multiple conceptualizations of fit (Kristof-Brown et al., 2005). Wheeler et al.’s (2005) model posited that multiple dimensions of fit allow employees to buffer misfit in any one conceptualization of misfit. They also proposed that should misfit occur across dimensions of fit, in the event that job alternatives are undesirable, misfitting employees will potentially engage in several behaviors in order to survive in an organization in which they do not fit and are dissatisfied. Among these alternatives, Wheeler et al. proposed that some employees might engage in periods of inactivity, where they disengage from the job and the workplace. Poor fitting employees might also become more vocal in an attempt to change the organization in a way that might lead to greater fit. The area of misfit is wide open to researchers, and we offer just these few future research opportunities. Limitations of research The present research has limitations that should be noted. From an empirical perspective, we collected cross-sectional data. This limits our ability to interpret our results. While we do acknowledge this limitation, we also note that we derived our sample from a large pool of participants instead of from a single pool of participants (e.g. a single organization). That is, if we had sampled employees from a single organization, the culture of that organization and historical confounds inherent in that organization would have limited our ability to interpret our results. As our sample was drawn from participants working for a wide array of organization, historical confounds associated with cross-sectional data (Cook and Campbell, 1979) was limited.

P-O misfit and turnover

215

JMP 22,2

216

A second data collection limitation should be noted. As we utilized a single method for collecting our data (e.g. a survey), some might argue that common method bias might inflate correlations between variables (see Kristof-Brown et al., 2005, for a summary related to P-O fit research). While we acknowledge the possibility of common method bias, we point to recent research on this topic that examines this frequently cited limitation to survey research (Spector, 2006). Spector noted that common method bias can either attenuate or inflate correlations among variables, so that to assume common method bias only inflates correlations ignores the potential attenuating, and thus conservative, effects of common method bias. Second, should common method bias account for significant correlations among variables measured via a single method, the bias would reveal itself in the form of significant correlations among all variables measured by that method. Quickly glancing at Table I of the present study, we note that some of the variables collected via survey in our study do not correlate at all. Moreover, as we controlled for the effects of two demographic variables in our moderated regression analysis without significantly altering our results, this further suggests that common method bias alone could not have accounted for our significant results. Thus, we feel confident that common method bias likely does not account for our significant findings. The last limitation we note is that we were unable to gather actual turnover information or actual participant marketability data. That is, collecting these variables would entail gathering data from numerous organizations across the country, and we did not find this feasible for our study. Thus, we relied upon proxies of these constructs, namely intent to turnover and perceived job mobility. Fortunately, these variables are commonly utilized in organizational studies, and we hope to collect this more objective data in future studies. Conclusion We have attempted to explore the relationships between P-O fit and two heavily studied outcomes of P-O fit, job satisfaction and intent to turnover. Our research suggests that these relationships are more complex than previously thought. We asserted and found support for our hypothesis that P-O fit indirectly predicted intent to turnover by more directly predicting employee job satisfaction. Moreover, we asserted that these relationships are influenced by perceived job mobility. In doing so, we have attempted to explain why P-O misfit and job dissatisfaction do not always predict employee intent to turnover. While more research needs to be conducted, our research provides an intriguing starting point for researchers interested in the concept of misfit. Notes 1. Kristof (1996) outlines two conceptualizations of P-O fit. Supplementary fit describes when an individual offers characteristics to the work environment that are currently found in that environment, and complementary fit describes when an individual adds new characteristics to the organization that are not currently present. Kristof (1996) further describes several operationalizations of P-O fit, including value-goal congruence between individuals and organizations. The value-goal congruence operationalization of P-O fit is considered a supplementary fit conceptualization. We ascribe to this supplementary conceptualization of P-O fit.

2. Verquer et al. (2003) found that P-O fit researchers measure the supplementary fit conceptualization of P-O fit utilizing three distinct methods: subjective measures, profile comparison measures, and objective measures. Their findings indicate that subjective measures of P-O fit yield the strongest effect sizes; thus we sought to capitalize on this finding by using a subjective measure in our study.

P-O misfit and turnover

References Aiken, L. and West, S. (1991), Multiple Regression: Testing and Interpreting Interactions, Sage, Newbury Park, CA. 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. Beach, L.R. (1990), Image Theory: Decision Making in Personal and Organizational Contexts, Wiley, Chichester. Cable, D.M. and Judge, T.A. (1996), “Person-organization fit, job choice decisions, and organizational entry”, Organizational Behavior and Human Decision Processes, Vol. 67 No. 3, pp. 294-311. Cable, D.M. and Parsons, C.K. (2001), “Socialization tactics and person-organization fit”, Personnel Psychology, Vol. 54, pp. 1-23. Cammann, C., Fichman, M., Jenkins, D. and Klesh, J. (1979), The Michigan Organizational Assessment Questionnaire, University of Michigan, Ann Arbor, MI. Chatman, J. (1991), “Matching people and organizations: selection and socialization in public accounting firms”, Administrative Science Quarterly, Vol. 36, pp. 459-84. Cook, T.D. and Campbell, D.T. (1979), Quasi-experimentation: Design and Analysis Issues for Field Settings, Houghton Mifflin Co., Boston, MA. Dineen, B.R., Ash, S.R. and Noe, R.A. (2002), “A web of applicant attraction: person-organization fit in the context of web-based recruitment”, Journal of Applied Psychology, Vol. 87, pp. 723-34. Ferris, G.R., Treadway, D.C., Kolodinsky, R.W., Hochwarter, W.A., Kacmar, C.J., Douglas, C. and Frink, D.D. (2005), “Development and validation of the political skill inventory”, Journal of Management, Vol. 31, pp. 126-52. Hellman, C.M. (1997), “Job satisfaction and intent to leave”, Journal of Social Psychology, Vol. 137, pp. 677-89. Hulin, C.L., Roznowski, M. and Hachiya, D. (1985), “Alternative opportunities and withdrawal decisions: empirical and theoretical discrepancies and an integration”, Psychological Bulletin, Vol. 97, pp. 233-50. Judge, T.A., Higgins, C.A. and Cable, D.M. (2000), “The employment interview: a review of recent research and recommendations for future research”, Human Resource Management Review, Vol. 10, pp. 383-406. Kolodinsky, R.W., Hochwarter, W.A. and Ferris, G.R. (2004), “Nonlinearity in the relationship between political skill and work outcomes: convergent evidence from three studies”, Journal of Vocational Behavior, Vol. 65, pp. 294-308. Kristof, A.L. (1996), “Person-organization fit: an integrative review of its conceptualizations, measurement, and implications”, Personnel Psychology, Vol. 49, pp. 1-49. Kristof-Brown, A.L., Zimmerman, R.D. and Johnson, E.C. (2005), “Consequences of individuals’ fit at work: a meta-analysis of person-job, person-organization, person-group, and person-supervisor fit”, Personnel Psychology, Vol. 58, pp. 281-342.

217

JMP 22,2

218

Lauver, K.J. and Kristof-Brown, A. (2001), “Distinguishing between employees’ perceptions of person-job and person-organization fit”, Journal of Vocational Behavior, Vol. 59 No. 3, pp. 454-70. Lee, T.W. and Mitchell, T.R. (1994), “An alternative approach: the unfolding model of voluntary turnover”, Academy of Management Review, Vol. 19, pp. 51-89. Lee, T.W., Mitchell, T.M., Wise, L. and Fireman, S. (1996), “An unfolding model of voluntary employee turnover”, Academy of Management Journal, Vol. 39, pp. 5-36. Lee, T.W., Mitchell, T.R., Holtom, B.C., McDaniel, L.S. and Hill, J.W. (1999), “The unfolding model of voluntary turnover: a replication and extension”, Academy of Management Journal, Vol. 42, pp. 450-62. Lee, T.W., Mitchell, T.R., Sablynski, C.J., Burton, J.P. and Holtom, B.C. (2004), “The effects of job embeddedness on organizational citizenship, job performance, volitional absences, and voluntary turnover”, Academy of Management Journal, Vol. 47, pp. 711-22. McAllister, D.J. (1995), “Affect- and cognition-based trust as foundations for interpersonal cooperation in organizations”, Academy of Management Journal, Vol. 38, pp. 24-59. Mitchell, T.R. and Beach, L.R. (1990), “‘Do I love thee? Let me count’ toward an understanding of intuitive and automatic decision making”, Organizational Behavior and Human Decision Processes, Vol. 47, pp. 1-20. Mitchell, T.R., Holtom, B.C., Lee, T.W., Sablynski, C.J. and Erez, M. (2001), “Why people stay: using job embeddedness to predict voluntary turnover”, Academy of Management Journal, Vol. 44, pp. 1102-21. Mobley, W.H. (1977), “Intermediate linkages in the relationship between job satisfaction and employee turnover”, Journal of Applied Psychology, Vol. 62, pp. 237-40. Saks, A.M. and Ashforth, B.E. (2002), “Is job search related to employment quality? It all depends on the fit”, Journal of Applied Psychology, Vol. 87, pp. 646-54. Schneider, B. (1987), “The people make the place”, Personnel Psychology, Vol. 40, pp. 437-53. Schneider, B., Goldstein, H.W. and Smith, D.B. (1995), “The ASA framework: an update”, Personnel Psychology, Vol. 48, pp. 747-73. Seashore, S.E., Lawler, E.E., Mirvis, P. and Cammann, C. (1982), Observing and Measuring Organizational Change: A Guide to Field Practice, Wiley, New York, NY. Spector, P.E. (2006), “Method variance in organizational research: truth or urban legend?”, Organizational Research Methods, Vol. 9, pp. 221-32. Spell, C.S. and Blum, T.C. (2005), “Adoption of workplace substance abuse prevention programs: strategic choice and institutional perspectives”, Academy of Management Journal, Vol. 48, pp. 1125-42. Tett, R.P. and Meyer, J.P. (1993), “Job satisfaction, organizational commitment, turnover intention, and turnover: path analyses based on meta-analytic findings”, Personnel Psychology, Vol. 46, pp. 259-93. Trevor, C.O. (2001), “Interactions among actual ease-of-movement determinants and job satisfaction in the prediction of voluntary turnover”, Academy of Management Journal, Vol. 44, pp. 621-38. Van Vianen, A.E.M. (2000), “Person-organization fit: the match between newcomers’ and recruiters’ preferences for organizational culture”, Personnel Psychology, Vol. 53, pp. 113-49. Verquer, M.L., Beehr, T.A. and Wagner, S.H. (2003), “A meta-analysis of relations between person-organization fit and work attitudes”, Journal of Vocational Behavior, Vol. 63, pp. 473-89.

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, pp. 82-111. Weinzimmer, L.G., Mone, M.A. and Alwan, L.C. (1994), “An examination of perceptions and usage of regression diagnostics in organization studies”, Journal of Management, Vol. 20, pp. 179-92. Wheeler, A.R., Buckley, M.R., Halbesleben, J.R., Brouer, R.L. and Ferris, G.R. (2005), “‘The elusive criterion of fit’ revisited: toward an integrative theory of multidimensional fit”, in Martocchio, J. (Ed.), Research in Personnel and Human Resource Management, Vol. 24, Elsevier/JAI Press, Greenwich, CT, pp. 265-304. Wilk, S.L. and Sackett, P.R. (1996), “Longitudinal analysis of ability-job complexity fit and job change”, Personnel Psychology, Vol. 49, pp. 937-67. Corresponding author Anthony R. Wheeler can be contacted at: [email protected]

To purchase reprints of this article please e-mail: [email protected] Or visit our web site for further details: www.emeraldinsight.com/reprints

P-O misfit and turnover

219