Cooperation in Groups: Procedural Justice, Social Identity, and Behavioral Engagement 1841690066, 9781841690063

Citation preview

COOPERATION IN GROUPS

Essays in Social Psychology General Editors: MAHZARIN BANAJI, Yale University, and MILES HEWSTONE, University of Cardiff

Essays in Social Psychology is designed to meet the need for rapid publication of brief volumes in social psychology. Primary topics will include social cognition, interpersonal relationships, group processes, and intergroup relations, as well as applied issues. Furthermore, the series seeks to define social psychology in its broadest sense, encompassing all topics either informed by, or informing, the study of individual behavior and thought in social situations. Each volume in the series will make a conceptual contribution to the topic by reviewing and synthesizing the existing research literature, by advancing theory in the area, or by some combination of these missions. The principal aim is that authors will provide an overview of their own highly successful research program in an area. It is also expected that volumes will, to some extent, include an assessment of current knowledge and identification of possible future trends in research. Each book will be a self-contained unit supplying the advanced reader with a wellstructured review of the work described and evaluated. Published titles Dweck: Self-Theories Gaertner and Dovidio: Reducing Intergroup Bias Sorrentino and Roney: The Uncertain Mind Van der Vliert: Complex Interpersonal Conflict Behaviour

Titles in preparation Bodenhausen and Macrae: Stereotype Use Carnevale: The Psychology of Agreement Kruglanski: The Psychology of Closed-Mindedness Mackie: Emotional Aspects of Intergroup Perception Semin and Fiedler: The Linguistic Category Model

COOPERATION IN GROUPS Procedural Justice, Social Identity, and Behavioral Engagement

Tom R. Tyler and Steven L. Blader New York University

Essays in Social Psycology

USA

Publishing Office:

PSYCHOLOGY PRESS A member of the Taylor & Francis Group 325 Chestnut Street Philadelphia, PA 19106 Tel: (215) 625-8900 Fax: (215) 625-2940

Distribution Center:

PSYCHOLOGY PRESS A memba of the Taylor & Francis Group 7625 Empire Drive Florence, KY 41042 Tel: 1-800-634-7064 Fax: 1-800-248-4 724

UK

PSYCHOLOGY PRESS A member of the Taylor & Francis Group 27 Church Road Hove E. Sussex, BN3 2FA Tel: +44 (0) 1273 207411 Fax: +44 (0) 1273 205612

COOPERATION IN GROUPS: Procedural Justice, Social Identity, and Behavioral Engagement Copyright © 2000 Taylor & Francis. All rights reserved. Printed in the United States of America. Except as permitted under the United States Copyright Act of 1976, no part of this publication may be reproduced or distributed in any form or by any means, or stored in a database or retrieval system, without prior written permission of the publisher. I 2 3 4 5 6 7 8 9 0

Printed by Braun-Brumfield, Ann Arbor, MI. 2000. Cover design by Rob Williams. A CIP catalog record for this book is available from the British Library. The paper in this publication meets the requirements of the ANSI Standard 239.48-1984 (Permanence of Paper).

0

Library of Congress Cataloging-in-Publication Data Cooperation in groups : procedural justice, social identity, and behavioral engagement I edited by Tom R. Tyler and Steven L. Blader. p. em. -- (Essays in social psychology) Includes bibliographical references and index. ISBN 1-84169-006-0 (alk. paper) I. Social groups. 2. Social interaction. 3. Human behavior. 4. Social psychology. Tyler. Tom R. 11. Blader, Steven L. Ill. Series.

1.

HM716.C66 2000 302.3'4--dc21 00-032354 ISBN 1-84169-006-6(case) ISSN 1367-5826

CONTENTS

About the Authors Acknowledgments

vii ix

I. OVERVIEW

Chapter

1

Introduction

Chapter

2

The Design of This Study

17

. II. THE ANTECEDENTS OF COOPERATIVE GROUP BEHAVIOR

Chapter

3

Why Study Cooperative Behavior in Groups?

23

Chapter

4

Instrumental Motivations for Engaging in Cooperative Behavior

35

Internally Driven Cooperative Behavior

51

Chapter

5

Ill. THE INFLUENCE OF JUSTICE: PROCEDURAL JUSTICE AND COOPERATION

Chapter

6

The Influence of Justice-Based Judgments

69

Chapter

7

Procedural Justice and Cooperative Behavior

77

IV. THE MEANING OF PROCEDURAL JUSTICE: THE FOUR-COMPONENT MODEL

Chapter

8

Relational Models of Procedural Justice

Chapter

9

A Two-Component Model of Procedural Justice: Quality of Decision Making and Quality of Treatment

89

103

v

vi

Contents

Chapter

v.

10

Creating a Four-Component Model of Procedural justice: Adding the Distinction Between Formal and Informal Sources of justice

125

SOCIAl IDENTITY AND COOPERATIVE BEHAVIOR: STATUS AND PSYCHOLOGICAl ENGAGEMENT :. ~~;.:

Chapter

11

Social Identity and Cooperative Behavior

143

Chapter

12

justice and Group Status: The Antecedents of Status Evaluations

169

Psychological Engagement with the Group

179

Chapter

13

VI. CONCLUSION

Chapter

14

Understanding Group Behavior from a Noninstrumental Perspective

189

References

201

Appendix: Employee Satisfaction Questionnaire

215

Author Index

225

Subject Index

231

ABOUT THE AUTHORS

Tom R. Tyler is Professor of Psychology at New York University. He is also an Adjunct Professor at the New York University School of Law. His work explores authority relations in legal, political, and managerial groups. His past books include: The Social Psychology of Procedural Justice (Plenum 1988, with E. Allan Lind); Why People Obey the Law (Yale, 1990); Trust in Organizations (Sage, 1996, with Roderick Kramer); Social Justice in a Diverse Society (Westview, 1997, with Robert Boeckmann, Yuen Huo, and Heather Smith); The Psychology of the Social Self (Erlbaum, 1999, with Roderick Kramer and Oliver John); Cooperation in Modern Society (Routledge, 2000, with Mark Van Vugt, Mark Snyder, and Anders Biel) and Social Influences on Ethical Behavior in Organizations (Erlbaum, 2001, with John Darley and David Messick). Steven L. Blader is a doctoral student in social pyschology at New York University. His research interests involve how people perceive and react to procedural justice. He has presented his work at numerous conferences, including the annual meetings of the Academy of Management, the Society for Industrial/ Organizational Psychology, and the International Society for Justice Research.

vii

ACKNOWLEDGMENTS

The authors would like to thank Deborah Allen-Barber, Ramona Bobocet John Darley, Geraldine Downey, Wayne Kerstetter, and E. Allan Lind for comments on the manuscript. The authors also benefited from feedback during presentations of this work at the University of Michigan, Columbia University, the Kurt Lewin Institute of The Netherlands, and the Kennedy School of Government at Harvard University. This book was written while Steven Blader was supported by Grant T32-MH19890 from the National Institute of Mental Health.

ix

Introduction

People vary in how much effort they exert on behalf of the groups to which they belong. It probably is easy for most of us to think of people who always are ready to step up and help their group (i.e., to volunteer and to take on responsibility for meeting the group's needs). It may be even easier to think of those who never voluntarily take on any added obligations, and who are generally uninvolved in their groups. What predicts these differences in people's behavior in groups? Why do some people engage more in the groups to which they belong than do others? In this book, we present and test a psychological model for understanding why people cooperate with the groups, organizations, or societies to which they belong. In other words, we are concerned about exploring the factors that shape cooperative behavior within groups. While many factors may be important in shaping people's behavior, we focus specifically on people's motivations. Individuals usually have considerable latitude to determine the extent and nature of their behavioral engagement with the groups to which they belong. They may choose to expend a great deal of effort on promoting the goals and functioning of the group, they may take a passive role, or they may engage in behaviors targeted toward harming the group and its goals. The choices people make regarding their behavioral engagement with the group have important implications for the group's functioning and viability. Our goal is to understand the factors that influence these choices. In order to understand the phenomena we are concerned with, consider an example from a type of group with which we are familiar: an academic department within a university. There is tremendous variation in the way that employees can act within such a work setting. Many faculty and staff take a generally passive stance toward their work. They do their jobs-no more and no less. Others put in extra energy by coming in on weekends, staying late at night, helping other people's students and, in many other ways, doing things that help the university to function more effectively. Most of us can think of the one or two people within any office or department who generally are viewed as making the group operate

1

2

Cooperation in Groups

smoothly because of their exceptional willingness to go beyond the formal requirements of their jobs. Likewise, many of us may be familiar with workplaces where the overall level of cooperative behaviors is higher than it is elsewhere. The forces that drive these differences in motivation between individuals and social settings are the phenomena we seek to understand in this book. We recently experienced this distinction ourselves when our computer malfunctioned while we were trying to complete a talk for an upcoming conference. One computer technician began to help fix the problem, but then noticed that it was approaching 5 p.m.-time to quit! He expressed sadness, but told us we would have to wait for repairs. As it turned out, this was not a problem. Another technician stepped forward and volunteered to stay late. In fact, he called a friend and postponed a dinner engagement until our problem was solved. This behavior was not a required part of the technician's job, but illustrates the willingness of one person to cooperate to promote the overall success of the group. Such cooperative efforts on the part of individual group members not only are important in work groups; they matter in a wide variety of types of groups. Consider the example of communities. Neighborhoods and cities depend heavily on the willingness of some citizens to take on civic responsibilities, ranging from participation in neighborhood watches and charity drives to involvement in government committees. Most of these activities involve issues or problems that interest the citizens involved, but their participation aids the general community more broadly. Hence, while this discussion will focus on cooperation in work settings, we mean to be making a more general argument about the benefits of cooperation within all types of groups. Cooperative behavior can take many forms, all of which help to make groups more effective and viable. Groups simply function better when the people within them engage in prosocial acts of cooperative behavior that help the group. These prosocial behaviors help the group irrespective of why people engage in them. We do not know why the technician in the example above was willing to stay late and help us. He may have been motivated by his loyalty to the department, by his personal liking of us, or by the hope that we would recommend him for a promotion. Regardless of his motivation, the group benefited from this behavior, since we moved forward and prepared our paper in a timely fashion. Although cooperative behavior of the type we have outlined is valuable irrespective of why people choose to engage in it, our goal here is to explore the reasons that people have for helping or not helping the groups to which they belong. In other words, our concern here is with understanding the motivations that underlie the willingness to cooperate. Through such an understanding, we can better explain: (a) why it is that people value group membership and (b) what groups mean to individuals. Based on this information, groups may be able to promote initiatives that foster greater engagement by people in their groups. Furthermore, considering these underlying motivations is valuable because it facilitates our comprehension of human behavior more generally. Before launching into our analysis of cooperation with groups, it is important to clarify what we mean by the term cooperation. The term cooperation is used in various research areas of psychology. In the negotiation and decision-making lit-

Introduction

3

eratures, for example, the term cooperation is opposed to the idea of competition. Negotiations are viewed as mixed motive situations in which both cooperative and competitive motivations are involved. Cooperation develops from similarity of preferences while competition develops from divergent preferences. This conceptualization of cooperation refers to behavior that jointly benefits both parties (Pruitt & Carnevale, 1993). So, if a person feels that there is an agreement in which both he or she and the other party will gain something at the same time, he or she will cooperate in the sense of taking actions designed to make that agreement happen. These might involve making concessions to the other person or trying to problem solve. The relationship between a person and a group to which he or she belongs does not involve cooperation versus competition. Instead, people have to make a decision about how actively to engage themselves in the group by taking actions that will help the group to be effective and successful. This type of cooperation is similar to what social psychologists refer to as helping behavior or proactive social behavior (Derlega & Grzelak, 1982). These literatures focus on the degree to which people within groups, organizations, and societies engage in such beneficial behaviors. In other words, we define cooperation in a different sense than do those concerned with mixed motive games. In our terms, cooperation refers to whether or not people act to promote the goals of the group. We do not expect the alternative to this form of cooperation to be competition with the group but, rather, a lack of cooperation. By not cooperating, people are acting on their own and without regard to the group as an entity. When cooperating, however, they are acting to help promote the group and its goals. So, in our sense of the word, cooperation primarily refers to people's behavior vis-a-vis the group. In both senses of cooperation, there are some potential costs to the individual, be it in terms of the potential resources forsaken or in terms of the costs of working on behalf of the group. But the focus of these two senses of cooperation is different, as are the literatures from which they originate. We want to emphasize the point that we are using the word cooperation only as it relates to group members' behavioral decisions about whether or not to act in the interests of the group. Our concept of cooperation is similar to that in the literature on social dilemmas, which recognizes that there is a conflict between individuals' immediate personal or selfish interests and the actions that maximize the interests of the group (Komorita & Parks, 1994). Put differently, social dilemmas can be defined as situations in which the reward or payoff to each individual for a selfish choice is higher than that for a cooperative one, regardless of what other people do; yet all individuals in the group receive a lower payoff if all defect than if all cooperate. (Smithson & Foddy, 1999, pp. 1-2)

Hence, the group urges the person to put aside his or her immediate concerns and to act on behalf of the group by cooperating. The key issue, as in our study, is with the degree to which people engage in cooperative behavior on behalf of their group.

4

Cooperation in Groups

The paradox that underlies the social dilemma literature is that, in the long term, people have an interest in the well-being of the group. Hence, short-term self-interest, if widely acted on, hurts group members. This situation exists in many groups. People's true interests lie in maintaining groups over the long term. For this reason, we would not want to treat self-interest in an oversimplified way. People have both a short-term and a long-term self-interest. The advancement of long-term self-interest may require efforts to limit the pursuit of short-term selfinterest, even by self-interested actors.

D

Types of Cooperation in Groups

Ultimately, our analysis is based on judgments about the type of cooperative behaviors that are wanted from the members of groups. To address this question, we first make a basic distinction between two forms of cooperation.

Two Forms of Cooperative Behavior We differentiate between discretionary cooperative behavior and mandatory cooperative behavior. Both forms of cooperative behavior involve acts of cooperation, but these two forms of cooperation describe different classes of cooperative behavior. Mandated cooperation occurs when people engage in behavior that is dictated or required by group rules or norms. Some rule or policy of the group prescribes the terms and guidelines of the behavior. In contrast, discretionary behavior occurs when people engage in behavior that is not directly required by the rules or norms of group membership. Thus, our first distinction between types of cooperative behavior involves the source of the behavior. Mandatory behaviors originate from external sources (group rules), while discretionary behaviors originate within the group members themselves. For example, carrying out the duties prescribed in one's job description is mandated behavior. In contrast, picking up trash on the office floor or fixing paper jams in the photocopier is something that people may or may not do (unless they are employees for whom these tasks are mandatory). It is worth taking a moment to explain our view that the performance of mandated behaviors can be cooperative in nature, since this notion may not seem intuitive. What is cooperative about doing what one is required to do? As we discussed earlier, group members usually have considerable latitude in how they carry out their group-related tasks. This latitude includes some freedom in the energy that they put into doing what is required and in the quality with which they carry out those tasks. So, for instance, employees can do their jobs in an adequate manner, carrying out their duties as specified without much emphasis on the quality of their work or the nature of the tasks specified in their job descriptions. Alternatively, employees can perform the same mandated behaviors with an emphasis on high performance and with a concern that the tasks be completed in the best manner possible. Since people have the option of performing mandated behaviors in either of these ways, we believe that cooperation with the group is

Introduction

5

a relevant concept to consider when examining mandated behaviors. People are cooperating with the groups to which they belong when they perform mandated behaviors with vigor and zeal, rather than with a focus on what is sufficient. The enthusiasm with which people engage in their jobs has been recognized to be important across a wide variety of settings, even for individuals that work at low level jobs. Consider, for example, recent studies of the work engaged in by the working poor, many of whom have marginal jobs. Newman (1999) noted that, within menial jobs such as being a hotel maid, a central dimension differentiating employees was the degree to which they did or did not "invest themselves in [their] job any more than was necessary to get by" (p. 151). This observation led Newman to study fast food employees (Newman, 1999). Even in these jobs, in which tasks generally are routinized and spelled out by training manuals, managers "cannot run a smooth operation if they have to micromanage every detail of the organization's operation" (p. 176). In this setting, "ninety percent of [a manager's] job involves coaxing his workforce to abide by the dozens of rules the firm imposes over the preparation of food. There are regulations covering virtually every move a worker makes in the production process" (p. 176). One manager interviewed indicated that, while he has the authority to discipline workers who fail to cooperate, [he] has discovered what most managers come to know in time: a willing workforce is much easier to supervise. And as [he] points out, you cannot keep a constant watch on everyone. (p. 179)

So, managers look for employees who are "willing to go the extra mile ... who can be instructed in the general goal-customer satisfaction-and let loose to achieve it" (Newman, 1999, p. 176). As one manager put it: If they develop [properly], it's not less work for you, but the things you cannot see

they will do. You forget something, they will try to work on it. They will let you know ... "Why don't you do it this way? Because I think it is better," so you will always get input from them. (p. 176)

These valuable employees were contrasted to those who "had no intention of doing more than was necessary to get by" (p. 177). As these anecdotes suggest, even jobs that seem defined largely by rules and procedures depend heavily for their success on the motivation of employees. Managers cannot effectively supervise all aspects of job performance, or can do so only by investing large amounts of their time and effort in surveillance and instruction. As a consequence, it is important that the employees be motivated to do their jobs. The distinction between two types of cooperative behavior-mandated and discretionary-is important to us because we expect to find that mandated behavior is motivated to a greater extent by instrumental judgments and concerns. Instrumental judgments involve people's assessments of the likelihood that engaging in cooperative behavior will be rewarded or that failing to engage in cooperative behavior will be punished. In contrast, we believe that discretionary behavior is motivated primarily by people's attitudes and internal values. That

6

Cooperation in Groups

is, we think that discretionary behavior is influenced by a person's sense of what is desirable or right and appropriate to do. One reason we predict that mandated behaviors will be instrumentally motivated is that rewards and punishments most typically are linked to the degree to which people engage in the behaviors that are required by their group. For example, salaries of employees are based on whether and how well they do their jobs. The group monitors how well people perform their required tasks because those are the criteria that define the behaviors which the group expects from the individual. Further, expectations are developed by the group member that doing one's job well will lead to rewards. In contrast, discretionary behaviors are not specified by the group and, hence, typically are not rewarded or punished by the group. The degree to which they are enacted is therefore likely to be more strongly dependent on whether people feel some internal motivation to engage in such discretionary behaviors. These internal motivations develop from attitudes and values, such as feelings about the legitimacy of group authorities or about commitment to the group. These attitudes and values provide people with personal reasons for acting cooperatively, as opposed to extrinsic reasons like the possibility of gaining rewards or the risk of being punished.

The Two Functions of Cooperative Behavior A second dimension along which we distinguish behaviors is between two different functions of cooperative behavior. These involve the role that the behavior plays in the group·s functioning. Some cooperative behavior involves the performance of desired behaviors by group members. For example, job specifications and descriptions tell people what they should do on their jobs to facilitate the goals of their work groups. This class of cooperative behaviors, if performed, advances the group toward achieving its goals. Another way in which people cooperate with groups involves limiting the extent to which they engage in undesirable behaviors. In other words, groups often have rules that restrict or regulate member behavior to ensure that people do not act in ways that hinder the functioning of the group. For example, people need to refrain from stealing from the group or destroying group property. In sum, the second distinction we draw between two functions of cooperative behavior differentiates between those that proactively advance the group's goals and those that limit behaviors which are obstacles to achieving the group's goals. What this suggests is that people cooperate with groups both by doing things that help the group in a proactive manner and by refraining from doing things that hurt the group. We refer to both of these types of behavior (or nonbehavior) as cooperation, since people make choices along each of these dimensions as to whether they are going to do things that help (or do not hurt, as the case may be) the group. The combination of these two dimensions of cooperative behavior leads to the classification shown in Table 1-1. Our goal will be to examine the factors that motivate group members to engage in each of these four types of cooperative behavior.

Introduction

7

TABLE 1-1. Types of cooperative behavior Forms of cooperative behavior

D

Function of the behavior

Mandatory

Discretionary

Promoting the group's goals Limiting behaviors that harm the group

In-role Compliance

Extra-role Deference

Two Types of Motivation for Behavior

There are two basic reasons that people might engage in cooperative behaviors_ These reasons are defined within the basic model of human motivation articulated by Lewin in his writings on field theory (Gold, 1999). One source of motivation is the basic human desire to maximize gain and minimize loss in interactions with others (i.e., people's "instrumental" or "self-interested" motivation). This motivation leads people to shape their actions to be consistent with the environmental contingencies within a particular situation. The assumption underlying instrumental models is that the structure of the group, and of situations within the group, affects the behavior of group members by shaping the contingencies associated with various types of behavior. This assumption about the nature of human motivation leads to management strategies that seek to change behavior within the group by altering the contingencies within particular situations. Reward-driven incentive systems encourage desired behaviors by rewarding those behaviors. Such strategies focus on the social facilitation of behavior that benefits the group, but may not benefit the individual. Punishment-driven sanctioning systems discourage undesired behavior by punishing those behaviors. They focus on the social regulation of behavior that is beneficial individually, but that hurts the group. The roots of these approaches in management theory lie in the work of F. W. Taylor (1911), who argued for the need for clearly specified job tasks managed by highly structured lines of command and motivated by clear and specific systems of incentives and sanctioning. These models continue to dominate thinking within industrial settings (Braverman, 1974; Haslam, in press; Kanigel, 1997; Locke, 1982; Merkle, 1980; Thompson & Warhurst, 1998; Waring, 1991). They are found widely in discussions of "pay for performance," "reinforcement-based leadership," and many other instrumental management models. Increasingly, however, social scientists have recognized the limits of these command-and-control approaches to managing conflict. This is true of organizational theories concerned with the management of employees (Pfeffer, 1994). It also is true of political and legal authorities interested in social regulation {Tyler, 1990) and in the encouragement of voluntary civic behavior (Green & Shapiro, 1994).

8

Cooperation in Groups

All of these researchers suggest that it is difficult to manage groups which rely only on the ability to reward or punish group members. The second type of motivation, originally outlined by Lewin in his discussion of "field theory," is internal motivation (Gold, 1999). Internal motivations develop from within the individual. They reflect the desire of people to behave in particular ways. Such individual predispositions are long term and are seen as a tendency for a person to behave in particular ways across situations and over time. This tendency is distinct from the particular external reward-cost contingencies found in specific situations. Two types of internal motivation will be considered in this analysis. The first are attitudes. Attitudes reflect people's feelings and desires-the things that they want to do. It is expected that these motivational forces will influence the willingness of group members to act out of their own volition in ways that help the group. That is, these motivational forces are voluntary and encourage selffacilitation. · The second type of internal motivational force is a person's values-the things that they feel that they ought to do. People learn values and feelings of obligation, which discourage them from engaging in actions that might be personally beneficial or desirable but that violate group rules. So, a person's moral values or feelings of obligation to the group and its rules tell him or her not to cheat or steal. These forces lead to self-regulatory behavior. In both the case of attitudes and of values, the key difference with internally motivated behavior is that the behavior involved is distinct from environmental contingencies. Instead, it is shaped by internal motivational forces. In that sense, it is voluntary or self-regulating. People help their group because they feel personally committed to group success, and they follow group rules because they feel a responsibility to do so. In either case, the group does not need to shape behavior by creating environmental contingencies which create incentives or risks that impact on people's expected gains and losses. Our argument is that cooperative behavior of each of the four types outlined potentially can be influenced by four motivational factors: attitudes, values, incentives, and sanctions. Although each of these four factors potentially can influence any of the four cooperative behaviors outlined, we expect that the relationship between attitudes, values, and cooperative behavior will be especially strong for discretionary behaviors. Since discretionary behaviors are not linked directly to incentives and sanctions, the extent to which they occur is most likely to be related to the degree to which there are internal motivational reasons for engaging in these cooperative behaviors. Although this study explores the role of these motivations on people's connections to group authorities and institutions, these same two types of motivation also can be seen in other settings. Consider the case of the informal social groups that exist within formal groups. People also may be motivated by their friends and coworkers. Informal social groups can motivate individuals by using incentives or sanctions. For example, people that work harder than is dictated by informal group norms may be informally sanctioned by their group, while those who follow those norms may receive the help of others when they encounter

Introduction

9

problems. In addition, informal groups can motivate individuals by encouraging the development of internal values linked to the standards of the group. In this case, people do not break informal group rules because they have come to agree with the underlying reasons for those rules; that is, they may have internalized the importance of the rules. Further, when people break social rules, they may feel shame that is linked to their internalized sense that other people will disapprove of their actions and think that those actions reflect badly on their character (Scheff, 1988, 1995, 1997). These feelings of shame can serve as a motivation for generating and maintaining voluntary adherence to informal group norms (Braithwaite, 1989). The first goal of our study is to directly examine the influence of attitudes, values, incentives, and sanctions on each of the four types of cooperative behavior that have been outlined.

D

justice and Cooperation

Our second goal is to examine the influence of judgments about the nature of the group on the behavior of group members. Here, our interest is in the dynamics of authority frameworks within groups. In particular, we will examine the role of justice motivations in shaping people's reactions to groups. This issue is theoretically important because it allows us to explore the degree to which people's sense of justice or injustice shapes what they think, feel, and do in groups. If justice motivates thoughts, feelings, and behaviors, then this indicates one important way in which the structure of a group influences the behavior of the people within that group. It does so by shaping their views about the fairness of the group. These findings also point to policy implications. They suggest that we can encourage desirable behavior from the people in groups by creating group frameworks that are experienced as fair. When people are making a decision about how much and in what way they will invest themselves in the group, they make evaluations of group characteristics. We will compare the role of three such judgments in encouraging supportive attitudes and values and in stimulating cooperative behaviors. Those three judgments are: (a) the degree to which the decisions made in the group lead to favorable outcomes for the person, (b) the degree to which the decisions made in the group lead to fair outcomes for the person, and (c) the degree to which decisions in the group are made through fair processes. One common image of people in groups is that their behavior is motivated primarily by the outcomes they receive from those groups. This image (which we will refer to as the self-interest perspective) develops from the assumption, rooted in social exchange theory, that people interact with others as part of an exchange of resources. In that exchange process, people are viewed as being motivated to maximize their gain in resources for themselves or, at least, to ensure that they will receive a fair number of resources relative to others (Thibaut & Walker, 1975). This perspective argues that people will be sensitive to the personal resources they receive during their interactions with the group. From a social exchange

10

Cooperation in Groups

perspective, the extent to which people will cooperate with other people or within groups is a reflection of the degree to which they view cooperating as a way to facilitate the acquisition of personally desired resources. We will compare this image to a procedural justice perspective that suggests people are influenced by their judgments about the fairness or unfairness of the procedures used to make decisions in their group. A core finding in the prior procedural justice literature is that people are motivated by their concerns about the justice or injustice of group decision-making procedures in ways that are distinct from their judgments about their personal self-interest. In fact, prior research has suggested that procedural justice judgments often have more influence on people's attitudes and behaviors than do their assessments of their personal self-interest (Lind & Tyler, 1988; Tyler, Boeckmann, Smith, & Huo, 1997; Tyler & Smith, 1997). We will test this procedural justice argument by examining the degree to which engagement in one's group, as indexed by the willingness to cooperate, is linked to judgments about the justice of group procedures. Further, we will compare the role of procedural justice to the influence of personal gain judgments as predictors of the willingness to cooperate. While our focus will be on the fairness of procedures, there is another aspect of the justice perspective that we also will consider. That is the argument that people are concerned not about maximizing personal gain, but about the fairness of the outcomes that they receive (i.e., a distributive justice perspective). As we have noted, this perspective is suggested by Thibaut and Walker (1975), who viewed distributive justice as the goal people seek in social interactions. It also is central to discussions of equity theory (Walster, Walster, & Berscheid, 1978). To test this perspective, we will evaluate the degree to which people react to outcome fairness, as opposed to outcome favorability and to procedural fairness. Each of these three perspectives (personal gain, procedural justice, and distributive justice) presents a plausible basis for people's evaluations of their groups. Beyond determining the criteria by which people may evaluate their groups, each of these perspectives suggests a plausible set of predictors concerning what motivates people to act cooperatively. Based on our prior research, we predict that procedural justice judgments will dominate people's relationships with their groups. As will be seen in our review of the procedural justice literature, a wide variety of studies have demonstrated that people's associations with groups, organizations, and societies are strongly linked to their judgments about the fairness of group procedures. These associations include people's internalization of the group's goals, evaluations of group authorities and institutions, willingness to defer to those authorities, and willingness to engage in voluntary cooperative behaviors that aid their group. In other words, we predict that procedural justice judgments will be the primary determinants of group members' attitudes, values, and cooperative behaviors. While prior research findings have linked procedural justice to a variety of attitudes and behaviors, there has been no effort to fully articulate and test an integrated model of the behavioral effects of procedural justice judgments. We

Introduction

11

articulate and test just such a model here. One aspect of our model specifies that the attitudes that are predicted by procedural justice account for some of the observed relationship that exists between procedural justice and cooperative behavior. Further, as we test our model, we will include the possible alternative underpinnings of cooperative behavior described above (distributive justice and personal gain) in order to compare their effects with those of procedural justice. Consistent with our earlier discussion about rewards and mandated behavior, we expect that these outcome-focused constructs may primarily predict mandated cooperative behavior, while discretionary behavior will find its origins in procedural evaluations.

D

The Meaning of Procedural Justice

In addition to examining the importance of procedural justice, we also will be concerned with understanding its meaning. In other words, what do people mean when they indicate that an organization's procedures are fair? We will distinguish among three models that potentially might define the meaning of procedural justice. First, people might define the fairness of procedures in terms of the favorability of their outcomes from the procedure. In this case, fair procedures would be those that lead to outcomes people desire. Second, people might define the fairness of procedures in terms of the fairness of their outcomes from the procedure. In these terms, fair procedures would be those that lead to fair outcomes. These factors already are familiar, since we outlined their potential importance in the Justice and Cooperation section. The third set of elements involves aspects of the way that decisions are made. They define the fairness of procedures through elements of procedures that communicate process-based information. These elements provide information about the procedures of the group. In other words, procedures are evaluated not by the outcomes to which they lead but by the relational information that they convey. We have argued that "relational" criteria include assessments of neutrality, the trustworthiness of the authorities, and the degree to which people are treated with dignity. We will collectively refer to these as the relational elements of procedures (Tyler & Lind, 1992). In this analysis, we will draw on the relational framework of Tyler and Lind (1992) and examine the impact of their three relational constructs. We expect this analysis to replicate prior findings that it is these relational issues which are key to defining the meaning of procedural justice, over and above the influence of either of the outcome assessments described above. Second, we will discuss a new conceptual framework for understanding the meaning of procedural justice that we have outlined in detail elsewhere (Blader & Tyler, 1999a). This new framework is the four-component model of procedural justice shown in Table 1-2. In this new framework, we will distinguish between two aspects of processes: procedural criteria that are related to the quality of decision-making processes and procedural criteria that reflect the quality of the treatment people experience.

12

Cooperation in Groups

TABLE 1-2. The four-component model of procedural justice Source

Procedural element

Quality of decision-making processes Quality of treatment

Rule of the group (Formal)

Actions of supervisor (Informal)

Formal quality of decision making Formal quality of treatment

Informal quality of decision making Informal quality of treatment

Our goal in introducing this new framework is to change the level on which we understand the meaning of procedural justice. The relational model links fairness to particular procedural elements, such as neutrality, but does not organize the elements in terms of an overall conceptual framework. In our new model, we link people's concerns to two core conceptual issues: how decisions are made and how people are treated. By creating basic conceptual categories, we are separating the components of procedural justice from the particular elements of the relational model. We expect that, in all situations, people will have basic concerns about: (a) the quality of the decision-making processes that occur and (b) the quality of the treatment that they receive. Some criteria, such as neutrality and consistency, are particularly linked to how decisions are made. Other criteria, such as respect for rights, are particularly linked to how people are treated. However, these specific elements may be organized in people's minds in different ways within different types of situations. Irrespective of these particular elements, however, we expect that there always will be a factor reflecting quality of decision-making concerns and a factor reflecting quality of treatment concerns. Other elements plausibly might be found to be linked to either type of concern or to both. One example is trustworthiness, which may be linked to the judgment that high-quality decision-making procedures are being used, that people are receiving high-quality treatment during the decision-making process, or both. The degree to which each possibility is true will depend on the situation in which trust is being studied. These models assume that people make distinct judgments about outcome favorability; outcome fairness; and quality of decision-making procedures, and treatment. We will test this assumption. We then will examine the impact of each of these judgments on people's evaluations of the fairness of group procedures, in order to best understand the meaning of procedural justice. We also will compare the utility of our current conceptualization to that of past procedural justice research. In elaborating our model of procedural justice, we will make a further distinction between two sources of experience with procedures. The first source is

Introduction

13

the formal rules of the group. The second is the particular authority to which a group member is responsible or with whom the group member interacts. We expect that people distinguish between these different sources of their experiences. Specifically, we predict that people will respond to agents of the organization in more personal terms than they respond to group rules. These two sources of experience also may differ in other ways, such as their relative influence on different behaviors and attitudes. For instance, since group authorities may be more proximal determinants of group members' experience, they may predict reactions that are based on the immediate environment. At the same time, more formal influences, which tend to be more stable, may predict reactions that are based on views of the group structure. These two dimensions can be considered together in a table of the possible antecedents of procedural fairness (Table 1-2). From our perspective, the key contribution of this model is its creation of a conceptual framework within which procedural elements, such as those identified by the relational model, can be placed. We would expect some elements to be more closely linked to one category, while other elements may be organized differently, depending on the situation. However, in any situation, we would expect people to be concerned about both of these basic categories of concern.

D

The Reasons for Cooperation: Resource Versus Identity Concerns

We also will consider why people cooperate with groups and why people are concerned about issues of procedural justice and injustice from the broader perspective of the images that social psychology has about why people interact with others in groups.* Toward this end, we will consider two possible images regarding the motivations that underlie both concerns about justice and cooperative behavior. The two images we will compare are linked to two basic psychological models that provide distinct images of the motivations which lead people to interact with others and that shape their behavior during such interactions. The first image connects people's justice judgments and cooperative behaviors to their concerns about the favorability or the fairness of the outcomes that they receive from their group. These images are social exchange perspectives on the motives underlying cooperation. They are based on the argument that cooperation develops from the exchange of resources within groups. As we have noted above, social exchange theory argues that people interact with others to exchange resources. If so, then their judgments about their relationship to other people in the group and to the group itself should be dominated by their assessments of the resources gained and lost through participation in a *Our concern is particularly about procedural justice. We have noted elsewhere that the reasons people have for being concerned about procedural and distributive justice are not identical (Tyler, 1994).

14

Cooperation in Groups

group. People ought to define justice through the lens of gains and losses. The exact nature of these outcome-based judgments either can focus exclusively on whether those outcomes maximize an individual's gains or on whether the outcomes themselves are perceived as fair. In either case, the individual is focused on outcomes, and should view all aspects of his or her group membership, including the fairness of procedures, from this vantage point. A variation of this model is rooted in the conception of procedural justice articulated by Thibaut and Walker (1975, 1978)-procedural justice as a mechanism for delivering just outcomes. In an extended discussion about the purpose of focusing on procedural justice, these authors argued that people view the use of fair procedures as the most effective possible way to ensure the attainment of just outcomes. This conceptual viewpoint emphasizes the notion that people define procedural justice according to their evaluation of the control that procedures allow them over evidence presentation and, through that control, over outcomes. By attaining high levels of control, people are able to ensure for themselves the attainment of outcomes with which they are happy. Like the outcome favorability and distributive justice models, this model assumes that people's ultimate concern is with the nature of the outcomes they receive from others. In this case, it is suggested that people are concerned about the fairness of the outcomes they obtain from their group. The models of procedural justice that are based on outcome favorability and outcome fairness both view people's concerns about procedural justice, as well as their cooperative behaviors, as linked to their interest in the outcomes they obtain from groups. People cooperate with groups, according to these models, to secure desired outcomes; their focus on procedural justice is a way of ensuring that the distribution of those outcomes is reasonable and fair. This view can be contrasted to the model based on social identity theory which argues that people interact with others to develop and maintain a group-based social identity (Hogg & Abrams, 1988; Tajfel & Turner, 1979). Unlike the outcomeoriented models outlined above, this model has a fundamentally different perspective on the goals of social interaction. It does not view people as being primarily motivated by the desire to exchange resources. Instead, people are viewed as being motivated to create and maintain a favorable identity. They are aided in that effort by positive information about their relationship with the groups to which they belong. In our prior work, we have used identity-based models as a basis for understanding why people use procedural justice as their central cue in evaluating their relationship to a group, organization, or society (Tyler, Degoey, & Smith, 1996). This approach to understanding why people care about procedural justice is articulated in both our group-value model of procedural justice (Lind & Tyler, 1988) and our relational model of authority (Tyler & Lind, 1992). It can be extended to address the question of why people cooperate in groups (Tyler, 1999a). In either case, the core argument is that people cooperate with groups because those groups perform an important role in sustaining a favorable social identity. The resource-based and identity-based perspectives lead to very different predictions about the factors underlying individuals' cooperative behaviors and

Introduction

15

justice-based judgments. Our prior work leads us to predict that people will care about procedural justice because procedural justice provides information that they view as relevant to their identities, and that these concerns will outpace the influence of resource-based antecedents of procedural justice. If people are focused on identity concerns, we predict that cooperative behavior will flow from status judgments, and procedural justice will be used as an index through which people evaluate their status. Two status judgments are of concern in this study: pride and respect. Pride indicates people's feelings about the status of their group. Respect indicates people's feelings about their status within their group. Both judgments reflect status evaluations that are linked to group membership. As such, they reflect the role of group membership in defining the "social" self-the aspect of the self and self-identity that is linked to the groups to which a person is a member (Tyler, Kramer, & John, 1999). We predict that people will cooperate with groups if they feel proud of group membership and respected by their group. The influence of status judgments will be compared to people's assessments concerning the level of resources they receive from their group. It is a central premise of the identity argument that status judgments impact attitudes, values, and behaviors. In addition, to examine the role of pride and respect in encouraging cooperation, we will explore the role of identification with the group in shaping the same behaviors. Identification with a group reflects psychological engagement with that group-the merger of self and group. People who identify with a group think of themselves in terms of group membership-(e.g., "I am a Harvard professor," "I am a woman"). We will examine the degree to which such identity-based engagement leads to behavioral engagement in the group. We also expect people's justice judgments to be associated with status assessments. From our perspective, the central role that procedural justice plays in shaping values, attitudes, and behaviors reflects the fact that people use procedural justice as a cue to evaluate their status. They use it to evaluate both the status of their group and their status within the group. When group processes are seen as fair, people perceive both the group's status and their relative status within the group to be positive. Consequently, we predict that justice judgments will be linked to direct indicators of status. We also expect to find evidence that procedural justice directly influences people's identities. In other words, we predict not only that procedural justice is used as a measure of status, but that it also has a direct impact on the content of one's identity. To test this argument, we will examine the degree to which fair treatment leads to greater identification with the group. Our concern here is with the degree to which people draw their sense of self from their membership within a particular group (i.e., from their "social identity").

D

The Overall Model

Our overall framework is presented in Figure 1-1. In that conceptual figure, we show how each of the constructs and questions we have presented in this

16

Cooperation in Groups

THE MEANING OF PROCEDURAL JUSTICE

Quality of decisionmaking process Quality of treatment

----/)

PROCEDURAL Jl!STICEAND COOPERATION

CONNECTIONS TO THE GROUP

~10TIVATIONS

TYPES OF COOPERATIVE BEHAVIOR

FOR COOPERATION

Procedural justice

Forms of cooperative behavior Outcome fairness

Outcome favorability

Resources received from the group

Incentives, sanctions

I

FIGURE 1-1. The group engagement model.

chapter relate to one another and how we propose they will come together to determine individual behavior in groups. Although many of the issues we address have been examined in prior research, our goal is to provide a framework that integrates all of these issues within a common conceptual model. We call it the group engagement model because it provides a framework for understanding why people engage in the groups to which they belong. We will explore this question throughout this book and will provide support for the views we have outlined in this introduction.

The Design of This Study

In addition to presenting an overall conceptual framework, we will test the validity of our model within one type of group environment-a work organization. We will perform this test using a study of 404 employees who were asked to make a variety of judgments about their respective work organizations. The study that forms the basis for this book is part of an ongoing research program on cooperation in groups. Two earlier studies also examined this issue using samples of employees (Tyler, 1999a). The first sample included 409 employees living in the Chicago area. Each employee was interviewed over the telephone about their current job. The second sample contained 649 employees in the United States, Hong Kong, and Germany. These employees completed questionnaires presented to them in various ways. Both of these studies focused on particular experiences that employees had with their work supervisors. In contrast, this study focuses on general judgments about the workplace. The various studies of this research program have two common elements. First, they all focus on the micro-level. That is, we are concerned about the individual person, not about the overall structure of the group. Second, we have a subjective focus. We are interested in what is going on inside the heads of the people we study (i.e., their perceptions), as opposed to any "objective" reality of group functioning. We are concerned about people's subjective feelings and thoughts because we believe that those feelings and thoughts shape their behavior in group settings. In this analysis, we are proposing and testing a general conceptual model that we feel applies to all groups, organizations, and societies. Therefore, this model potentially might be tested in a variety of group settings, varying in their formality and nature. The test we present here in a work setting is but one of many possible tests of our model. Our goal is to use it to examine the basic plausibility of our approach. In the long run, of course, the value of the model can be evaluated only by testing its applicability to a wide variety of settings. The reason we are conducting this test in a work setting is that it is widely believed that people are especially instrumental in their relationships to their 17

18

Cooperation in Groups

jobs. Hence, deviations from instrumentality should be especially striking in such a setting. It is more counterintuitive that people would care about justice in their jobs than, for example, in their dealings with the courts or their community, since we believe that justice is central to politics and law. For this reason, we focus our attention on a work setting, as it provides a particularly stringent test of the model we outlined in the previous chapter. The employees used in this study worked in a variety of settings that were chosen to maximize variation in the level of instrumental investment in the work environment. Likewise, people's positions within their groups varied. Some of those who completed questionnaires were low income and part-time employees, whose involvement in their work, and certainly their material benefits, are low. On the other hand, others were long-time and highly compensated executives and technical staff. Each person was asked to complete a questionnaire assessing his or her attitudes, values, and behaviors. For example, they might be asked to indicate whether they agreed or disagreed (agree strongly to disagree strongly) that "I am treated fairly where I work." The exact items asked in the questionnaire, and the response scales used, are shown in the Appendix. Each person in the study completed this questionnaire anonymously. As we examine this data throughout this book, our concern will be with the impact of people's judgments about their group (in this case, their work organization) on their cooperative behavior within the group. Using the results of these employee questionnaires, we test the key premises of our model of cooperative behavior.

D

The Sample of Employees

The sample consisted of 404 employees. Respondents were recruited from a variety of sources and included people approached in public areas (e.g., train stations, parks, outside office buildings) or through the mail, as well as employed undergraduate psychology students. Our general motivation was to try to find a heterogeneous group of employees, which included both high-income, career-track employees and part-time, low-income employees. By assembling such a heterogeneous sample, we attempted to address any concerns that the generalizability of our results was limited to any particular organization or type of employee sampled. People approached to participate in the study were asked to complete a questionnaire about their current job. Respondents were given $10 in appreciation for their participation. Introductory psychology students were given credit toward meeting their research requirement in lieu of the $10. Analysis of the demographic data from the sample confirms that data were collected from a wide variety of employees. Our sample was almost evenly split between the genders, with 55% male respondents and 45% female respondents. The average age for our sample was 30 years old. Almost 40% had completed

The Design of This Study

19

some college, 29% had completed their bachelor's degree, and the remaining 30% had completed some graduate education. People worked for a variety of types of organizations: 35% worked in small offices or stores with only a single location, 19% worked for organizations with several locations, 19% worked for organizations with offices throughout the United States, and another 25% reported working for multinational firms. The median size of our sample's work organizations was 250 employees. Respondents reported a great deal of variation in the income they derived from work. Of those who completed questionnaires, 26% made less than $10,000 a year, 35% made between $10,000 and $40,000, 26% made between $40,000 and $90,000, and 13% made over $90,000. These figures represent the total amount earned from the respondent's work organization, unadjusted for full- or part-time status. The high percentage of low-income workers is primarily attributable to college student part-timers that participated in the study. On average, employees worked about 40 hours, and had been at their jobs for an average of 3 years and 10 months. (Tenure varied greatly. Median tenure was equal to about 2 years.) Much of this demographic data must be interpreted in light of the fact that many (n = 123) of the 404 respondents were undergraduate students who were also employed (part time, in most cases). To assure that these respondents had an adequate basis for answering the survey, we required that they be employed at their current jobs for at least 3 months and that they work at least 20 hours per week. Nevertheless, this explains the high percentage of low-income workers and the high percentage of individuals who had not yet completed their bachelors degree. While this group may have been overrepresented in the sample as compared to the population, we feel that this does not pose a problem for the conclusions we draw from the data. In fact, we believe that the high number of students in the sample actually provides a more stringent test of our hypotheses than would be achieved with a more representative sample of the U.S. workforce. Part-time, student workers tend to not be highly involved in their jobs, and are more likely to have a purely transactional, instrumental relationship with their employer. They work primarily as a means of supporting themselves while they get their education, and few intend to build careers at their current jobs, which often are menial in nature. The relational model underlying cooperative group behavior that we have outlined would seem to apply least to such a group of employees. As a result of this composition of our sample, we note that the findings we report here actually may be a very strict test of our ideas, making the results we will discuss and present all the more compelling.

D

Limitations of the Study

We also think it is important to acknowledge several limits to the study presented. Because of these limits, we are the first to recognize the importance of further tests of the ideas outlined in this model.

20

Cooperation in Groups

The Measurement of Cooperative Behavior First, in this study, people's cooperative behavior in their work organization was assessed through employee self-report. We acknowledge that self-report is an imperfect measure of behavior. It is important to recognize this potential limit to the findings of this study. On the other hand, evidence in prior research on cooperative behaviors suggests that self-report can be an effective and reasonably reliable way to assess people's actions. (See a review of empirical research on this issue in Tyler, 1990.) Ultimately, this model needs to be tested using objective indexes of people's behavior. We think there needs to be such a test of the validity of behavioral self-reports conducted by comparing self-reports of cooperative behavior to the ratings of employee cooperative behavior made by work supervisors or to some other objective measure of employee behavior, such as observation. Further, it is important to note that we have used the procedures that researchers have found to most effectively minimize the biases associated with the self-report of behavior. All questionnaires were completed anonymously. No names were collected. When respondents were paid, they indicated their names and addresses separately on sheets of paper that were not attached to their questionnaires (if they completed the questionnaire via mail) or signed a sheet listing those who had received payment (if they completed the questionnaire in person). It is possible that some mail respondents felt that the researchers might try to connect their names to their questionnaires, since those who returned their questionnaires via mail included both the questionnaire and the form with their name and address in the same envelope. However, none of the respondents indicated where they worked, so the connection to their employer was remote. Further, people who felt uncomfortable with issues of confidentiality were free to return the payment form in a separate envelope, or to decline to complete the form indicating their name (of course, those people could not be paid). All of these caveats aside, there is no substitute for actual behavioral data. For this reason, our more recent research seeks to replicate the findings reported here in a setting in which we can collect such data. As a result, we soon will be able to test our model using objective performance indicators.

Model Limitations It is equally important to recognize the limits of our test of the causal framework we outline. Correlation does not demonstrate causality, so it is important to remember that the order of causality shown in the data reflects that outlined within our model. The analysis can show that the data is or is not consistent with our model, but consistency does not prove that our model is correct. A failure of the data to be consistent with our predictions would suggest that our model is not correct, but consistency with the model provides only preliminary support for that causal framework. It is possible that the data is equally, or even more, consistent with another framework that we do not test, or is influenced by unmeasured third variables.

The Design of This Study

21

In the long run, we feel that further research is needed to demonstrate causal validity for our model. Such further research should involve both experiments and longitudinal field studies. We view this effort as only a first step in a program of research designed to validate the model outlined in this book. If, on the other hand, the data presented here is not consistent with our model, then further research to demonstrate internal validity seems unnecessary. Support for our model in the type of analysis presented here therefore is necessary, but not sufficient, to demonstrate that our model is valid. With these caveats in mind, let us proceed with our argument regarding the relationship and importance of justice, identity, and engagement in group contexts.

Why Study Cooperative Behavior in Groups?

It would be hard to overemphasize the importance of the level and type of

cooperative behavior engaged in by group members in shaping the extent to which groups can function efficiently, effectively, and, ultimately, successfully. In general, when the people within groups engage in more cooperative behaviors, the groups to which they belong become more efficient, effective, and viable. Hence, it is in the group's best interest to encourage cooperative behavior. Social psychologists have long recognized the value of facilitating the frequency of the cooperative behavior that occurs within groups and have studied its occurrence in a variety of social settings (Derlega & Grzelak, 1982). These range from trying to understand the level of the help that people give to particular group members who are in distress (Latane & Darley, 1970) to exploring what shapes the willingness of people to cooperate with the groups, organizations, and communities to which they belong (see the social dilemma literature, Dawes, 1980; Komorita & Parks, 1994). In both situations, the root issue is the same: Why do people engage in actions that benefit their group, organization, or society? In other words, why do people act prosocially toward the various groups to which they belong? As we have noted, people's behavior can be viewed as a continuum that ranges from doing nothing to acting in many ways that help one's group. We will examine this issue in the context of people's behavior in organized groups; in this case, work organizations. Recent discussions of organized groups have demonstrated that there is a link between cooperative behavior and group effectiveness (Pfeffer, 1994). It has been argued that organized groups benefit when their members are "willing to expend extra effort on behalf of the organization" (Pfeffer, 1994, p. 24). For example, work organizations are more efficient and effective when their employees cooperate by working harder or longer at group tasks. This includes both engaging in superior in-role behavior (the duties, activities, and accomplishments which are considered

23

24

Cooperation in Groups

part of the job) and by participating in extra-role behavior (activities which are beyond the explicit job description). The idea that cooperative behavior on the part of group members is advantageous to the group has an obvious, intuitive appeal to it. Research in various areas of organizational behavior also has empirically supported this intuition. For example, in the context of work organizations, research has shown that the behavior of employees measurably influences organizational productivity and economic performance (Freund & Epstein, 1984; H. C. Katz, Kochan, & Weber, 1985). The literature on organizational citizenship behaviors is predicated on the idea that having people in an organization who act beyond their required roles facilitates the superior performance of the organization. Organizational citizenship behavior has been empirically shown to improve organizational performance. Podsakoff, Ahearne, and MacKenzie (1997) cited four studies that support the relationship between more organizational citizenship behaviors and higher organizational performance. Across these four samples, organizational citizenship behaviors accounted for approximately 19% of the variance in performance quantity, 18% of the variance in performance quality, 25% of the variance in financial measures of performance, and 38% of the variance in customer service measures of performance. Furthermore, when Podsakoff, Ahearne, and MacKenzie (1997) divided organizational citizenship behaviors into various dimensions, their findings showed that the dimension of helping behaviors has a stronger relationship with these four performance criteria than do sportsmanship and civic virtue, a distinction we will return to below. In sum, their work showed that organizational citizenship behaviors do have positive consequences and that this effect is robust across several different criteria of group performance. That cooperative behaviors are desirable from the vantage point of the group is implicitly shown by the value placed on them by group leaders. For instance, research has shown that, in work organizations, involvement in organizational citizenship behaviors has favorable consequences on evaluations of the individual performing them (Borman & Motowidlo, 1997; MacKenzie, Podsakoff, & Fetter, 1991, 1993) and on the value placed on their contributions (Orr, Sackett, & Mercer, 1989). Some research has identified perceived affective commitment and liking as mediators of the relationship between organizational citizenship behavior and group performance (Allen & Rush, 1998; Shore, Barksdale, & Shore, 1995), providing insight into the processes that may underlie how individual evaluations are affected by engagement in cooperative behaviors. This evidence supports the notion that group authorities value such behavior because of the positive influence it has on group functioning. Interestingly, this result leads to a confound with the original meaning of extrarole behavior, since employees were originally considered to have shown extrarole behavior only if they engaged in behavior that was not being rewarded. However, as managers have come to see the value of "voluntary" behavior, they have increasingly attempted to reward those who engage in it. Consequently, employees may be engaging in such behavior because they know it will be

Why Study Cooperative Behavior in Groups?

25

rewarded. In this study, we seek to avoid this problem by asking respondents to report on the extent to which they engage in behaviors that are not required by their work organization. In other words, the discretionary nature of these behaviors lies in their not being required, although they may be rewarded in some sense. In our analyses, we will be able to see what the effects of rewards may be on eliciting these behaviors. It is more likely that directly rewarded behaviors will be perceived as required. We attempt to circumvent the potentially confounded nature of extra-role behavior by focusing respondents' attention on behavior that is not required. Additional research is required to directly test the extent to which the rewarding of behaviors leads to the perception that they are "required." Through what mechanisms does cooperation benefit groups? The importance of cooperation is linked to the central functions performed by groups, organizations, and societies. Those functions involve coordinating the actions of the individuals within the group. For instance, the benefits of organizational citizenship behaviors have been attributed to the ability of such cooperative behaviors to "lubricate" the social machinery of the organization, reducing friction, and thereby increasing efficiency (Podsakoff et al., 1997). In other words, cooperation may improve group performance via direct methods such as the attainment of group goals and the freeing up of group resources, as well as improving the social coordination of individuals within the group. Understanding how coordination exists in groups is therefore important to grasping how groups achieve their goals and the role of cooperative behavior in this process. There are (at least) two ways in which groups may work toward achieving their goals by coordinating the behaviors of group members. One way is through group members engaging in behaviors that promote the group's wellbeing. These behaviors have a direct impact on the group's success, since they are directly tied to achieving the group's objectives and goals. Groups may coordinate these behaviors formally through the specification of appropriate actions and tasks for people who occupy particular positions or roles within the group. This specification allows the behavior of individuals to be directed toward the achievement of common objectives, since roles are assigned on a basis that ensures crucial group tasks are accomplished. In the case of work organizations, job descriptions often stipulate particular forms of cooperative behavior that are defined as part of the job. In fact, it frequently is the work of industrial psychologists to create these job descriptions by analyzing the needs of the group and identifying the qualifications necessary to fulfill those roles (a process called "job analysis"; see Harvey, 1991). The roots of this work lie in the ground-breaking studies of F. W. Taylor (1911). By properly defining job roles, the work organization ensures that actions and behaviors needed for it to be successful are performed. For example, someone's job description indicates that they should pick up and empty the trash cans in the office. This job description facilitates the occurrence of a behavior that is necessary and therefore benefits the group. Because of the job description, someone has a role in the group that involves engaging in this behavior. For the gains associated with the creation and maintenance of groups to be realized, people must engage

26

Cooperation in Groups

in the behaviors mandated by the group to be part of their job. Proper definition of job roles also is important because it ensures that group members' efforts are utilized efficiently (i.e., that tasks are not repeated unnecessarily and that work flows through an appropriate process). The prescription of duties associated with certain roles not only occurs in work groups. Other types of groups, such as families, social groups, and even political systems, all prescribe behaviors that are associated with different roles in the group. In work groups, prescribed actions are referred to as in-role behaviors. In legal and political systems, they are defined by laws and group policies. In social groups, which tend to be less formal in nature and structure, these duties may be more implicit than explicit. Nevertheless, expectations regarding each individual's role and contribution develop even in social groups. In a family, for example, there are clear prescriptions for the appropriate behaviors associated with the roles of father, mother, and child. Usually, groups also benefit when people within the group engage in behaviors that promote the group's goals, but that are not necessarily prescribed. Indeed, if people restricted their behavior to only those actions that are formally stipulated by the group's norms and rules, then the group's functioning would be considerably stifled. There are several reasons why it is important for group members to perform nonprescribed behaviors. First of all, it would be difficult to stipulate or establish roles and norms for every behavior that contributes to a group's well-being, given the complexity that exists in carrying out group functions. Second, groups need to be flexible since they are entities that exist in dynamic environments. This flexibility requires that members adapt to new situations and often may require them to take initiative and act out of their own internal motivations. These types of cooperative behavior are referred to in work settings as extra-role behaviors (O'Reilly & Chatman, 1986). Consider the example of in-role behavior we have outlined above-picking up the trash. Clearly, it is someone's job to pick up the trash in the trash cans, but many situations are more ambiguous. For example, Xerox rooms are a notorious mess because people leave loose papers, paper clips, staples, and other assorted junk around in them, although often not in the trash cans. There is wide variation in the degree to which either the cleaning people or others in the group move beyond the clearly specified rules and expand their conception of their group roles to include cleaning around the room (but outside the trash can) as well as on the machines. Yet, the group obviously would benefit from people's willingness to be flexible enough in their conception of their roles to expand them to deal with the problems posed by issues such as trash that is not in the trash can. Of course, more important examples of this same idea can be imagined, such as flexibility with changing client demands, or, in family situations, everyone chipping in when finances are tight. In sum, groups benefit both when people do things that are beyond their required roles and when they put their efforts toward performing their required duties well. Importantly, research confirms the independence of the function of inrole and extra-role behaviors in shaping overall performance (e.g., Motowidlo &

Why Study Cooperative Behavior in Groups?

27

Van Scatter, 1994). This supports our conceptual distinction between mandatory and discretionary behaviors in group settings. While both types of behavior help the group, they have distinct conceptual underpinnings that distinguish them from each other. Together, we will refer to both in-role and extra-role behaviors as promoting behaviors since, in both cases, group members' conduct is in the interest of promoting the group's goals. Both are active ways by which group members promote the achievement of the group's objectives. For instance, in order for the gains associated with organized work groups to be realized, people must both do their jobs and do things that help the work group but are not required. It is the challenge of groups to encourage and coordinate the performance of both of these types of behavior. Other than by encouraging the promoting behaviors described above, the second way that coordination of actions among group members may facilitate group success is through limiting behavior. Limiting behavior involves decreasing the frequency of people's undesirable behaviors within groups. To increase the effectiveness of behavior-limiting approaches, people are directed by group rules and group authorities away from engaging in actions that harm the group. The way in which in-role behaviors are to be carried out is also specified (and therefore limited) by group rules and authorities. Much of our understanding about strategies for limiting undesirable behavior comes from the study of social regulation. Although issues of social regulation are important in all groups, the study of social regulation has tended to focus on legal and governmental authorities (Braithwaite, 1985; Bryner, 1987; Hawkins, 1984; Kagan, 1978; S. Kelman, 1981; Melnick, 1983; Mitnick, 1980; Noll, 1985; Reichman, 1998; Scholz, 1984; Stone, 1975; Vaughan, 1983; Wilson, 1980; Yeager, 1991). These studies have focused primarily on government regulation of work groups, rather than on the internal efforts of work groups to regulate themselves. As we might expect, the efforts of the group to place limits on the behavior of its members often develop in situations in which people's self-interest leads them to want to engage in behaviors that do not benefit the group. It might, for example, seem advantageous to a particular group member to take the computers from the psychology department's student laboratory and sell them on the street. That individual certainly would benefit from the income. However, his or her behavior would hurt the group, which would have to use group resources to replace those computers. Within social psychology, the social dilemma literature details the many situations in which the short-term interests of individuals and groups diverge, with individuals feeling self-interested motivations to engage in actions that are not in the interests of the groups to which they belong. Groups respond to such conflicts in a variety of ways, each of which is designed to lower the frequency of individual self-interested behaviors. Group authorities may limit behavior for several reasons. First is the motivation of preventing self-interested actions that may harm the group. Authorities also are concerned with increasing the coordination of individual group members' efforts to complete tasks. This requires the specification of rules regarding how those actions are to be carried out and how group members are to coordinate their

28

Cooperation in Groups

behavior to achieve particular ends. For example, every employee in a company would prefer to set his or her own work hours. However, to make it possible for people to work together, the group must set common hours, for example, 9 a.m. to 5 p.m. This type of social coordination involves some restrictions on selfinterest, since people with children might prefer to work at one time and young unmarried people at another. However, self-interest is not as directly opposed to group interest as in the case of stealing. Similarly, social rules coordinate public actions like driving on one side of the street. It does not make any intrinsic difference whether cars drive on the right or the left side of the street, but it is very important that everyone within a single group or society drive on the same side of the street. To assure this desirable group goal, rules limit people's ability to drive wherever they want. These types of efforts at social coordination are less likely to be difficult to implement, since people can successfully pursue their personal self-interest within a variety of frameworks of social coordination. Here, there is little loss of the ability to pursue self-interest, since people can get to where they want to go equally quickly by driving on either side of the street. In this analysis, we consider all of these forms of effort to restrict behaviors that hurt the group (and have to be regulated) and hurt the coordination of group members' behavior (and have to be coordinated) to be equally "undesirable" from the vantage point of the group. Hence, groups benefit when either of these types of behavior is less frequent. We will collectively refer to this goal as that of limiting the frequency of undesirable behavior within the group. Whether the goal is preventing harmful acts against the group or constraining individual behavior to facilitate the social coordination of group members, regulatory systems are needed to limit individual behavior. For the gains of a regulatory system to be realized, people must obey the limits specified by the rules (Tyler, 1990). This coordination function involves the limiting of behavior, since it is concerned with encouraging people to bring their behavior into line with group rules. Here, rules are designed to discourage people from engaging in behavior that hurts the group and to encourage behavior that facilitates individuals working together for the group, again limiting their freedom to act as they wish. An additional word may be needed to more explicitly differentiate this latter idea from the earlier notion of prescribed promoting behaviors. Rules that limit behaviors in the interest of promoting social coordination specify how tasks are to be carried out. They may be rules specifying what time everyone is to show up for work, the sequence with which tasks are to be carried out, that students are to sit quietly through class, or that committee members are to meet deadlines for decisions. Rules also may limit behaviors with the hope of preventing group members from harming the group by antisocial acts such as stealing group resources. In either case, the rules are designed with the goal of avoiding problems with the group's functioning. We distinguish this from promoting behaviors, which tend to be focused on an affirmative statement about what needs to be done rather than a set of rules about how things are done. By focusing on what needs to be done to meet

Why Study Cooperative Behavior in Groups?

29

the group's goals and needs, the emphasis in these types of behaviors is less on social coordination and the avoidance of problems (which is the focus of limiting behaviors). Instead, promoting behaviors emphasize the achievement of the group's success and increased prosperity. Earlier, we used the example of job descriptions to describe what these prescribed promoting behaviors might be. Such behaviors are referred to as "roles" and they identify the tasks to be performed in a job, the rules that direct an individual's activities (D. Katz & Kahn, 1966). For this reason, these behaviors often are referred to as in-role behaviors (i.e., the things involved in "doing one's job"). Job descriptions usually indicate what an individual is to do at work, but not how those duties are to be performed or the parameters surrounding their performance. Of course, we want to emphasize that these different elements of cooperative behavior-promoting and limiting-can never be seen as wholly independent. To some extent, it is impossible, and certainly unimportant, to make a complete distinction between how and what people do when fulfilling group norms or following group rules. Further, situations may arise where promoting and limiting behaviors conflict, such as when novel circumstances require one to ignore group rules and act beyond his or her role in the interest of promoting the group's success. Our point is that both limiting and promoting behaviors are necessary to achieve the overall success of the group. Nonetheless, there exists a conceptual difference that we believe makes distinguishing between them, even if incompletely, worthwhile. Cooperation in both of these domains-promoting behaviors and limiting behaviors-on the part of group members is important because it helps the group to more effectively manage its problems, maintain its effectiveness, and sustain its viability. From the group's perspective, all such cooperative behavior by members might be collectively labeled "value-added" behavior, since the occurrence of such behavior increases the viability and effectiveness of the group. A group whose members work together on promoting behaviors and who are willing to limit their behaviors is likely to more successfully coordinate the actions of group members and to be more efficient in dealing with and completing group tasks. When we observe the people within groups, we typically see many examples of each of the types of cooperative behavior we have outlined above. People generally engage in job-related promoting behaviors that facilitate achieving group goals and they also usually follow group rules that limit their behavior. However, no matter what type of group we examine, the occurrence of these behaviors cannot be taken for granted. Consider again the case of work organizations. Employees cannot always be counted on to do their jobs well. They may slacken in their efforts, cut corners in their work, or fail to put their best effort into work tasks. Additionally, they may refuse to perform extra-role behaviors. The same may be said of social and political groups. Even in families (i.e., groups with long histories and close emotional ties), parents may neglect to perform their parental duties and children may shirk their family roles and responsibilities. In all of these cases, individuals are not engaging in behaviors that promote their respective group's goals.

30

Cooperation in Groups

Likewise, with limiting behavior, whether people in work groups are using drugs on the job, stealing office supplies, ignoring procedures, or engaging in industrial sabotage, they are hindering the success of their group. Similarly, if children are staying out too late or stealing money from their homes, they may be harming the functioning of their family. In sum, people often act in ways that are contrary to group rules. That is, they are defying the limitations on their behaviors. Since group rules often limit people's ability to engage in personally rewarding behaviors, people may simply ignore them. For example, people may drive faster than the speed limit or steal items from stores. More extremely, people may be antisocial in their behavior and may engage in actions that are motivated not by pure self-interest, but by the desire to hurt the group (Giacalone & Greenberg, 1997). Sabotage is an example of such a behavior, since people are intentionally damaging the resources that allow the group to function, without any apparent material benefit to themselves. An example of this type of behavior is when employees attempt to harm a work organization's computer networks or delete important files from computers. In society, vandalism represents this type of behavior.

D

Four Types of Cooperative Behavior

We have noted, in this chapter, two basic types of cooperative behavior: promoting behaviors and limiting behaviors. Promoting behavior is concerned with facilitating those behaviors that help the group to achieve its goals. This can occur when group members are encouraged to perform behaviors that have been defined by the group as being important to its success, or when group members step beyond their traditional roles and act on behalf of the group. The distinction between prescribed and nonprescribed promoting behaviors maps onto the mandatory-discretionary distinction discussed in the introductory chapter. The performance of stipulated tasks can be construed as mandatory behavior, since those tasks have been identified by the group and labeled as necessary for its success. Group members likely perceive these tasks as being required. Discretionary behavior, on the other hand, is more consistent with the idea that underlies the performance of extra-role promoting behavior, the nature of which is not clearly prescribed. Limiting behavior is concerned both with discouraging the occurrence of behaviors that hurt the group and with setting rules to ensure that the different components of the group work in coordination. The issue of limiting rules leads the issue of social regulation to be more process-oriented than promoting behaviors. People acting in accord with these limits on their behavior are engaging in cooperative behavior insofar as they are acting on behalf of the group's overall good, instead of their own, and they are facilitating the coordination of individuals within the group. It is for these reasons that we identify the following of group rules as a cooperative group behavior. Groups benefit when people coop-

Why Study Cooperative Behavior in Groups?

31

erate with the rules of all the groups to which they belong (assuming, of course, that those rules are not misguided). Limiting behaviors also can be mapped onto the mandatory-discretionary distinction we drew above. People may follow group rules because they believe that such rule following is mandatory. In such cases, they will follow those rules only when they believe their rule breaking may be detected (i.e., when there is some risk that sanctions might be applied). On the other hand, they may follow rules (i.e., engage in limiting behaviors) out of their own volition. In this case, their rule following occurs irrespective of the possible detection of rule breaking behavior; they follow group rules because they believe that it is the right thing to do, regardless of whether anyone will know whether they did or not. It thus is discretionary, as opposed to mandatory, in nature. Throughout the rest of this book, we will refer to mandatory limiting behavior as compliance, while discretionary limiting behavior will be described as deference. When we combine the distinction between promoting behaviors and limiting behaviors with the distinction between mandated and discretionary behavior that we outlined in Chapter 1, we end up with the four categories of cooperative behavior shown in Table 1-1. This table defines four types of cooperative behavior: compliance (mandated-limiting), deference (discretionary-limiting), in-role behavior (mandated-promoting), and extra-role behavior (discretionary-promoting). Our goal in this analysis is to examine motivations that shape the occurrence of each of these four types of cooperative behavior. In other words, these are the cooperative behaviors we will examine in our study. Although we have been considering employee behavior in work organizations and using examples from typical private work organizations, the ideas being outlined can be applied more broadly. For example, they can be applied to public management. Consider the case of police officers. One thing that we want police officers to do is to refrain from engaging in illegal activities such as taking bribes. To some extent, this activity can be limited by sanctions for rule breaking. However, we give the police a great degree of discretion to deal with citizens, and we have to rely heavily on their efforts to self-limit their behavior. For example, the police have discretion over their use of deadly force. Although they can be brought to account for wrongful shootings, the threshold of punishment is very high, and citizens usually trust the motives of the police in ambiguous situations. The major controls on police wrongdoing are the internal moral values held by police officers. We also want the police to do their jobs. When they see a problem, we want them to step in and work to solve it. To some extent, such behavior is promoted by the specification of the job assigned to police officers. However, again in many situations, police officers have much discretion. If they see a crime in progress, they can stop and intervene or they can drive by and ignore the problem. To have an effective police force, we need officers who want to handle problems and resolve issues that they encounter on their job. An important motivation for doing so is commitment and loyalty to the job. Beyond public management, the concerns we have outlined above are important in all groups. In a family, for example, parents want their children to follow

32

Cooperation in Groups

family rules by coming home on time. They also want them to help around the house by cleaning, carrying groceries, and so forth. As in all groups, some of these actions can be specified by rules. But many are discretionary and parents also rely heavily on the goodwill and voluntary cooperation of their children.

0 This Study The four types of cooperative behavior outlined above were assessed among the 404 employees interviewed for this study. Table 3-1 indicates people's average responses to each of the items reflecting compliance, in-role behavior, deference, and extra-role behavior. An examination of the mean levels of people's self-reported limiting behavior suggests that people generally indicated that they follow group rules and that they avoid actions that are against those rules. However, there is considerable variance in the extent of this self-reported behavior, and it is certainly true that most employees acknowledged engaging in at least some rule-breaking behavior. Similarly, most employees indicated that they generally engaged in promoting behaviors, although again there was variation among respondents in the degree to which they engaged in these behaviors. Interestingly, people reported breaking rules more often than they acknowledged not doing their jobs. The average frequency found for reported discretionary behaviors (e.g., willingness to defer to authority and to engage in extra-role behavior) was lower than that of mandated behaviors, suggesting that people were less likely to voluntarily defer to group rules than they were to obey them and that they felt less willing to do extra-role behaviors than to do their jobs. This seems reasonable, since discretionary behaviors are voluntary and involve a decision among those interviewed about their willingness to consent to cooperating with the group. Engaging in mandatory behaviors is more likely to be necessary for ensuring continuance as a group member, while engaging in discretionary behaviors is not. Further, these empirical differences in the reported frequency of discretionary and mandatory behaviors are logical if one considers that those engaging in discretionary behaviors also are likely to meet the criteria for engaging in mandatory behaviors, while the opposite is not as likely to be the case. Table 3-2 shows the average level of the scales created to reflect each of the four types of cooperative behavior outlined above, as well as showing the correlation among the four scales that reflect each of the four types of cooperative behavior. The means indicate that people are most likely to indicate that they engage in mandated in-role behaviors, and least likely to indicate that they engage in discretionary rule following (deference) and discretionary helping (extra-role) behaviors. The results show that the four types of cooperative behavior are somewhat related (mean correlation= 0.39). Interestingly, the mandatory-discretionary distinction shows a smaller association (mean correlation = 0.35) than the promoting-limiting distinction (mean correlation= 0.45), suggesting that people tend to engage in behaviors that serve similar functions more than they tend to engage in behaviors that they view as similarly required or at their discretion.

Why Study Cooperative Behavior in Groups?

33

TABLE 3-1. Cooperative behavior measures

M (SD) Compliance (6 = very often, 1 = never) How often do you: Comply with work-related rules and regulations? Follow the policies established by your supervisor? Carefully try to carry out the instructions of your supervisor? Take inexpensive items such as office supplies from work for personal use?a Use sick leave when not sick?a Receive pay for hours or overtime not worked?a Use the organization's resources for personal projects?a Spend time at work doing personal things?a In-role behavior (6 = very often, 1 =never) How often do you: Fulfill the responsibilities specified in your job description? Perform the tasks that are expected as part of your job? Meet the performance expectations of your supervisor? Adequately complete your required work projects? Exert your full effort when getting your job done? Deference (6 = very often, 1 =never) How often do you: Willingly follow your organization's policies? Follow organizational policies even when you do not need to do so because no one will know whether you did or not? Do what your supervisor expects of you, even when you do not really think it is important? Willingly accept the decisions made by your supervisor? Defer to your supervisor's decisions even when you could go to others to try to change them? Implement your supervisor's decisions even when he/she will not know whether you did? Accept decisions even when you could complain to higher authorities about them? Extra-role (6 = very often, 1 = never) How often do you: Volunteer to do things that are not required in order to help your organization? Volunteer to help to orient new employees? Make innovative suggestions to help improve your work setting? Volunteer to help others when they have heavy workloads? Help your supervisor, even when not asked to do so? Put an extra effort into doing your job well, beyond what is normally expected of you? Lend a helping hand to others at work? Do work that is not the best you could do because you are angry at your employer?a Consider bringing issues at work to the attention of outside agencies like the courts?a Try to find ways to hinder or undermine your work supervisor?a aReversed to create a scale.

5.23 5.21 5.25 2.74

(0.63) (0.67) (0.67) (1.38)

2.34 1.49 2.67 3.25

(1.37) (1.02) (1.43) (1.32)

5.64 5.70 5.49 5.58 5.18

(0.55) (0.52) (0.65) (0.59) (0.94)

5.02 (0.71) 4.56 (1.03) 4.89 (0.79) 4.87 (0.70) 4.37 (1.14) 4.72 (0.98) 4.47 (1.18)

4.33 (1.14) 4.22 (1.36) 4.40 (1.16) 4.42 (1.05) 4.42(1.11) 4.75 (0.99) 4.79 (0.90) 2.12 (1.13) 1.60 (1.12) 1.38 (0.81)

34

Cooperation in Groups

TABLE 3-2. Relationship among types of behaviors Cooperative behaviors M (SD)

Compliance In-role Deference Extra-role

4.78 5.52 4.70 4.72

(0.68) (0.49) (0.62) (0.65)

Compliance

In-role

Deference

Extra-role

.76 .32 .41 .43

.79 .30 .48

.77 .37

.80

Note. Entries are Pearson correlations (n = 404). Diagonal entries are the alphas for each scale. Scales range from 1 to 6, with high numbers indicating more cooperative behavior.

D Summary This chapter outlines the cooperative behaviors that we will try to understand in this book. It also establishes the theoretical distinctions between these different types of cooperative behavior that will be relevant to testing the hypotheses of our model. It is clear that there is wide variation in the degree to which the respondents in the current study indicated that they engaged in various cooperative behaviors within their work groups. Further, while the various forms of cooperation were found to be interrelated, they also were found to be distinct, suggesting that it is worthwhile to examine each of these types of cooperative behavior separately. As we have noted, our goal is not to describe behavior within any particular type of group. Instead, we want to present a general theory of cooperative behavior. In the case of promoting behavior, for example, work groups as well as community organizations, families, and society itself all seek to encourage people to act in ways that facilitate the completion of group tasks. Irrespective of whether people are being encouraged to join a neighborhood watch, to vote, or to do their jobs, the group seeks to encourage actions. Similarly, all of these types of groups structure and regulate the behavior of group members. When people are being encouraged not to speed or steal, not to take office supplies or 3-hour lunches, and not to take the family car out for a joyride, the group is seeking to regulate the behavior of the individual.

Instrumental Motivations for Engaging in Cooperative Behavior

As we explore the underpinnings of cooperative group behavior, we will draw a distinction between two types of motivation, each of which might potentially lead people to engage in cooperative behavior. Those two types of motivation are instrumental, or environmentally driven, motivation and internally driven motivation. We will first focus on instrumental, environmentally driven, motivations, the topic of this chapter; internally driven motivation will be discussed in chapter 5. Instrumental motivation is concerned with cases in which people cooperate with others because they expect rewards for cooperation or fear punishments if they fail to cooperate. These instrumental reasons are central to both incentive approaches to encouraging desired behavior and to deterrence or threat-based approaches to discouraging undesired behavior. Every era in history has it own unique perspective on how to maintain social order and effectively manage the people within groups. Over the past half century, the social sciences have been dominated by an instrumental view of people's motivations for acting on behalf of groups, whether informal groups, formal organizations, or societies. This view of motivation suggests that people engage in desirable cooperative behavior because they think such behavior is likely to be rewarded by the group. Conversely, they refrain from rule breaking when they think they will be caught and punished for such undesirable behavior. The underlying assumptions of this instrumental or resource-based model are described by social exchange theory. Social exchange theory links people's reactions to organizations to the resources that they receive, or expect to receive, from them (Homans, 1961; Kelley & Thibaut, 1978; Thibaut & Kelley, 1959). There are two key assumptions of the model. First, that group membership and participation is rooted in the exchange of resources. This approach emphasizes material resources, although some social exchange theorists have recognized a broader group of potentially important resources (e.g., Foa & Foa, 1974; Kelley &

35

36

Cooperation in Groups

Thibaut, 1978). Second, the model assumes that people are motivated to maximize their personal gain. Within economics, this model is referred to as the theory of rational choice. While this theory was originally developed to explain economic choices in market settings, it has been applied widely to the study of law, political science, and management. Within these various contexts, the fundamental underlying assumption of rational choice theory, as with social exchange theory, is that people seek to maximize personal gain. Further, rational choice models are typically oriented toward material gains and losses. Although a review of the wide ranging literature on rational choice is beyond the scope of this book, it has been highly influential in a wide variety of areas of the social sciences. Gain-loss arguments also have been used by a variety of social psychologists as possible explanations for the motivations underlying people's willingness to help others in groups. The willingness to help others has been linked to the perceived benefits and costs of helping (Latane & Darley, 1970; Piliavin, Piliavin, & Rodin, 1975), while cooperation within groups has been linked to estimates of the likelihood that others will reciprocate such cooperative behavior (Komorita & Parks, 1994; Rousseau, 1995; Tyler & Kramer, 1996; 0. J. Williamson, 1993). Expectancy theory similarly links work motivation to expected payoffs (Vroom, 1964), as does goal setting theory (Locke & Latham, 1990). An example of the application of these assumptions to cooperation in groups also is found in the psychological literature on leadership. Cooperation with leaders reflects cooperation in groups, since leaders typically represent groups and articulate group rules. The literature on motivations for following leaders has argued that leader-follower relations depend on the exchange of rewards. According to this perspective, if leaders make good decisions that lead to success and to the gain of resources for group members, followers respond by obeying the directives of their leaders (Levine & Moreland, 1995). For example, some studies of leaders have emphasized the importance of their task competence (Hollander, 1980; Hollander & Julian, 1978; Ridgeway, 1981), suggesting that people will follow those leaders who they feel can solve group problems in a way that will lead to personal gain for group members. Similarly, transactional theories of leadership suggest that leader-follower relations depend on resources received from leaders in the past or expected in the future (Bardach & Eccles, 1989; Dasgupta, 1988; Komorita, Chan, & Parks, 1993; Komorita, Parks, & Hulbert, 1992; Wayne & Ferris, 1990; 0. J. Williamson, 1993). One example of such a theory of leadership is vertical dyad linkage theory (Dansereau, Graen, & Haga, 1975) which explores the nature of the exchange relationships between organization members and their leaders (Chemers, 1983, 1987; Duchon, Green, & Taber, 1986; Vecchio & Gobdel, 1984). Such exchanges vary in the nature of the resources exchanged, although theories typically focus on material rewards and costs (Dansereau et al., 1975; Dienesch & Liden, 1986; Graen & Cashman, 1975; Graen, Wakabayashi, Graen, & Graen, 1990; Liden & Graen, 1980). Of course, expected gain-and-loss judgments in group settings not only are made about the immediate situation. People have long-term relationships with

Instrumental Motivations for Engaging in Cooperative Behavior

37

groups and they make long-term judgments about the expected costs and benefits of group membership. In the context of ongoing groups, these more long-term judgments of expected rewards and costs guide people's behavior within their group. In making such long-term judgments about what types of behavior will be rewarding, people evaluate the overall quality of the outcomes they are receiving from the group, across situations, relative to their available alternatives, as well as judging the degree to which they already have invested resources in the group. An example of the application of long-term resource-based approaches to the study of behavior in groups is provided by the investment model (Rusbult & Van Lange, 1996), which studies the factors shaping people's decisions to leave or remain within groups (their "loyalty" to the group). The investment model predicts that the key factor shaping personal decisions about whether to exit a group is the degree of dependence an individual feels on the group for obtaining personally valued resources. Dependence judgments involve considerations of one's immediate and expected long-term reward level, the quality of one's alternatives, and the amount that one has invested in a group. Studies based on the investment model have suggested that greater dependence on a group or relationship leads to heightened loyalty, with people being less willing to leave groups that provide them with high levels of desired resources or in which they have already invested resources (Farrell & Rusbult, 1981; Rusbult, 1980, 1983; Rusbult & Farrell, 1993; Rusbult, Farrell, Rogers, & Mainous, 1988; Rusbult, Johnson, & Marrow, 1986a, 1986b; Rusbult & Lowery, 1985; Rusbult, Zembrodt, & Gunn, 1982). These studies have supported the argument that one way to understand people's behavior in groups is through an instrumental perspective. They have emphasized the value of such an instrumental approach being linked to overall and long-term assessments of resources obtained from the group, as well as to the immediate gains or risks found within any particular situation.

0

Rule Functions

In this book, we make a basic distinction between two functions of authority systems within groups: encouraging desired behaviors (promoting behaviors) and discouraging undesired behavior (i.e., encouraging limiting behaviors). Each of these two functions is associated with a particular type of instrumental strategy for facilitating cooperative behavior. Strategies for encouraging desired behavior typically are linked to the use of reward-driven incentive systems. If we wish group members to engage in particular types of cooperative behavior, we can encourage that goal by providing pay and promotions to those who engage in the prescribed behaviors, as well as by linking engagement in such behaviors to a wide variety of group "perks," such as more office space, greater access to the use of group resources, and other types of resource-based or financial incentives. Providing all of these rewards to group members based on their willingness to engage in role-based cooperative behavior has the goal of motivating desired cooperative behavior within the group.

38

Cooperation in Groups

Strategies to discourage the occurrence of undesired behavior usually are associated with threatening or punishing people who engage in such behaviors. These are punishment-driven sanctioning systems. Groups typically have rules that define those undesirable behaviors and indicate that they should not be engaged in. Groups then discourage people from engaging in such undesirable behaviors by attempting to catch and punish those who do not follow the rules. Such deterrence or social control models typically are found in the context of social regulation, such as the operations of police and the courts (Tyler, 1990; Tyler, Boeckmann, Smith, & Huo, 1997; Tyler, Huo, & Lind, 1999). Although usually associated with encouraging different types of behavior, it is important to see rewards and punishments as two sides of the same instrumental coin. They both are means of operationalizing an incentive-driven instrumental perspective. As such, irrespective of whether we are concerned with the use of incentives or punishments, there are several key issues that must be addressed when evaluating the use of instrumental systems by groups. Those issues include: (a) whether it is possible to effectively evaluate if desired cooperative behavior is occurring, (b) how the use of the instrumental system influences the behavior of the person, and (c) what effect the use of an incentive system has on the group.

D

Incentive Systems

The first question to ask about incentive systems is whether it is possible to evaluate if desired cooperative behavior is, in fact, occurring under incentive systems. One of the positive features of incentive systems is that they encourage people to bring their behavior to the attention of group authorities. Since people seek the incentives offered by the group, they want authorities to know about the cooperative behavior they are engaging in, and there are few problems associated with observing cooperative behavior. On the contrary, people are likely to want to bring their cooperative behavior to the attention of those group authorities who are responsible for dispensing group rewards. On the other hand, while they will seek to bring their behavior to the attention of group authorities, people also are motivated to distort their behavior in ways that emphasize the positive or desirable aspects of their actions. In other words, people have a tendency to try to corrupt or undermine systems of incentive-based control over their behavior (Darley, 1999). They do so in a wide variety of ways, the common purpose of which is to allow them to receive incentives without actually engaging in the cooperative behavior desired by the group. People are motivated to distort their cooperative behavior in ways that exaggerate or even fabricate the connection between their behavior and the behavior desired by the group. Darley provides a number of examples of the corruption of "objective" indexes of performance and cooperation within real-world groups. Why might group members focus more on making it appear that they performed the rewarded behavior rather than on actually performing it? Group members might do this because incentive systems that encourage particular types of cooperative group behavior focus their attention on obtaining the reward for

Instrumental Motivations for Engaging in Cooperative Behavior

39

the behavior itself, as opposed to the potential intrinsic value of the desired behavior and the reasons why the desired behavior might be important to the group. Rather than promoting an understanding among group members about why they should engage in a particular behavior, incentive systems manifest a concern with how to best achieve the reward. The behavior that the reward is intended to encourage consequently becomes peripheral in the group member's mind, compared to the reward that is associated with it. This leads us to addressing our second issue of concern to the group, which is how the use of incentive systems influences the behavior of the person. One problem with encouraging cooperation by rewarding it is that incentive approaches do not encourage people to exercise initiative or to become committed to the spirit underlying the desired behavior. Since incentives often are tied to specified outcomes, they lead to a focus on "doing one's job" (i.e., doing what it takes to achieve the outcome specified by the incentive, and nothing more). In other words, even when incentive systems do not become corrupted by group members' efforts to make it seem like they have engaged in the desired behavior when they actually have not, incentive systems still encourage only a narrow range of cooperative behavior. People become focused on the concrete behaviors specified by the incentive plan and lose sight of other behaviors that might aid the group. Similarly, although groups can make efforts to reward positive out-of-role behaviors, incentives generally have the effect of reducing risk taking by group members. Incentive based strategies encourage group members to focus narrowly on the specified task (Kohn, 1993). Since the general effect of incentives is to focus people's attention on doing what is indicated by group rules, their ability to encourage discretionary cooperative behaviors thus is severely limited. People are less likely to take on added challenges and to be creative when solving group tasks that have a incentive attached to them, since they instead will be focused on how to acquire the incentive as quickly as possible. Incentive systems also are not designed for situations in which the contribution that the group desires from the person is difficult to specify in advance, as in cases where the group may try to reward positive out-of-role behaviors. Of course, groups can make efforts to cast a broad net over desired behavior-providing incentives for "showing initiative" or "going beyond what your job requires." That is, the group can seek to provide incentives for acting in ways that are not clearly mandated by one's group role. However, these types of incentive systems leave themselves open to criterionrelated problems and biased evaluations, which can have severe repercussions both on individual and group morale. Implementing such broad incentive systems in a way that will lead all group members to feel that decisions are made in a fair, unbiased manner is a difficult, if not impossible, task. Managers are caught in a dilemma. They can create vague and flexible incentive procedures and risk being seen as dispensing incentives in arbitrary and unfair ways, or they can make the rules very clear and limit employee willingness to be flexible, creative, and innovative. As a consequence, it is easy for incentive-based systems to breed resentment and hurt group functioning.

40

Cooperation in Groups

Incentive systems also are inadequate for influencing the behavior of the group member over time. Since the specified behavior becomes associated with a reward, the behavior will desist once the reward is removed. If people come to believe that they are performing a behavior for a reward, they will require that reward in order to perform the behavior. This means that groups must continually maintain reward systems to motivate employee behavior, which may be costly for many groups. Furthermore, rewards often are subject to habituation, meaning that the motivating force of the same reward becomes lost over time. Research in other areas has confirmed the idea that people often habituate to monetary incentives and conditions (Brickman, Coates, & Janoff-Bulman, 1978; Csikszentmihalyi, 1999), so ever greater levels of rewards are needed to maintain the same level of instrumental motivation. In other words, not only do groups have to maintain incentive systems once they have set them in place, but they actually may have to continually increase the reward that is offered in order to sustain its power to motivate the desired behavior. It also is clear that greater levels of resources or rewards are not as strongly related to subjective well being as some might expect (Diener, Horwitz, & Emmons, 1985). As a result, the magnitude of the motivating power of incentives may not be as great as is often believed (Jenkins, Mitra, Gupta, & Shaw, 1998). Another negative effect of using incentives to encourage cooperative behavior has been described by the large literature in social psychology suggesting that shaping behavior through the use of incentives diminishes people's intrinsic motivations for cooperating. This leads people to be less willing to act because of their own internal enjoyment of cooperation or due to internal values about what is appropriate or ethical to do (Deci, 1975, 1981). For example, in work organizations, seeking to shape behavior through rewards communicates a negative message to employees by suggesting that work is not enjoyable or interesting. Rewards can undermine any internal drive an individual may have to perform his or her job. This leads people to spend less time performing tasks, to report less enjoyment of their work, and to be less creative and original in doing their work (Bordin, 1979). As Deci, Koestner, and Ryan (1999) pointed out in their meta-analysis on the relationship between extrinsic rewards and internal motivation, although rewards can control behavior, their primary negative effect is that they undermine people's self-regulation. The usefulness of rewards on motivating desired work behaviors has been examined empirically. A recent meta-analysis by Jenkins et al. (1998) investigated the effectiveness of financial incentives for improving work performance, and found mixed results for this relationship. These researchers examined 47 studies that tested the relationship between incentives and performance and found that, while incentives had some relationship with performance quantity (corrected correlation = 0.34), there was no significant relationship with performance quality. These findings support our view that incentive systems can be problematic for motivating desirable work behavior, and that the domain in which financial incentives may be useful is limited (Pfeffer, 1997, pp. 111-112). Further, this domain may be shrinking as work roles become less clearly defined and jobs become more complex. Such changes in the nature of work require

Instrumental Motivations for Engaging in Cooperative Behavior

41

individuals not only to follow task descriptions to get the job done, but also to focus on the nature and quality of their work. Further, the observed magnitude of the relationship between financial incentives and performance quantity, while impressive, still leaves open the likelihood that other factors also are at play in explaining what motivates individual performance. On the other hand, the connection of groups and group authorities with the allocation of rewards has the advantage of creating positive affective associations for the group member. Conditioning theories have shown that people associate the feelings they experience in the presence of others with those other people. In the case of rewards, such feelings are positive. Over time, therefore, people will develop positive associations with the group and with group authorities if they receive rewards from the group for cooperating. Lastly, there is the issue of how using incentives changes the group and influences its effectiveness and viability. Kohn (1993, 1999) notes several reasons why the use of incentive systems to motivate behavior may have negative consequences on relationships among group members. Incentives lead to competition among group members, which leads group members to view each other as obstacles to their own success and thus discourages cooperation between individuals. Further, competition leads to there being winners and losers. This feeling of having lost breeds poor morale among employees and obviously is negative for motivating cooperative behavior. Just as rewards congratulate those whoreceive them, they punish those who do not receive them. These problems with incentives are as true of incentive systems that reward individual performance as those that reward team performance since, in both cases, there are those that remain "unrewarded." These morale issues are similar to those we raised earlier regarding the consequences of incentive systems that lack clear, specified criteria. In sum, there are various mechanisms by which the use of rewards can erode social relations. The use of incentives also leads to the inefficient allocation of resources in groups. This misallocation of resources hurts group viability. One effect of using incentive strategies for exercising authority in work organizations, for example, is to increase the percentage of employment costs accounted for by bonuses and incentive payments (Arthur, 1994). At extreme levels, this can lead to the type of unequal pay levels that are the focus of recent concerns about the high level of chief executive officer (CEO) pay in U.S. companies (Kerr & Bettis, 1987). For instance, a recent Business Week study found that the ratio of CEO pay to that of the average blue-collar worker was 419 to 1 in 1998, up from 44 to 1 in 1965 (Reingold, 1999). Such unequal pay levels often are linked to inefficiency in companies (Barkema & Gomez-Mejia, 1998; Bloom, 1999; Huselid, 1995; Jensen & Murphy, 1990; Kerr & Bettis, 1987), as well as to increased turnover (Pfeffer & Davis-Blake, 1992; Pfeffer & Langton, 1988). The problems inherent in highly unequal pay have been recognized and addressed by formal attempts to alleviate those inequalities. For example, proposed legislation in the United States House of Representatives, called the Income Equity Act, would limit the tax deductibility of compensation for CEOs to 25 times that of the organization's lowest paid worker. Further, several large companies

42

Cooperation in Groups

have put resolutions before their shareholders to determine if the CEO's pay should be restricted to a specific multiple of what the lowest paid full time employee earns (D. M. Gold, 1999). In sum, the problem of growing discrepancies between executive compensation and that of the other employees in the organization is one that is becoming widely recognized, as are the negative morale consequences of such systems. The origins of such systems often can be tied back to incentive-based approaches, such as stock options and bonuses. Noncompressed pay structures of the type examined in the discussion of high CEO pay are a problem for the efficiency of work groups because broad differences in pay influence group cohesion and morale, lead to widespread dissatisfaction, and heighten efforts by other members of the organization to seek and gain outside offers. It is those members that an organization least wants to lose to other groups who are most likely to be lost, since the more experienced and qualified employees are those most likely to receive outside offers. Although work organizations often can hold members within them by matching or exceeding any offers they receive from other organizations, this exacerbates compensation inequalities among coworkers, which further diminishes morale and commitment. In other words, reward strategies lead to morale problems within the organization and the loss of valued employees. In contrast, organizations with strong organizational cultures that can maintain employee loyalty in noninstrumental ways will have more efficient compressed pay structures, thereby keeping group morale intact. Incentive systems also lead to unstable groups because groups must have resources available to use in shaping people's behavior. While always costly and inefficient, reward mechanisms are especially problematic during times of transition and crisis. When organizations are in the midst of difficulties or change, they are highly dependent on the cooperation of their members, but least able to provide organizational members with immediate rewards. Hence, dependence on resources for cooperation leads to high levels of organizational vulnerability. In sum, there are a quite a number of arguments against the use of incentive systems. Despite this, they remain at the bedrock of how groups, and work organizations in particular, motivate people. Why might this be so? As Kohn (1999) pointed out, incentives may remain attractive for the simple reason that they are easy to implement and they often result in short-term successes. That is, they often illicit desired behaviors in the immediate context, particularly when those desired behaviors are guided by quantitative performance criterion (such as sales) as opposed to quality-related criterion. It is primarily in the long term that many of the negative side effects of incentives occur, making them less likely to be detected. Additionally, popular notions regarding the motivating power of incentives (and, more generally, of self-interest) lend an intuitive appeal to such approaches and, to some extent, makes their use necessary so that people can justify their own behavior to themselves in terms they will find acceptable (Miller & Ratner, 1996).

Instrumental Motivations for Engaging in Cooperative Behavior

D

43

Sanctioning Systems

Sanctioning systems are used primarily as environmentally driven instrumental means of preventing undesirable behavior by group members through the use of punishments. We pose the same three questions we addressed regarding reward systems to punishment-based systems. The first question is whether it is possible to evaluate if cooperative behavior is occurring under these systems. One of the central problems with sanctioning systems is that they require extensive surveillance of the behavior of group members. Surveillance is needed because people are motivated to hide their behavior from group authorities. They do so because they want to avoid being punished for their actions. Groups therefore must have some system of regulatory authorities through which they can find rule-breaking behavior. The key issue in most regulatory systems is the difficulty of observing behavior when people within the group are trying to hide it. To implement sanctionbased systems, managers need to be able to effectively engage in surveillance of employees' behavior. Sometimes surveillance is easy, because the structure of the situation makes it easy. For example, wage earners' income is easy for the government to monitor, since businesses withdraw money from each paycheck and send it to the government. That makes surveillance of tax payment among this group straightforward. On the other hand, sometimes surveillance is difficult. The police, for example, have great difficulty monitoring public behavior to identify those who are using illegal drugs. Irrespective of its effectiveness, the use of surveillance systems creates difficulties in the social climate of groups. The use of surveillance implies distrust, which lessens people's good feelings about themselves and their groups (see Kramer & Tyler, 1996). Further, people perceive intrusions into their lives as procedurally unfair, leading to negative feelings that will be elaborated later in this book. Whether it works or not, surveillance has social costs. Studies have suggested that the most important issue to the people who are deciding whether or not to break group rules is their estimate of the likelihood of punishment for their actions, not the expected severity of their punishment (see Tyler, 1996, 1997). As a consequence, groups cannot enforce rules simply by developing draconian (i.e., serious) punishments. They must engage in the more costly surveillance strategies involved in increasing the likelihood of detection for rule breaking. For example, they must increase the number of supervisors who are watching employees, or increase the size of the police force that is watching citizens. Leaders, of course, would prefer to increase punishment severity, as in recent efforts to expand the use of the death penalty, since the use of severe sanctions involves less social resources than does the effort to increase the likelihood of punishment. Given sufficient resources, there is nothing inherently untenable about controlling people's behavior via threats of punishment. For example, the U.S. criminal justice system devotes considerable resources to finding those who commit murder. As a consequence, the objective likelihood of being caught and punished for

44

Cooperation in Groups

murder is around 50% (Robinson & Darley, 1997). Studies have indicated that this level of likelihood of detection is sufficiently high to deter most possible murderers. Hence, murder is a crime that can be largely, although obviously not completely, controlled through a deterrence approach. However, that effectiveness occurs because society devotes a large amount of its societal resources to the effort to catch murderers. Many other types of rule breaking are similarly controllable using punishment-based strategies if a group is willing to devote sufficient group resources to the task. However, the level of sufficient resources for running such a system should not be underestimated. As an example, despite the use of considerable societal resources in the form of increased police efforts, deterrence strategies largely have been ineffective in securing public compliance with laws relating to drug use (MacCoun, 1993). Preventing certain behaviors apparently may require an inordinate amount of the group's resources. As we have noted above, punishment-based strategies are especially likely to be effective when social authorities are able to operate in controlled environments in which surveillance is possible. In work settings, for example, employees often can be closely observed by their supervisors or via television cameras or the monitoring of their telephone calls or locations. While the ability to control and observe an environment makes surveillance more effective, problems like employee theft remain major group problems even in these environments. In other environments, surveillance is almost impossible. Consider the problem of software and entertainment piracy (Tyler, 1996-1997). Because most stolen software and pirated music and films is in private homes, held by private individuals, it is very difficult to detect and punish the illegal use of such material. As a consequence, illegal use of such products is widespread. The second question we will consider is how the use of sanctioning systems influences the behavior of the person. As with incentive systems, sanctioning systems undermine people's intrinsic feelings of obligation to the group and to group rules. Consequently, they lessen people's internal motivation to obey rules and make it more necessary to use group resources for sanction-based approaches to shaping rule-following behavior in the future. Studies of regulatory authorities have emphasized that trying to regulate behavior by threats undermines people's commitment to rules and authorities (Frey, 1994; Frey & Oberholzer-Gee, 1997). The third question is how the use of sanctioning systems affects the group. The costs associated with pursuing a deterrence strategy are harmful for the group, since they lead to distortions in the nature of a group's allocation of resources away from productive uses. Groups have limited resources. Those resources can be used productively to develop and implement new projects, to educate group members, and to help those in need. The same resources also can be used to monitor, control, and incarcerate group members (Tyler & Boeckmann, 1997). The use of deterrence strategies introduces inefficiencies associated with the allocation of resources to monitoring and controlling group members. These lessen the ability of an organization or society to function effectively. Consider the case of government efforts to control crime using deterrence strategies. Resources devoted to armies, police forces, and prisons do not create value. They may be

Instrumental Motivations for Engaging in Cooperative Behavior

45

necessary to maintain social control, but it would be more beneficial to a group to maintain control in other ways and free up these resources for more productive purposes. This is not to say that sanctioning approaches cannot work. In fact they can, if sufficient resources are devoted to their implementation. For example, at the time of its collapse, East Germany had over 10% of its population in its police force, and had placed video monitors on almost all city street corners (G. Bierbrauer, personal communication, July 20, 1997). This large expenditure of resources to monitor people, while potentially effective in minimizing rule-breaking behavior, led to extreme inefficiency in the use of state resources and, hence, made the state itself inefficient. This inefficiency led to the collapse of the East German government when external resources no longer sustained it. The system worked, but its inefficiency made it unsustainable without external support.

Summary Our basic argument is that the use of instrumental strategies is a difficult and costly, but not necessarily ineffective, approach to securing cooperative behavior from group members. Instrumental approaches can be effective, but they have a number of difficulties and costs that lead them to be nonoptimal approaches to the problem of gaining cooperation. The question is whether there exist alternative models of social control that might be better. We will turn to that question in the next chapter.

D This Study In our current study, we assess instrumental judgments about a person's relationship to the group on three distinct levels. First, we asked people to indicate the general level of the resources they receive from their job (e.g., pay, benefits, long-term rewards and opportunities). The items used to make this assessment are shown in Table 4-1. They reflect people's overall judgments about their current and future resources from the group. That is, they assess the general level of instrumental connection between the individual and the group. When asked about the general level of resources received, people gave intermediate ratings. People also were asked to evaluate the degree to which they were sanctioned for not following group rules (sanctions) and the degree to which they were rewarded for doing their jobs well (incentives). These items directly examine the contingency between what people do in the context of the group and the rewards or punishments they receive. The items used to make these assessments also are shown in Table 4-1. People gave intermediate ratings on the scale reflecting their estimate of the likelihood of sanctions for noncompliance (M = 3.79), and slightly lower ratings on the scale reflecting the likelihood of being rewarded for doing their jobs well (M = 2.91). In other words, people were more likely to think that they would be punished for rule breaking than that they would be rewarded for doing their jobs well.

46

Cooperation in Groups

TABLE 4-1. Instrumental measures

M (SO) Total resources received from the group (6 =strongly agree, 1 = strongly disagree) Overall, I receive excellent pay and benefits where I work. If I wish to, I think I can have a secure job with my current work organization well into the future. I am satisfied with my pay. It would be very difficult for me to find another job with the pay and benefits of my current job. I have good opportunities for promotion where I work. My future opportunities for pay increases are not very favorable. 3 Costs of noncompliance (Sanctions) (6 =a lot, 1 = very little) How much attention does your supervisor pay to whether or not you follow work rules? If you were caught breaking a work rule, how much would it hurt you (e.g., you would lose pay or your chances for promotion would be hurt)? If you were caught breaking a work rule, how much would your supervisor care? Rewards for performance (Incentives) (6 =a lot, 1 = very little) If you do your job well, how much does that improve your pay and benefits? If you do your job poorly, how much does that hurt your pay and benefits? Outcome favorability of organizational rules & decision making How often do these rules and procedures lead to decisions and outcomes favorable to you? (6 =always, 1 =rarely) How much do the decisions your organization makes benefit you? (6 =a lot, 1 =not at all) Outcome favorability of supervisor's decision making How often does your supervisor make decisions that are favorable to you? (6 =always, 1 =rarely) How much do the decisions your supervisor makes benefit you? (6 =a lot, 1 =not at all)

3.85 (1.41) 4.37 (1.39) 3.48 (1.47) 2.77 (1.44) 3.34 (1.48) 3.55 (1.58) 3.86 (1.42) 3.28 (1.39)

4.24 (1.26)

2.82 (1.64) 2.99 (1.65)

3.77 (1.15) 3.65 (1.19)

4.36 (0.91) 4.21 (1.00)

aReversed item.

Finally, people were asked to indicate the degree to which the decisions made within the group favored them. They were asked this question in two ways: They were asked whether the rules of the group led to decisions that were favorable to them, and they were asked whether the decisions made by their particular supervisor favored them. These items capture the extent to which people feel that group-relevant processes tend to lead to positive outcomes; that is, when decisions are made in the group, they "win." Note that this is distinct from the assessments of sanctions and punishments, since it does not reference how their own behavior relates to their outcomes. Group rules might favor a person irrespective of whether or not they work hard, obey rules, or act in other ways. The items used to make each of these two assessments are shown in

47

Instrumental Motivations for Engaging in Cooperative Behavior

Table 4-1. In the case of organizational rules, people indicated that group rules are slightly favorable to them (M = 3.72), while they thought the decisions of their supervisors were somewhat more favorable to them than were the group's rules (M = 4.29). Table 4-2 shows the intercorrelation among the five indexes reflecting each aspect of people's instrumental relationships with their group. Interestingly, the correlation among these indexes was not especially high. The average correlation among all five indexes was 0.18. This suggests that there was not a single index of instrumentality and that this concept is multidimensional in nature. The degree to which people feel that they have a positive instrumental engagement with the group appears to depend on the manner in which this question is asked. We will examine all of these five levels of assessment during this analysis. We do so to maximize the possibility that, if instrumental effects occur, they will be identified. We outlined the four types of cooperative behaviors that are our ultimate concern in chapter 3. In Table 4-3, we examine the ability of the five instrumental indicators outlined to explain these behaviors. We include all five instrumental indicators in the analysis. Those indexes are: the overall level of resources that people feel they are receiving from the group, the subjective likelihood of sanctions for noncompliance, the subjective likelihood of incentives for performance, the degree to which the group rules lead to decisions favorable to the person, and the degree to which the person's supervisor makes decisions favorable to them. The analysis is shown in Table 4-3. It indicates that the instrumental indicators explain approximately 11% of the variance in cooperative behavior.

TABLE 4-2. Relationship among the instrumental scales Instrumental Scales

M (SD)

Total resources Sanctions Incentives Outcome favorabilityrules Outcome favorabilitysupervisor

3.55 3.79 2.91 3.72

Total resources

OF-

OF-

Sanctions

Incentives

rules

supervisor

(0.95) (1.09) (1.43) (1.05)

.72 .44 .37

.72 .24 .13

.67 .26

.77

4.29 (0.88)

.33

.01

.13

.42

.11

.83

= outcome favorability. Entries are Pearson correlations (n = 404). Diagonal entries are the alphas for each scale. Scales range from 1 to 6, with high scores indicating that: (a) one is receiving favorable rewards from the job, (b) the costs of punishment for rule breaking are high, (c) the likelihood of rewards for rule following are high, (d) organizational rules lead to favorable outcomes, and (e) the decisions of the employee's supervisor lead to favorable outcomes.

Note. OF

48

Cooperation in Groups

TABLE 4-3. Ability of the instrumental indicators to explain cooperative behavior Dependent variable Compliance

In-role

Deference

Extra-role

-.12* .26*** .1 7** .16* -.08

-.03 .09 .22*** .04 .16**

-.07 .24*** .11 .09 .11*

-.00 .07 .19** .03 .19**

12%

10%

10%

10%

Total resources Sanctions Incentives Outcome favorability-rules Outcome favorability-supervisor Total adjusted R2 *p