organizational citizenship behavior

Organizational Citizenship Behavior: Construct Redefinition, Measurement, and ValidationAuthor(s): Linn Van Dyne, Jill W. Graham, Richard M. DieneschSource: The Academy of Management Journal, Vol. 37, No. 4 (Aug., 1994), pp. 765-802Published by: Academy of ManagementStable URL: http://www.jstor.org/stable/256600Accessed: 14/10/2010 13:20

Your use of the JSTOR archive indicates your acceptance of JSTOR's Terms and Conditions of Use, available athttp://www.jstor.org/page/info/about/policies/terms.jsp. JSTOR's Terms and Conditions of Use provides, in part, that unlessyou have obtained prior permission, you may not download an entire issue of a journal or multiple copies of articles, and youmay use content in the JSTOR archive only for your personal, non-commercial use.

Please contact the publisher regarding any further use of this work. Publisher contact information may be obtained athttp://www.jstor.org/action/showPublisher?publisherCode=aom.

Each copy of any part of a JSTOR transmission must contain the same copyright notice that appears on the screen or printedpage of such transmission.

JSTOR is a not-for-profit service that helps scholars, researchers, and students discover, use, and build upon a wide range ofcontent in a trusted digital archive. We use information technology and tools to increase productivity and facilitate new formsof scholarship. For more information about JSTOR, please contact [email protected].

Academy of Management is collaborating with JSTOR to digitize, preserve and extend access to The Academyof Management Journal.

http://www.jstor.org

http://www.jstor.org/action/showPublisher?publisherCode=aom

http://www.jstor.org/stable/256600?origin=JSTOR-pdf

http://www.jstor.org/page/info/about/policies/terms.jsp

http://www.jstor.org/action/showPublisher?publisherCode=aom

? Academy of Management Journal 1994, Vol. 37, No. 4, 765-802.

ORGANIZATIONAL CITIZENSHIP BEHAVIOR: CONSTRUCT REDEFINITION, MEASUREMENT,

AND VALIDATION

LINN VAN DYNE Michigan State University

JILL W. GRAHAM RICHARD M. DIENESCH

Loyola University of Chicago

Organizational citizenship behavior (OCB) is reconceptualized in terms of civic citizenship as described in political philosophy. We used this theoretical foundation to derive substantive categories of OCB and to propose a nomological network of its potential antecedents. We also propose a new measurement of organizational citizenship behavior based on these substantive categories. Finally, data from 950 employees in diverse organizational and occupational contexts support the construct validity of the measurement of OCB proposed here.

Work behavior that is in some way beyond the reach of traditional measures of job performance but holds promise for long-term organizational success is receiving increasing theoretical attention as the challenge of global competition highlights the importance of organizational innovation, flex- ibility, productivity, and responsiveness to changing external conditions. In the last decade, many terms have been used to describe such behavior, including organizational citizenship behavior (Graham, 1991; Organ, 1988, 1990; Schnake, 1991), prosocial organizational behavior (Brief & Motowidlo, 1986), extra-role behavior (Van Dyne & Cummings, 1990), organizational spontaneity (George & Brief, 1992), and even counter-role behavior (Staw & Boettger, 1990). In addition and not surprising, researchers have employed diverse operational definitions of the new concept (e.g., Bateman & Organ, 1983; George, 1991; Graham & Verma, 1991; MacKenzie, Podsakoff, & Fetter, 1991; Moorman, 1991; Motowidlo, 1984; Motowidlo, Brief, Atieh, & Ash-

We gratefully acknowledge research support from the Social Sciences and Humanities Research Council of Canada (Grant 494-85-1018) and from a summer research grant from Loyola University of Chicago's School of Business Administration. Benjamin Foo, Paul Lim, Beverly Moon, David Shea, and Debbie Stayburg provided excellent technical support. Robert Bies and L. L. Cummings collected the data for the M.B.A. student sample and permitted its reporting here. Ron Anton, S.J., L. L. Cummings, Michael Keeley, Dennis Organ, and Judi McLean Parks offered helpful comments on earlier versions. The first two authors made equal contributions to this article.

765

Academy of Management Journal

worth, 1987; O'Reilly & Chatman, 1986; Organ & Konovsky, 1989; Pierce, Gardner, Cummings, & Dunham, 1989; Podsakoff, MacKenzie, Moorman, & Fetter, 1990; Puffer, 1987; Scholl, Cooper, & McKenna, 1987; Scholl & Mc- Kenna, 1983; Smith, Organ, & Near, 1983; Williams & Anderson, 1991; Williams, Podsakoff, & Huber, 1986).

A common theme of these diverse conceptualizations is an attempt to identify work behavior that contributes, at least in the long run, to organizational effectiveness, but which is sometimes overlooked by the traditional definitions and measures researchers use to assess job performance. Graham (1991) identified two fundamentally different approaches to conceptualizing these sometimes overlooked behaviors. The first approach proposes that organizational citizenship behavior (OCB) and traditional conceptualizations of job performance are separate constructs. Early OCB researchers defined citizenship behavior as separate from in-role job performance and emphasized that OCB should be viewed as both extra-role and organizationally functional (Bateman & Organ, 1983; Smith et al., 1983). Graham (1991) argued that these criteria put researchers in the awkward position of determining what is in-role and what is extra-role, an inconstant distinction that varies across persons, jobs, and organizations and over time and with circumstances for individual job incumbents. To avoid this difficulty, Graham proposed a second approach based on the theoretical heritage of civic citizenship research in philosophy, political science, and social history. From that perspective, civic citizenship is viewed as including all positive community-relevant behaviors of individual citizens. By extension, Graham argued that organizational citizenship can be conceptualized as a global concept that includes all positive organizationally relevant behaviors of individual organization members. Thus, this broader conceptualization of organizational citizenship includes traditional in-role job performance behaviors, organizationally functional extra-role behaviors, and political behaviors, such as full and responsible organizational participation, that typically have been omitted from previous studies of citizenship.

In this study, we proposed and tested a conceptualization of organizational citizenship behavior based on an application of political philosophy to organizational settings. Such an approach has two key advantages. First, the intellectual heritage of civic citizenship suggests substantive categories of organizational citizenship behavior. Second, political philosophy also suggests a nomological network of related constructs. In summary, this approach provides a theory-grounded definition of OCB, a more comprehensive set of substantive citizenship behaviors than has existed, and a theoretical foundation for future research on organizational citizenship.

THE ORGANIZATIONAL CITIZENSHIP BEHAVIOR CONSTRUCT

Graham's (1991) review of classical philosophy and modern political theory highlighted several beliefs and behavioral tendencies that together comprise what Inkeles described as the "active citizenship syndrome"

766 August

Van Dyne, Graham, and Dienesch

(1969: 1139). Three interrelated substantive categories of civic citizen responsibilities make up this syndrome. The first category, obedience, involves respect for orderly structures and processes. Responsible citizens recognize rational-legal authority and obey the law. The second category, loyalty, expands parochial welfare functions to include serving the interests of the community as a whole and the values it embodies. Loyal citizens promote and protect their communities and volunteer extra effort for the common good. The third category, participation, entails active and responsible involvement in community self-governance in whatever ways are possible under the law. Responsible citizens keep themselves well informed about issues affecting the community, exchange information and ideas with other citizens, contribute to the process of community self-governance, and encourage others to do likewise. Political philosophers dating back to Aris- totle have recognized that citizenship includes multiple related responsibilities and have emphasized that responsible civic citizenship requires balanced engagement in obedience, loyalty, and participation (Cary, 1977; Inkeles, 1969; Janowitz, 1984; Lane, 1965; Pateman, 1970; Rossiter, 1950; Wolin, 1960).

Graham (1991: 255) extended this political philosophy perspective on civic citizenship and applied the political categories of obedience, loyalty, and participation to citizenship in organizational settings. She positioned OCB as a global concept composed of several correlated substantive categories modeled after Inkeles's (1969: 1122-1123) definition of the active citizenship syndrome, defining the categories as follows: Organizational obedience reflects acceptance of the necessity and desirability of rational rules and regulations governing organizational structure, job descriptions, and personnel policies. Obedience can be demonstrated by respect for rules and instructions, punctuality in attendance and task completion, and steward- ship of organizational resources. Organizational loyalty is identification with and allegiance to an organization's leaders and the organization as a whole, transcending the parochial interests of individuals, work groups, and departments. Representative behaviors include defending the organization against threats, contributing to its good reputation, and cooperating with others to serve the interests of the whole. Organizational participation is interest in organizational affairs guided by ideal standards of virtue, validated by an individual's keeping informed, and expressed through full and responsible involvement in organizational governance. Representative activities include attending nonrequired meetings, sharing informed opinions and new ideas with others, and being willing to deliver bad news or support an unpopular view to combat "groupthink" (Janis, 1982).

A Nomological Network for Organizational Citizenship Behavior

Our extension of political philosophy suggests that organizational citizenship is a multidimensional construct that is composed of several different, but correlated, substantive categories. Here, we introduce and test this expanded conceptualization of organizational citizenship. Over time, re-

1994 767


searchers can develop separate and more detailed nomological networks for the citizenship categories, each of which most likely has somewhat different antecedents and consequences. Because at this time the conceptualization of citizenship based on political philosophy is in its early stages, it would be premature to hypothesize different relationships for each category. To facilitate the future development of separate nomological networks for each category, however, the discussion section of this article explores the multidimensionality of the citizenship construct by examining the different relationships revealed by the empirical analysis.

Political philosophy suggests that the nature of the relationship citizens have with their government is critical to their citizenship behavior. The active citizenship syndrome is based on covenantal relationship, which is characterized by open-ended commitment, mutual trust, and shared values (Bromley & Busching, 1988; Elazar, 1980; Graham, 1991; Graham & Organ, 1993; Grover, 1982). Covenants describe relationships of mutual commitment in which specific behaviors required to maintain the relationship or pursue common ends are not specifiable in advance (DePree, 1989). A covenant is "not a bargain but a pledge" (Rowley, 1962: 1515), a mutual promise by individuals to do their best to serve common values for an indefinite period. In contrast to contractual, exchange, or other instrumental relationships (Blau, 1964), covenants are existential; they focus on a state of being and involve intrinsically motivated effort rather than earning something or getting somewhere. In a covenantal relationship, "the partners do not auto- matically live happily ever after, but they are bound by covenant to struggle toward such an end" (Elazar, 1980: 10). This means that covenantal partners can disagree about particulars without threatening the existence of the relationship and can forgive each other should disappointing performance occur. The more strongly a person identifies with the collective entity (such as a particular relationship or community) and feels valued and values the connection, the less that individual will rely on legal sanctions to resolve difficulties (Macneil, 1985) and the more he or she will be an active contributor to the community (Almond & Verba, 1963; Verba & Nie, 1972). Thus, covenant is conceptualized as a reciprocal relationship based on ties that bind individuals to their communities and communities to their members (Kanter, 1968).

Macneil (1985) described a continuum that ranges from exchange contracts to relational contracts. The latter are open-ended, nonarticulated, and subjective, and they evolve over time (Rousseau & McLean Parks, 1993). Several different types of relational contracts, including covenants, social exchange, and psychological contracts, can be differentiated from each other and from transactional, or exchange, contracts (Bromley & Busching, 1988). Graham and Organ (1993) contrasted social exchange contracts and covenants. Social exchange is based on a long-run exchange of fairness that does not require a precise accounting and is based on reciprocity in the sense of diffuse obligations to reciprocate fairly. Trust and good faith guide the form and timing of reciprocating gestures and, according to Organ (1988), will

768 August


lead to citizenship behaviors. Covenants are a more extreme form of relational contract. A covenantal relationship is based on commitment to the welfare of both parties to the exchange and is also based on values. Thus, a covenantal relationship differs from a social exchange because the former is based on more than a general notion of fairness (Graham & Organ, 1993). Instead, covenantal relationship has a normative-affective foundation that differentiates it from other types of relational contracts (Etzioni, 1988; Gor- don, Anderson, & Bruning, 1992).

Rousseau (1989) defined a psychological contract as an individual's belief in a reciprocal obligation between self and organization. Both covenants and psychological contracts are special forms of contracts. Both are based on individual employee perceptions or beliefs regarding their cross- level and reciprocal relationships with their employing organizations. Both are unwritten and subjective. Both go beyond traditional economic exchange relationships and are based on ongoing relationships with indefinite param- eters. Both represent unarticulated and diffuse unilateral beliefs held by individual employees. Both covenant and psychological contract are based on more than traditional notions of organization commitment because both are based on beliefs about reciprocity (Rousseau, 1989). Covenants and psychological contracts differ, however, in that covenants imply acceptance and internalization of organizational values (Etzioni, 1988) but psychological contracts need not involve values (Rousseau, 1989).

Covenantal relationship is important in organizational contexts because it goes beyond influencing traditional affective states, such as satisfaction and commitment, and also influences behaviors that have long-run positive consequences for organizations. McLean Parks (1992) argued that the incom- pleteness of covenantal contracts and their emphasis on trust, mutuality, and shared values will lead to high levels of citizenship behavior, perhaps because their open-endedness and lack of specificity raise motivation and encourage internally driven (intrinsic) motivation. Graham and Organ (1993) described employee obedience, loyalty, and participation in organizational contexts that are results of covenantal relationship. These proactive behaviors represent the "responsibilities of citizenship" (Graham & Organ, 1993: 494) and are based on employee perceptions of mutuality in their relationships with their organizations.

Eisenberger, Huntington, Hutchison, and Sowa (1986) stressed the importance of reciprocity in employee-organization relationships and demonstrated that employee perceptions of organizational support (the commitment of an organization to its employees) were linked to employee attendance (a form of employee commitment to the organization). Building on this idea of reciprocity, Eisenberger, Fasolo, and Davis-LaMastro (1990) demonstrated relationships between perceived organizational support and attachment to an organization, performance, and innovation. In other words, the global perception that an organization supported its members and valued their contributions was an important correlate of employee behavior and affective states. Finally, Gordon and colleagues' (1992) empirical study dem-

1994 769


onstrated a relationship between perceptions of a partnership or two-way commitment between an organization and its employees and perceptions of co-worker citizenship behavior. Thus, the mutual affirmation, trust, and open-ended commitment that are characteristic of covenantal relationship have powerful motivational effects that create and release energy and effort to serve a collectivity. Within organizational contexts, this energy and effort can be expressed in various acts of organizational citizenship. Thus,

Hypothesis 1: Covenantal relationship will be positively associated with the substantive categories of organizational citizenship behavior.

Recently, researchers have begun to investigate the mediating processes that account for the relationship between OCB and its antecedents. For example, Moorman (1991) assessed the potential mediating role of satisfaction in the relation between fairness and citizenship. Similarly, Podsakoff and colleagues (1990) examined trust as a potential mediator of the relationship between leader behavior and citizenship, and Konovsky and Pugh (1994) demonstrated that trust mediated the effects of procedural justice on citizenship. Extending this work on the mediating processes that lead to citizenship behavior, we propose covenantal relationship as a mediator linking a variety of antecedents and citizenship behavior. Figure 1 presents this model.

In the model proposed herein, three basic types of antecedents have effects, mediated by covenantal relationship, on organizational citizenship. Following Podsakoff and colleagues' (1990) recommendation that research-

FIGURE 1 Model of Organizational Citizenship Behavior

+

4-

770 August


ers examine a variety of organizational citizenship behavior antecedents, we chose two variables to represent each of three basic types of antecedents in the model. The first is personal factors, which include an employee's affective state of satisfaction with a broad range of job-related dimensions and the individual's dispositional tendency to approach situations cynically. The second type of OCB antecedent is employee perceptions of situational factors in a workplace. Situational factors include perceptions of an organization's values and perceptions of the motivating potential of employee jobs. The third type of OCB antecedent is positional factors (Rousseau, 1978), which represent an individual's membership or position in an organization and include organizational tenure and hierarchical job level.

Although personal, situational, and positional factors are not an exhaus- tive set of categories, they do represent a broad variety of constructs with demonstrated relevance to employee behavior in organizations. Hypotheses 2 through 7 develop the relationships for representative variables from each of these three types of OCB antecedents. In each case, we propose covenantal relationship as a mediator. We note at this point our choice of the term antecedent rather than the term consequence or correlate. Although an experimental design or longitudinal research is required to demonstrate causality, each of the constructs we examined-job attitudes, disposition toward cynicism, organizational values, motivating jobs, organization tenure, and job level-is hypothesized to influence organizational citizenship behavior through the mediator of covenantal relationship and is thus is more likely to be an antecedent than a consequence of OCB.

Mediation of Citizenship Behavior

Personal factors: Positive job attitudes. Barnard (1938) suggested that satisfaction leads to individual willingness to cooperate and willingness to contribute to cooperative systems. A significant amount of empirical research has demonstrated that there is no consistent, positive relationship between job satisfaction and traditional measures of job performance (Bray- field & Crockett, 1955; Iaffaldano & Muchinsky, 1985). Organ (1977), however, argued that a consistent and positive relationship exists between satisfaction and broader conceptualizations of performance that include organizational citizenship behaviors in addition to more traditional measures of performance. In fact, past research has consistently demonstrated a relationship between positive job attitudes and citizenship behavior (Organ [1988: 44] reviews the empirical studies of satisfaction and OCB). Organ used social exchange theory to suggest that when employees are satisfied by their jobs, they reciprocate. This reciprocation includes attachment to the organization and behaviors such as organizational citizenship. Extending this reasoning, Pearce and Gregersen (1991) argued that the constraints placed on traditional measures of job performance by job descriptions and standard operating procedures make it more likely that employee reciprocity will occur as citizenship behavior rather than as traditionally measured aspects of job performance.

1994 771


Early research on organizational citizenship behavior, however, typically did not specify the process that related satisfaction and citizenship. Extending the more recent research on mediating relationships (Konovsky & Pugh, 1994; Moorman, 1991; Podsakoff et al., 1990) and combining it with the theoretical heritage of civic citizenship, we hypothesized that personal satisfaction regarding numerous aspects of a workplace, such as satisfaction with co-workers and supervisors, satisfaction with the work itself, and satisfaction with both immediate and long-term rewards, will promote covenantal relationship. As noted earlier, covenantal relationship is characterized by reciprocity and mutual commitment and the absence of predeter- mined inducements and contributions. By definition, citizenship behaviors are not all required by a job. When they occur, they are not explicitly rewarded, and when they do not occur, there is no punitive action (Organ, 1990). In other words, when a job meets an individual's personal needs (Locke, 1976), job satisfaction will lead to covenantal relationship, which will lead to citizenship behaviors. Thus,

Hypothesis 2: Positive job attitudes will affect the substantive.categories of organizational citizenship behavior through the mediator of covenantal relationship.

Personal factors: Cynicism. People vary in their capacity and inclina- tion to form relationships characterized by open-ended commitment and shared values. Cynics, for example, distrust others' motives and thus are disinclined to allow themselves to be vulnerable to others in open-ended relationships (Barber, 1983; Michalos, 1990). Although cynical attitudes may originate with regard to one target, they are often generalized across a variety of social and organizational targets (Bateman, Sakano, & Fujita, 1992; Crank, Culbertson, Poole, & Regoli, 1987; Regoli, Crank, & Rivera, 1990). Kanter and Mirvis (1989) found that cynicism can have pervasive and important effects on a variety of behaviors, including those in the workplace. Extending this view, we argue that those who are generally cynical about others will assess relationships at work in terms of their own personal advantage. As a result, these individuals will be unlikely to form a covenantal relationship and accordingly will engage in fewer citizenship behaviors. Thus,

Hypothesis 3: Cynical attitudes about human nature will negatively affect the substantive categories of organizational citizenship behavior through the mediator of covenantal relationship.

Perceived situational factors: Workplace values. Shared values are a critical component of covenantal relationship that distinguishes this type of relationship from traditional exchange relationships (Blau, 1964; Graham & Organ, 1993). Values that are socially sanctioned and noncontroversial (e.g., quality, innovation, cooperation, and wide participation) are easy to share (Fairholm, 1991) and can lead to close relationships, positive affect, and

772 August


attachment (O'Reilly, Chatman, & Caldwell, 1991; Posner, Kouzes, & Schmidt, 1985; Renesch & associates, 1992). For example, Argyris (1964, 1973) described the mutual benefits that accrue to employees and an organization when organizational values include respect for employees and their developmental needs. If employees believe that their organization values quality products or services, they will be more likely to become attached to it (develop covenantal relationship) and engage in behaviors that will contribute to high quality. If employees believe that their organization values participation, they will be more likely to feel as though their participation will make a difference. Consequently, they will be more willing to become attached and perform participative citizenship behaviors. If employees believe that their organization values innovation, they will be more likely to develop covenantal ties and feel as though they want to take the initiative in contributing their innovative or change-oriented ideas to the enterprise. Fi- nally, if employees believe that their organization values cooperation, they will be more likely to develop the embeddedness that characterizes covenantal relationship and subsequently emphasize cooperative behaviors. Overall, members who perceive that socially desirable values are an important part of their organization's culture will be more likely to relate to it in covenantal terms and consequently, to engage in citizenship behaviors. Thus,

Hypothesis 4: Perceptions of socially desirable workplace values will affect the substantive categories of organizational citizenship behavior through the mediator of covenantal relationship.

Perceived situational factors: Job characteristics. To the extent that a job is structured to provide regular feedback and autonomy as well as a sense of task completion, employees can monitor their own behavior and gain an increased sense of personal control (Greenberger & Strasser, 1986). Personal control is an individual's belief that he or she can effect a change, in a desired direction, on the environment. According to Lawler (1992), an increase in perceived control strengthens emotional bonds with an organization. Greenberger and Strasser (1986) wrote and Greenberger, Strasser, Cum- mings, and Dunham (1989) demonstrated that a heightened sense of personal control (such as that based on jobs with highly motivating characteristics) has positive consequences for employee attitudes and behaviors at work.

Motivating job characteristics like meaningful work, autonomy, and feedback "maximize the possibility for internal motivation" (Hackman & Oldham, 1976: 273). Thus, specific characteristics of a job can increase an employee's sense of felt responsibility and subsequently, the sense of attachment to the organization (Salancik, 1977). For example, understanding how one's job contributes to interdependent outcomes enhances feelings of embeddedness and accountability. Similarly, awareness of outcomes (feedback) can lead to a strong feeling of mutual responsibility, like that typically found

1994 773


in covenantal relationship. Proactive behavior, such as citizenship behavior, is likely to follow this heightened sense of responsibility and embeddedness generated by job characteristics. A job that allows a high degree of autonomy and the absence of close supervision or supervisory monitoring suggests a situation characterized by trust. Recent results of a study by Niehoff and Moorman (1993) support this view and demonstrate that monitoring negatively influences citizenship. Thus, the freedom associated with autonomy and low monitoring is balanced by the reciprocal response of responsibility and constructive behavior.

Hypothesis 5: Highly motivating job characteristics will affect the substantive categories of organizational citizenship behavior through the mediator of covenantal relationship.

Positional factors: Organizational tenure. Employees who have been with their employing organizations for a long time are more likely to have embedded relationships (Granovetter, 1985) and strong organizational ties (Becker, 1960; Mowday, Porter, & Steers, 1982; Rousseau, 1989; Rousseau & McLean Parks, 1993). In addition, those with longevity most likely have developed confidence and competence in performing their jobs and are likely to exhibit positive feelings and behavior toward their employing organizations (Organ & Greene, 1974; Salancik, 1977; Vivier, 1973). Meyer and Allen (1984) differentiated two types of commitment: Affective commitment is based on the positive force of attraction stemming from congruent values and emotional commitment to an organization, and continuance commitment is based on assessment of the costs of leaving or the scarcity of alternative jobs (Becker, 1960). When organizational longevity is based on a feeling of positive choice, it will increase affective ties to the organization (Mowday et al., 1982), such as covenantal relationship. When it is based on a feeling of lack of choice, individuals will develop attitudes consistent with their behavior (Festinger, 1964), and longevity will lead to "postaction justification of behavior" (Salancik, 1977) and "causal accounts" (Bies & Sha- piro, 1987), such as reporting a covenantal relationship.

Meyer, Allen, and Gellatly (1990) supported this reasoning by demonstrating that continuance commitment had a positive effect on the affective commitment of employees. In other words, especially for longer-tenured employees, accumulating binding investments to an organization can lead to self-justification or dissonance reduction through the development of affective attachment to the organization. Thus, attitudes become more consistent with behavior over time, and long-term employees report attitudes that generate legitimacy for their past behavior. Accordingly, we argue that individuals' organizational tenure will lead to their reporting stronger affective ties, such as those represented by covenantal relationship, and hence to employee citizenship behavior. In some cases, this behavior will be based on choice and positive affect (Mowday et al., 1982), and in others, on the need to justify or rationalize past behavior (staying with the organization) (Kiesler, 1971; Staw & Ross, 1987).

774 August


Hypothesis 6: High organizational tenure will affect the substantive categories of organizational citizenship behavior through the mediator of covenantal relationship.

Positional factors: Hierarchical job level. Studies of civic citizenship have consistently found socioeconomic status to be the single strongest pre- dictor of the active citizenship syndrome because high status tends to increase both the motivation and the ability to be actively involved (Campbell, Converse, Miller, & Stokes, 1966; Dahl, 1961; Milbrath & Goel, 1977; Verba & Nie, 1972). In organizations, those at high job levels generally have higher levels of organizational commitment than those at low levels (Mowday et al., 1982). This is because positions of power allow people to influence organizational decision making (Salancik, 1977), a high job level indicates high status (Sheldon, 1971) and formal authority (Hrebiniak, 1974) and perhaps competence, in that they may have had and rejected other employment alternatives (Becker, 1960), and their high level of responsibility shows that the organization recognizes their competence and values their contributions (Salancik, 1977).

In addition, those at higher levels of authority have more freedom concerning their behavior on the job. Lawler (1992) argued that choices like those available to high-level employees enhance an individual's sense of control over an environment and lead to increased affective attachment to an

organization. Similarly, Hrebiniak (1974) argued that high job level is associated with autonomy, opportunity for interaction, and decisional involvement, all of which increase attachment to the organization. Finally, those in high-level jobs may feel social pressure to report attachment to their organizations and may feel subject to expectations from both peers and subordinates that they will "go the extra mile." Thus,

Hypothesis 7: High hierarchical job level will affect the substantive categories of organizational citizenship behavior through the mediator of covenantal relationship.

In summary, the reconceptualization of organizational citizenship behavior based on political philosophy proposed here offers two key contributions. First, citizenship behavior has three distinct though related forms: obedience, which demonstrates respect for a rational structure of rules and regulations; loyalty, which displays allegiance to the organization as a whole; and participation, which is full and responsible involvement in organizational governance. Second, by positioning many of the antecedents of citizenship behavior as working through the mediating construct of covenantal relationship, political philosophy focuses attention on the importance in the organizational citizenship nomological network of reciprocity between organizations and their members.

METHODS

Data and Procedures

Results were based on questionnaire responses from six sets of respondents totaling 950 employees and 169 supervisors. Five of these sets were

1994 775


part of initial data collection effort, and the final data were collected later for cross-validation of the new instrument presented here, including assessment of construct validity, reliability, and factor structure (see Table 1). The original data set included respondents from a professional accounting firm (N =

90), a savings and loan organization (N = 131), a social service agency (N = 40), a construction products firm (N = 145), and students in an evening master's of business administration degree program (N = 132). All respondents participated voluntarily in this university research project that was approved by their managers or instructors. Assurances of anonymity were made and kept. Employee respondents at two of the organizations completed questionnaires at group meetings conducted on site by one of us. Employee respondents at the other two organizations and all supervisory respondents completed questionnaires on their own and returned them by mail. Overall, the response rate for these four organizations was 52.4 percent from employees and 30.7 percent from supervisors.

The cross-validation data were collected from 85 supervisors from 48 organizations who provided information on the organizational citizenship behavior of 412 subordinates. These data were used to assess the dimensionality and internal consistency reliability of the OCB instrument. A sub- group of the supervisors from this data set (N = 47, 55%) completed a second questionnaire on 233 employees four weeks after the first questionnaire to allow assessment of test-retest reliability. Of the 107 supervisors who initially agreed to provide data, 85 (79%) completed the first questionnaire and 47 (44%) completed both questionnaires.

TABLE 1 Breakdown of Samples and Analytic Procedures

Original Cross-Validation Type of Response Data Data Totals

Self-reporting employees 538a 538 Employees supervisors reported on 154b

Time 1 412c 566 Time 2 233d

Supervisors reporting on subordinates 84 Time 1 85 169 Time 2 47

Total employees 950e

a Number of employee self-reports for instrument development factor analysis. b Number of matched pairs of supervisor and subordinate responses for testing hypotheses

and initial internal consistency reliability. c Number of subordinates whose OCB was described by 85 supervisors for cross-validation

internal consistency reliability. d Number of subordinates whose OCB was described by 47 supervisors for test-retest reli-

ability. e Total employees, including 538 self-reports (154 matched pairs, 384 employee self-

reports) and 412 cross-validation supervisor reporting.

776 August


Both supervisor-reported and self-reported data on OCB were used to assess the reliability and factor analytic structure of the new organizational citizenship instrument. Tests of the relationships proposed in the nomological network were based on three field samples with matched responses from employees (N = 154) and their supervisors (N = 84). We used employee responses to assess the proposed antecedents in the model and supervisor responses to assess subordinate citizenship behavior. The number of obser- vations for hypothesis testing was reduced because managers at one research site declined to provide supervisory ratings and some supervisors at the other three organizations did not provide OCB ratings for their subordinates. Profiles of the 154 respondents for whom matched data were available in the original data set are as follows: 53 percent women; mean age, 38 years; average education, one to two years beyond high school; and average organizational tenure, 8.2 years. The following describes the 84 supervisors in the original data set who provided OCB data on their subordinates: 63 percent men; mean age, 40 years; average education, associate or technical degree; average organizational tenure, 8.6 years; and average work experience, 19 years. The profile of the supervisors in the cross-validation group is similar to that of the initial 84 supervisors: 55 percent men; mean age, 38 years; average education, a four-year college degree; average organizational tenure, 10 years; and average work experience, 18 years.

Construct Measurements

Using the three substantive categories suggested by political philosophy, we developed a new instrument to measure organizational citizenship behavior. Obedience was measured with 16 items describing conscientious work habits; we adapted and expanded these items from Smith and colleagues (1983). Loyalty and participation were measured by items developed for this study (16 for loyalty and 22 for participation) that were generated by focus group interviews held with employees from diverse job levels at three of the research sites. The initial OCB instrument contained 54 items that were measured on seven-point Likert scales. Approximately half the items were reverse-coded to avoid response set bias. Exploratory factor analysis on the original data and confirmatory factor analysis on the cross-validation data (described below) resulted in retention of 34 items.

The OCB antecedents represented in Figure 1 and described by Hypoth- eses 1-7 were measured by well-established, validated scales when such were available. The Appendix gives the items for new or uncommon measures. Covenantal relationship was measured with 23 items that represent a unique combination of cross-level relationships designed to capture the re- ciprocality of covenants. The first part of this measure (8 items) represents employee perceptions of an organization's relationship (or covenant) with employees in general. Five of these 8 items are from Bass's (1985) individualized consideration measure of leader behavior toward employees, and the other 3 items are from the Index of Organizational Reaction's measure of company identification (Dunham, Smith, & Blackburn, 1977; Smith,

1994 777


1976). The second part of the measure (15 items), representing an employee's relationship with an organization, was measured by Mowday and colleagues' (1982) Organizational Commitment Questionnaire, which measures shared values and goals, intent to remain a member of an organization, and willingness to exert extra effort on its behalf. Together, the items in these two subscales (see the Appendix) measure the reciprocal relationship between an employee and an organization. Because the number of items in the two subscales is unequal, the 8 items representing the organization's relationship and 15 items representing the employee's relationship were each standard- ized and then averaged to form the combined measure of covenantal relationship. The reliability (Cronbach's alpha) of the 23-item scale was .90, indicating a unidimensional construct that captures employee perceptions about the mutual relationship between employees and their organizations.

Overall job satisfaction was measured by 37 items from the Index of Organizational Reactions assessing satisfaction with supervision, coworkers, kind and amount of work, physical working conditions, pay, and career prospects (Dunham et al., 1977; Smith, 1976). Cronbach's alpha for this measure was .93. We measured cynicism by summing responses to three dichotomous response questions and then dividing by three to create a con- tinuous variable with a range of 1 to 2. These questions (see the Appendix) were taken from the General Social Survey (Davis, 1980), and they assess a general predisposition toward a cynical view of human nature. The Cron- bach's alpha was .74. Workplace values were assessed by 12 questions developed for this study concerning employee perceptions of the value their organizations placed on quality, innovation, cooperation, and wide participation in decision making (see the Appendix). Cronbach's alpha was .89. We used the Job Diagnostic Survey (Hackman & Oldham, 1975, 1976) to measure job characteristics, which we then combined multiplicatively to calculate the motivating potential score (MPS) for each respondent's job. Cronbach's alpha was .87. Organizational tenure was reported by respondents and was measured in months. We determined job level (low, medium, or high) by discussing job content with knowledgeable organizational informants.1 Ta- ble 2 contains scale reliabilities, descriptive statistics, zero-order correlations for the interval variables, and rank order correlations for the ranked variable of hierarchical job level based on the sample of 154 employees and supervisor ratings of their organizational citizenship behavior.

ANALYSES AND RESULTS

We first address psychometric properties of the citizenship scale and then report tests of the hypothesized relationships.

1 This procedure involved discussing job content for each general type of job included in the data and reaching a joint classification decision. The organizational informants provided the specific content knowledge for each job and the researcher explained the desired meaning of low, medium, and high and provided continuity in determining these classifications across the informants.

August 778

TABLE 2 Descriptive Statistics and Correlationsa

Variables Means s.d. 1 2 3 4 5 6 7 8 9 10 11

1. Obedience 5.60 1.08 (.88) 2. Loyalty 4.93 1.08 .51*** (.84) 3. Social

participation 4.60 0.99 .33*** .59*** (.68) 4. Advocacy

participation 4.96 1.05 .38*** .54*** .57*** (.86) 5. Functional

participation 5.35 1.18 .54*** .57**7*** 6*** .65*** (.75) 6. Covenantal

relationship 0.03 0.88 .14t .41*** .25*** .21** .32*** (.90) 7. Job satisfaction 3.58 0.52 .15* .32*** .17* .20** .28*** .76*** (.93) 8. Cynicism 1.19 0.32 -.05 -.13t -.10 -.09 -.24*** -.26*** -.20** (.74) 9. Values 4.91 1.02 -.01 .28*** .15* .09 .15* .75*** .66*** -.20** (.89)

10. Motivating potential score 160.21 65.96 .02 .llt .06 .16* .18* .34*** .46*** -.25*** .38*** (.87)

11. Tenure 97.95 110.36 .03 .17* -.05 -.05 -.14t .33*** .30*** .06 .34*** .03 12. Job level 2.18 0.73 -.05 .20*** .15** .21*** .12* .13* .25*** -.11t .25*** .33*** .04

a N = 154. Cronbach alphas are shown on the diagonal for all scales; a rank order correlation is reported for job level. t p < .10 * p < .05

** p < .01 *** p < .001

c-: CD

gP

r

C

CD

0

c- 0

CD

0

tz Cn


Psychometric Properties of the OCB Scale

Factor analysis. Self-report responses of employees (N = 538) to the 54-item OCB scale were factor-analyzed using principal component extrac- tion. We specified oblique rotation for the analysis because theory suggests that good citizenship is demonstrated by high levels of all three substantive categories of organizational citizenship behavior. Thus, obedience, loyalty, and participation should be positively correlated rather than orthogonally related.

Five factors met the selection criteria of eigenvalues greater than 1.0 and inclusion of at least three items. In addition, interpretation of the scree test suggested five factors. Finally, a five-factor solution also provided the most conceptually interpretable factor structure (Kim & Mueller, 1978). We de- leted items with multiple loadings (indicated by a differential of less than .20 between factors), leaving a total of 37 items. Table 3 lists the 54 items, their loadings, and the theoretical OCB category that each item was initially intended to measure, and Table 4 gives summary factor statistics.

Factor 1 contains 7 loyalty items, representing allegiance to an organization and promotion of its interests. Factor 2 contains 10 obedience items, representing respect for the rules and policies of an organization and willingness to expend appropriate effort on its behalf. The consistency of the items and the magnitude of the loadings for factors 1 and 2 provide strong empirical support for the substantive categories of loyalty and obedience.

The empirical results for the third substantive category of organizational citizenship behavior, participation, are more complex. Factors 3, 4, and 5 all reflect participation, but in differentiated modes. Factor 3 is the most diverse of the five factors in terms of the original substantive categories its items represent because it contains two participation items, two loyalty items, and one obedience item. All of these items, however, can be interpreted as concerning forms of participation that are noncontroversial and that involve interaction with other individuals. Examples include attending meetings, engaging in positive communications with others, and involvement in other affiliative group activities. In other words, these items describe participation in the form of interpersonal and social contact. In the political realm, similar behaviors include voluntary participation in civic boards and commissions, commemorative occasions, community social events, neighborhood coali- tions, and public ceremonies. In organizational settings, these are important interpersonal and affiliative behaviors. Therefore, we called factor 3 "social participation."

The eight items of factor 4 were all initially designed to measure participation. They describe innovation, maintaining high standards, challenging others, and making suggestions for change-behaviors targeted at other members of an organization and reflecting a willingness to be controversial. In the political realm, these are the behaviors of a political activist or social

August 780

TABLE 3 Results of Factor Analysis

~~OCR~~B ~Factorb Items Categorya 1 2 3 4 5

Retained items Represents organization favorably to outsiders Does not go out of way to defend organization against outside threatsc Does not tell outsiders this is a good place to workc Does not defend organization when employees criticize itc

Actively promotes organization's products and services Would accept job at competing organizations for more moneyc Would not urge coworkers to invest money in organizationc Rarely wastes time while at work Produces as much as capable of at all times

Always comes to work on time

Regardless of circumstances, produces highest quality work Does not meet all deadlines set by organizationc Is mentally alert and ready to work when arrives at work Follows work rules and instructions with extreme care Sometimes wastes organizational resourcesc

Keeps work area clean and neat Sometimes misses work for no good reasonc Only attends work-related meetings if required by jobc Shares ideas for new projects or improvements widely Keeps informed about products and services and tells others Works so personal appearance is attractive and appropriate Is not involved in outside groups for benefit of organizationc Frequently makes creative suggestions to coworkers Uses professional judgment to assess right/wrong for organization Encourages management to keep knowledge/skills current

Encourages others to speak up at meetings Helps coworkers think for themselves Keeps well-informed where opinion might benefit organization Does not push superiors to perform to higher standardsc

Loyalty Loyalty Loyalty Loyalty Loyalty Loyalty Loyalty Obedience Obedience Obedience Obedience Obedience Obedience Obedience Obedience Obedience Obedience

Participation Participation Loyalty Obedience Loyalty Participation Participation Participation Participation Participation Participation Participation

.75

.68

.68

.68

.60

.59

.54

.13

.27

.02

.09 -.09

.12

.05 -.03 -.08

.11

.06

.01

.12

.06

.07 -.02

.02

.12

.15 -.10

.17

.10 -.05

.05

.00

.05

.02 -.08

.66

.64

.60

.55

.53

.53

.52

.52

.50

.39 -.13 -.02

.16

.27 -.05

.00 -.01

.05

.16

.03

.02

- .04 .10

-.06 .05 .03

- .02 .12 .00

-.11 - .04 - .05 - .08

.05

.07

.14

.06 - .01

.67

.65

.62

.61

.48

.16

.02

.08

.08

.09

.09

.05

.09 -.19 - .03

.23 - .04 - .02

.10

.06 - .02

.14 - .08

.07

.09 - .07 - .02 -.13 - .04

.30

.26 -.12

.13

.66

.58

.56

.55

.55

.50

- .04 .14 .15 .10

- .08 - .13

.08 - .08

.08

.04 - .02

.23

.05 - .14

.13 - .05

.09

.26 - .04 -.15 - .18

.07 - .01

.11

.03

.00

.00

.22 -.06 -.10 -.15 .43 .05

c-A CD co

CD

a

03

Q-

SD

CO 0 a-

TABLE 3 (continued)

~~~~~OCB ~Factorb Items Categorya 1 2 3 4 5

Does not pursue additional training to improve performanceC Avoids extra duties and responsibilities at workc Does not work beyond what is requiredC Volunteers for overtime work when needed Has difficulty cooperating with others on projectsc

Dropped items: Exploratoryd Shows up for work early Does not work as fast as possiblec Uses organizational property for personal useC Is guided by high professional standards Perseveres until problems are solved Maintains confidentiality of information Uses slow periods to do personal businessC Won't stay overtime to finish a job if not paidC Searches for new ideas to improve operations Reports wrongdoing by others Expresses opinions honestly even if others differ Lets the supervisor know before things go wrong Avoids telling the supervisor unpleasant thingsC When everyone else agrees, keeps any doubts to selfC Does not keep knowledge and skills up to dateC Discourages others from challenging the supervisorC Does not let the supervisor know when things go wrongC

Dropped items: Confirmatoryd Anticipates and solves problems before required Does not volunteer for committees unless instructedC Does not read company bulletins conscientiouslyC

Participation Loyalty Loyalty Loyalty Participation

Obedience Obedience Obedience Loyalty Obedience Loyalty Obedience Loyalty Participation Loyalty Participation Participation Participation Participation Participation Participation Participation

Participation Participation Participation

-.15 .14 .14 .25

-.01

.01 -.06

.13

.10 -.02

.21

.00

.09

.26

.15

.15

.12

.10

.09 -.08 -.17

.12

-.01 .08 .09

-.04 .12 .05 .09 .04

.44

.42

.37

.31

.28

.26

.43

.03

.02

.11

.07

.16 - .08 -.21

.14 -.10

.16

- .02 - .05

.03 - .02

.02

.40

.09

.29 - .05 - .10

.11

.48

.43

.20

.01 - .04 -.13 - .15 - .07 - .12

.37 - .13

.18 -.06

.24 .30

.05 .00

.16 .64

.08 .57 -.06 .56

.00 .47

.00 .39

- .02 .09

-.12 .31 .27 .03

- .20 -.16 -.46

.26

.31

.30

.18

.19

.27 - .05

.30

-.16 .24

-.10 .12 .01 .06

- .08 .31 .21 .09

- .09 .08 .28 .18 .43

-.12

.08

.55 .00

.09 .51

.01 .45

a Categorizations are preliminary.

b Boldface indicates primary factor loadings and loadings with a differential of less than .20. c Item was reverse-coded. d Items were dropped on the basis of the factor analysis identified.

C:

o

S:,

a 0

c0 iz~

on 00

Ch r-t


TABLE 4 Summary Factor Statistics

Percentage of Cumulative Percentage Factor Eigenvalue Variance Explained of Variance Explained

1 8.77 16.2 16.2 2 4.25 7.9 24.1 3 2.70 5.0 29.1 4 2.15 4.0 33.1 5 1.64 3.0 36.1

visionary. In organizational settings, these behaviors are typical of an internal change agent. Consequently, we labeled factor 4 "advocacy participation."

Factor 5 represents a third type of participation. Four of the included behaviors were originally conceptualized as participation, and three were initially intended to measure loyalty. Each of these seven items, however, describes a form of participatory contribution in which individuals focus on themselves rather than others in their organizations but yet contribute to organizational effectiveness. These personally focused behaviors include participation through performing additional work activities, self- development, and volunteering for special assignments. In the political realm, these behaviors suggest the highly committed, hard-working, and highly valued party member. In an organizational context, these behaviors are typical of a dedicated individual contributor whose commitment, self- development, and participation add value to the functioning of the organization. Thus, we labeled factor 5 "functional participation."

In summary, the results of the exploratory factor analysis generally support the three substantive categories of organizational citizenship derived from the broader context of civic citizenship. The data indicate, however, that the respondents in this study had a more complex conceptualization of participation than we originally anticipated, with responses indicating significant distinctions among social, advocacy, and functional participation.

The dimensionality and factor structure of the new OCB scale were also assessed with confirmatory factor analysis on the cross-validation data (Hay- duk, 1987). This approach follows the recommendations of Schwab (1980) and DeVellis (1991) that researchers collect cross-validation data for any new scale on a new sample in order to assess the dimensionality of the scale, ascertain whether previously obtained exploratory factor analysis results were the result of sample-specificity, and avoid unnecessary proliferation of constructs. Harris and Schaubroeck (1990) and James, Mulaik, and Brett (1982) noted that assessment of a measurement model allows examination of the relationship between latent variables (causes) and the manifest or measured variables (effects) that serve as indicators of the latent variables. Ac- cordingly, we estimated the theory-based five-factor model on the 37-item citizenship scale using maximum likelihood techniques with LISREL VII.

1994 783


The initial assessment produced a chi-square of 2,337.14 (df = 619) and a goodness-of-fit index of .76 (Joreskog & Sorbom, 1989). Modification index values indicated that we could improve the model by dropping three items (Long, 1983), and theoretical examination of these items supported their elimination. For example, the item "Does not volunteer for committees at work unless instructed to do so by the supervisor" was originally classified under functional participation by the exploratory factor analysis. This item just met the retention criteria set in the initial factor analysis and had rela- tively high cross-loadings of .30 and .24. More important, the qualifying clause in this item "unless instructed" most likely confounded responses to the question. DeVellis (1991) recommended that scale items should not include qualifying phrases because respondents can misinterpret them. The second item with problematic loadings was "Does not conscientiously read technical and administrative bulletins and newsletters." The negative word- ing of this item may have confused respondents and most likely did not communicate clearly. The third and final item, "Anticipates and solves problems before required," is conceptually different from the rest of the advocacy participation items because it involves nonchallenging, noncontroversial, and non-change-oriented behaviors. Accordingly, dropping the item provided a cleaner measurement of advocacy participation congruent with its theoretical definition.

The five-factor model, measured by 34 items, resulted in a chi-square of 1,803.36 with 517 degrees of freedom and a goodness-of-fit index of .80. In addition, the difference-of-chi-square test (Long, 1983) demonstrated that the five-factor model produced a significantly better fit than a more traditional one-factor model (X2 = 1,777.40112), and the result of dividing chi- square by the degrees of freedom (J6reskog, 1970) exceeded the threshold of five suggested by Wheaton, Muthen, Alwin, and Summers (1977) and the threshold of two or three Carmines and McIver (1981) suggested.

Internal consistency reliability. We calculated scale reliabilities on the 34-item scale for each of the five categories of organizational citizenship using supervisory ratings of 154 subordinates' organizational citizenship behavior from three of the research sites. Cronbach's alphas were .88 for obedience, .84 for loyalty, .68 for social participation, .86 for advocacy participation, .75 for functional participation, and .95 for the entire OCB scale.

Following the recommendation of DeVellis (1991) for new scale development, we also calculated Cronbach's alpha on the cross-validation data in order to assess the internal consistency reliability of the scale for a different set of respondents. Eighty-five supervisors completed the 34-item questionnaire on the organizational citizenship behavior of their 412 subordinates. Cronbach's alphas were .83 for obedience, .79 for loyalty, .68 for social participation, .84 for advocacy participation, .75 for functional participation, and .91 for the entire 34-item OCB scale.

Test-retest reliability. Temporal stability of the OCB instrument was assessed with responses from a subset of the 85 supervisors described above. Forty-seven supervisors completed the citizenship behavior instrument on

784 August


233 subordinates four weeks after the first data collection (the average number of days later was 29.67, s.d. = 7.96). Test-retest results were .81 for obedience, .88 for loyalty, .76 for social participation, .83 for advocacy participation, .83 for functional participation, and .92 for the entire 34-item OCB scale.2

Evidence of construct validity. The foundation provided by the theoretical framework we developed is a strength of this research, especially since many of the other instruments in the literature have been developed without explicit theoretical frameworks. For example, Schnake noted that Smith and colleagues' (1983) and Bateman and Organ's (1983) instruments do not "represent a comprehensive list of citizenship behaviors" (1991: 738) and said that a more theoretically grounded approach was required for research to proceed rapidly and efficiently. Although the theoretical foundation provided by political philosophy differentiates our research from other OCB research, it is still important to explore the construct validity of our instrument. This exploration involved three sets of analyses: (1) Assessment of the convergent validity in the cross-validation data of the overall OCB scale through comparison of supervisor responses to the new scale with supervisor responses to three general items designed to assess the overall citizenship behavior of subordinates, (2) confirmatory factor analysis of the new measure on the cross-validation data with factor loadings based a priori on the theory presented here, and (3) assessment of the relationships proposed in our preliminary nomological network (the hypothesis testing reported later in this article). Although none of these steps can prove the construct validity of an instrument (Nunnally, 1967; Schwab, 1980), we suggest that the pattern of results achieved provides strong preliminary support for the construct validity of this measure. Results of all three sets of analyses indicate that the instrument measures what it was intended to measure.

The first analysis focused on the convergent validity of the new measure and three items designed specifically for this research to measure overall citizenship behavior. Pearson product-moment correlation analyses of responses from 85 supervisors on 412 subordinates in the cross-validation data supported convergent validity here (r - .69, p < .001). Second, the confirmatory factor analysis described above also provides preliminary support for the construct validity of the new citizenship scale.

The third and final set of analyses concerning the construct validity of the new OCB instrument is the hypothesis testing of the preliminary nomological network. Of the three approaches we used to explore construct validity, this third approach is both the most critical and the most profound for two reasons. First, we based the hypotheses on the theoretical framework

2 We note that the test-retest reliability for the entire scale is higher than the test-retest reliability for the subdimensions of the scale. This is because the subdimensions include fewer items and consequently are more sensitive to changes in responses than the full 37-item scale.

1994 785


provided by political philosophy. Nunnally (1967), Carmines and Zeller (1979), Schwab (1980), and Kerlinger (1986) all argued that assessment of a proposed set of theory-based relationships (a nomological network) is the key to examining construct validity. More specifically, Schwab suggested that researchers should use theory to differentiate constructs and that construct redundancy would decrease if constructs were developed and tested on the basis of theory. A key advantage, therefore, of our approach is that it avoids atheoretical expansion of the dimensions of citizenship such as that noted by Schnake (1991) and Moorman and Blakely (1992). Second, the relationships tested in the hypotheses provide the beginning of a theory- based nomological network for citizenship research. Empirical examination of this network can provide a firm foundation for future research by suggesting different relationships for each of the theory-driven dimensions of citizenship based on political philosophy. In addition, it can provide a continuing theoretical framework for future research on organizational citizenship behavior by specifying a domain with boundaries that will limit the proliferation of dimensions of OCB and simultaneously facilitate examination of the hypothesized dimensions in more detail.

Tests of the Hypothesized Relationships in the Nomological Network

Hypotheses 1-7 were designed to test the construct validity of the new 34-item citizenship scale. Drawing on political philosophy, we hypothesized specific relationships in the preliminary nomological network of organizational citizenship. Thus, we present these statistical tests as part of the construct validation of the measure rather than as evidence of substantive relationships (Schwab, 1980). For this reason, we applied the significance criterion of p < .10 that Achen (1982) and Hartwig and Dearing (1979) recommended for exploratory research. We tested the hypotheses using supervisory ratings of organizational citizenship behavior and employee responses for all other variables. This approach avoids artificial inflation of relationships by common source variance or the demand characteristics and pressure for positive self-presentation that are sometimes associated with self-reported OCB.

Hypothesis 1 predicted that covenantal relationship would be positively associated with the substantive categories of organizational citizenship behavior. We calculated the zero-order correlation between covenantal relationship and the five OCB categories derived from the factor analysis; the resulting significant correlation coefficients (p < .10) ranging from .14 for obedience to .41 for loyalty support Hypothesis 1 and justify further analyses to explore the hypothesized mediated relationships between OCB and its antecedents specified in Figure 1.

Hypotheses 2-7 proposed that covenantal relationship would mediate the relationships between organizational citizenship behavior and six antecedents. We tested mediation using the three-step mediated regression approach Baron and Kenny (1986) recommended: First, the mediator is regressed on the independent variable; second, the dependent variable is re-

786 August


gressed on the independent variable; and third, the dependent variable is regressed simultaneously on both the independent variable and the mediator. According to Baron and Kenny, mediation is indicated for a independent variable-mediator-dependent variable relationship if the following conditions are met: The independent variable must affect the mediator in the first equation; the independent variable must affect the dependent variable in the second equation; the mediator must affect the dependent variable in the third equation; and, finally, the effect of the independent variable on the dependent variable must be less in the third equation than in the second equation (Baron & Kenny, 1986: 1177). Full mediation is supported if the independent variable has no significant effect when the mediator is controlled, and partial mediation is indicated if the independent variable's effect is smaller but still significant when the mediator is controlled.

Table 5 presents mediated regression results for each of the five categories of citizenship: obedience, loyalty, social participation, advocacy participation, and functional participation. For step 1, we regressed covenantal relationship (the mediator) on each of the six predicted antecedents of citizenship: job satisfaction, cynicism, values, motivating job characteristics, tenure, and job level. The results are significant (p < .05) in every case. For step 2, we regressed each of the five substantive categories of citizenship on each of its six predicted antecedents. Overall, 19 of the 30 regression coefficients are significant (p < .10). For step 3, we regressed each of the five OCB categories simultaneously on covenantal relationship and one of the six antecedents, repeating the process six times. Of the 19 relationships that were significant at step 2, 17 passed the test at step 3 of a significant coefficient for the mediator and a decrease in the magnitude of the coefficient for the independent variable. Of these relationships, 13 represent the predicted full mediation and 4, partial mediation.

Overall, the results of the three-step regression analysis basically support the presence of mediated relationships for four of the substantive categories of citizenship: loyalty, functional participation, social participation, and advocacy participation. There is no support, however, for a mediated relationship for obedience. As expected from political philosophy, the pattern of mediated relationships varies across the substantive categories of organizational citizenship behavior and provides further support for its multidimensionality (Nunnally, 1967). For example, no significant mediated relationships were found for obedience, but the effects of all six antecedents on loyalty were mediated by covenantal relationship. Five of these six relationships for loyalty are fully mediated (job satisfaction, cynicism, values, motivating potential score, and tenure), and job level has a partially mediated effect on loyalty.

The pattern of relationships also varies within the three categories of participation and, like the factor analysis results, supports three distinct dimensions of participation (Carmines & Zeller, 1979). Results provide strong support for mediated effects on functional participation through covenantal relationship. Full mediation is supported for job satisfaction, values,

1994 787


TABLE 5 Results of Mediated Regression Analysisa

Categories of Organizational Citizenship Behavior

Independent Social Advocacy Functional Variables Obedience Loyalty Participation Participation Participation

Job satisfaction Step 1 .794*** .794*** .794*** .794*** .794*** Step 2 .160* .384*** .213** .173* .278*** Step 3 .103 .055 -.041 .106 .088 Step 3 .072 .414*** .320* .085 .226t R2 .028 .211*** .083** .033t .090** F 2.13 19.89*** 6.78*** 2.52** 7.37***

Cynicism Step 1 -.305*** -.305*** -.305*** -.305*** -.305*** Step 2 -.040 -.144t -.118 -.097 -.228** Step 3 .008 -.002 -.034 -.050 -.153t Step 3 .156t .457*** .277*** .153t .242** R2 .024 .210** .083** .031 .105*** F 1.82 19.75*** 6.81*** 2.36t 8.71**

Values Step 1 .730*** .730*** .730*** .730*** .730*** Step 2 -.002 .300*** .175* -.015 .135t Step 3 -.248* -.077 -.074 -.300** -.165 Step 3 .335** .514*** .341** .389*** .410*** R2 .052* .212** .085** .070** .096*** F 4.12* 20.08*** 6.95** 5.63** 7.94***

Motivating potential Step 1 .343*** .343*** .343*** .343*** .343*** Step 2 .044 .162* .138t .137t .205* Step 3 -.009 .010 .045 .091 .121 Step 3 .157t .455*** .271*** .138t .249** R2 .024 .210*** .084** .036t .097*** F 1.82 19.76*** 6.88*** 2.77t 7.98***

Tenure Step 1 .328*** .328*** .328*** .328*** .328*** Step 2 .018 .169t -.050 -.054 -.136 Step 3 -.030 .039 -.152t -.139 -.271*

Step 3 .146 .398*** .311*** .258** .408*** R2 .019 .170*** .088** .062* .167*** F 1.15 11.88*** 5.67** 3.86** 11.65***

Job level Step 1 .190** .190** .190** .190** .190** Step 2 .003 .238*** .171** .246*** .168* Step 3 -.042 .197** .161* .182* .071 Step 3 .162* .422*** .256*** .136t .276*** R2 .025 .247*** .107** .060** .089*** F 1.95 24.47*** 9.02*** 4.79** 7.24***

a The beta at step 1 is the same for each independent variable (for example, .794 for job satisfaction). Step 1 represents the regression of covenant on job satisfaction with no inclusion of the dependent variable.

t p < .10 p < .05

* p < .01 ** p < .001

788 August


motivating potential, and job level, and partial mediation for cynicism. Mod- erate support is indicated for social participation, with full mediation indicated for job satisfaction, values, and motivating potential, and partial mediation indicated for job level. In contrast, results indicate only weak support for mediated effects on advocacy participation. Of the six relationships that have advocacy participation as the dependent variable, only two suggest that the independent variable influences advocacy participation through covenantal relationship: motivating potential has fully mediated effects and job level has partially mediated effects. In conclusion, results of these hypothesis tests support the conceptualization of OCB as multidimensional. The pattern of results also provides support for the mediating role of covenantal relationship for loyalty, functional participation, and social participation.

DISCUSSION AND RECOMMENDATIONS FOR FUTURE RESEARCH

The current study had two major purposes. The first was theory building-specifically, to redefine organizational citizenship behavior and develop a nomological network for the construct based on political philosophy. Our second major purpose was to develop an instrument to measure this expanded conceptualization of organizational citizenship and test the proposed mediated model. Political philosophy suggested that citizenship is multidimensional, comprising obedience, loyalty, and participation (Inke- les, 1969). The data in this study provide strong support for the multidimensionality of citizenship behavior in organizations. In addition, results show that respondents differentiated three forms of participation (social participation, advocacy participation, and functional participation) instead of the single generalized category Inkeles (1969) and Graham (1991) described.

We predicted a fully mediated model (Figure 1) in which the antecedents of organizational citizenship behavior would have no effects beyond those that operate through covenant. In general, excluding obedience, results support covenantal relationship's mediation between OCB and its antecedents. Overall, loyalty had the strongest and most uniform pattern of full mediation. These results seem logical given the initial conceptualization of loyalty as identification with and allegiance to an organization as a whole. Thus, personal factors, situational factors, and position in an organization influence loyalty behaviors through the quality of the relationship they create between an organization and its employees.

In contrast, obedience was generally not related to the antecedents in the model (with the exception of job satisfaction). In addition, even the relationship between job satisfaction and obedience was not mediated by covenant. Examination of the items in the obedience scale suggests that obedience may be closer to traditional definitions of performance. This may help to explain its low correlation with covenantal relationship (r = .14) and the absence of any mediated relationships with the antecedents of citizenship. We speculate that norms, socialization, job instructions, and explicit

1994 789


incentives may cause obedience directly and that covenantal relationship may not be required for obedience to occur. More generally, it is possible that structural contingencies and expectations operate directly on traditional conceptualizations of performance (perhaps including obedience) but that the antecedents considered in this model operate on broader, nontraditional conceptualizations of performance (such as loyalty and participation) through the affective and cognitive states represented by covenantal relationship.

Another possible explanation for the lack of results for obedience may be our choice of antecedents with a plausible link through covenantal relationship. Five of the six antecedents in the model were unrelated to obedience (cynicism, values, motivating potential, tenure, and job level). Perhaps a different set of antecedents would be related to obedience and hence could have mediated effects on obedience. Schnake (1991) suggested several constructs as potential antecedents to conscientiousness, which is similar to our dimension of obedience, including self-monitoring, need for achievement, need for social approval, and work ethic. In addition, we suggest that the absence of certain demands outside of work such as medical or health problems, a second job, night school, or significant family responsibilities could allow more personal investment in covenantal relationship and generate higher levels of obedience, such as an unusually high quality and quantity of work, excellent attendance and punctuality, and continuing focus of attention on the job rather than on outside concerns. In other words, we are speculating that commitments outside an immediate work environment may influence the amount of energy that an employee has available to invest in covenantal relationship and subsequently in behaviors that regularly exceed expectations.

The results for participation are more complex than we initially anticipated, in terms of both dimensionality and mediated relationships. We discuss the suggested multidimensionality of participation first and then consider the results of the hypothesis testing. In most organizations, most employees have some degree of task interdependence with other employees. In contrast, in the civic arena, although most citizens depend on the infra- structure for services such as water, sanitation, and police and fire protec- tion, structural interdependence does not require citizens to interact with one another to obtain public services. Tax paying, while required, is an individual act. Thus, participation in organizations may be more pervasive and more salient than participation in civic settings. This contextual contrast may be responsible for the complex and highly differentiated conceptualization of participation indicated by the factor analysis reported here.

Social participation and functional participation fit the mediated theoretical model better than advocacy participation, with most of their antecedents having mediated effects (four out of six for social and five out of six for functional). In contrast, job satisfaction, cynicism, values, and tenure have no mediated influence on advocacy participation. Perhaps the less affiliative

790 August


and potentially more controversial and change-oriented nature of advocacy participation is based on a different set of antecedents that may or may not include the mediating state of covenantal relationship. For example, it is possible that advocacy participation is a function of general affective attachment combined with dissatisfaction with some specific feature of a job, supervisor, or situation. Alternatively, advocacy participation within a particular work environment might be based on covenantal ties with a collectivity larger than the employing organization, such as the local or national community or the "common good." In this case, advocacy participation may represent principled organizational dissent (Graham, 1986), whereby an internal change agent attempts to influence an organization to align its policies and practices with general principles that serve the larger collectivity. Fi- nally, advocacy participation can be viewed as challenging behavior and may entail some degree of personal risk. Consequently, it would be useful to examine additional constructs as potential antecedents to advocacy participation. Perhaps covenantal relationship mediates the effects of self-esteem, risk propensity, or job involvement on advocacy participation.

Unlike many relationships found in psychology and organization behavior (Baron & Kenny, 1986), the relationships that emerged here tended to be fully rather than partially mediated. We speculate that the presence of full mediation indicates that citizenship is deeply embedded in many different aspects of the relationship between organizations and their employees. Three of the six proposed antecedents (job satisfaction, socially desirable organizational values, and motivating potential) consistently have fully mediated relationships, and a fourth antecedent (job level) has partially mediated effects. For job level, three of the four mediated relationships had direct effects on citizenship as well as indirect effects through covenantal relationship. Perhaps this finding is a result of the high level of loyalty and participation that is generally expected of employees in high-level jobs. If so, such individuals may be motivated both extrinsically, by the expectations of others and organizational incentives, and intrinsically, as a result of covenantal relationship.

Two hypothesized organizational citizenship behavior antecedents, personal cynicism and organizational longevity, were not strongly supported by the data in this study. Cynicism had only one fully mediated and one partially mediated relationship (with loyalty and functional participation, re- spectively), suggesting that, despite Kanter and Mirvis's (1989) observation, cynicism is not a strong variable in this model. Long organizational tenure had only one significant mediated relationship-with loyalty. A possible explanation for why tenure operating through covenantal relationship does not influence the other forms of OCB is that affective and continuance commitment may have offsetting consequences for covenantal relationship and citizenship behavior. This perspective challenges our earlier argument that those whose longevity is based on continuance commitment might account for their perseverance by reporting stronger affective ties and by justifying

791 1994


their tenure through positive behavioral acts. Perhaps longevity is more a consequence than an antecedent to covenantal relationship. Future research might examine this relationship in more depth.

Several important implications can be drawn from these results. First, covenantal relationship, which emphasizes mutual trust, shared values, and open-ended commitment, is an interesting alternative to traditional exchange relationships (e.g., Blau, 1964) between organizations and their employees. Future research could test such competing models of interaction. In addition, research might examine when covenantal relationships are more likely to develop than transactional relationships. Second, since covenantal relationship implies a long-term mutual commitment between employees and employers, fostering such a relationship might be an effective vehicle for combating excessive emphasis on short-term results. People in covenantal relationships with their organizations or supervisors should be more likely to contribute consistently to long-term organizational welfare. This contribution could take the form of organizational citizenship behavior or of another employee behavior, such as staying with an organization. Finally, it is also possible that an exchange relationship can develop into a covenantal relationship (or vice versa). Given the long-term benefits of covenantal relationship suggested by this study, future research might benefit from examining its development and evolution over time.

Future research might also explore the implications of the multidimensionality of organizational citizenship behavior. Although the five OCB substantive categories are correlated, employees may choose among the categories rather than engage equally in all forms of citizenship. These choices could be based on a variety of factors, such as personality differences, structural constraints, incentives and recognition, supervisory style, supervisory expectations, and organizational culture. Future research should examine when and under what conditions particular forms of OCB occur. The empirical results presented here could serve as a foundation for a more detailed and differentiated model of the relationships for each substantive category of OCB within the overall nomological network. In addition, future researchers should examine participation both theoretically and empirically in order to refine the forms of participation identified in this study. In particular, social participation (factor 3) requires additional work, given the more difficult interpretability and lower reliability it showed relative to the other forms of participation we assessed. Finally, future researchers should examine obedience in more detail to develop a better understanding of why its relationships with other constructs in the nomological network are so different from those of the other substantive categories of citizenship.

As with all studies, the design of the current study is subject to limita- tions. The study is correlational and did not involve manipulation of variables. Consequently, the results can not indicate causality. It is possible that the model is misspecified and that engaging in citizenship behavior causes covenantal relationship. On the other hand, it is unlikely that covenantal relationship causes any of the independent variables, given the high stability

792 August


of the personal, situational, and positional factors. Thus, if any causal arrows are in the opposite direction, they most likely could only be those for the relationships between covenantal relationship and the substantive categories of citizenship. In view of this limitation, however, future research should include experimental designs in which key variables can be manip- ulated and causality assessed under more tightly controlled circumstances.

A second limitation of the present design is the preliminary status of the nomological network. The model most likely does not include all relevant or important constructs. Accordingly, future research should include deeper theoretical development of the nomological network for covenantal relationship. In addition, future researchers should explore the dimensionality of citizenship in more detail and develop specific hypotheses for each dimension and a variety of different constructs in the network. Additionally, potential moderators that may influence the relationships found in this study should be examined. A third limitation of the study concerns the results of the confirmatory factor analysis. Even after deletion of three items, the goodness-of-fit index reached only .80, suggesting the citizenship scale can most likely be further improved through additional iterative work that may lead to a more effective factor structure.

The study has, however, some notable strengths. First, the conceptualization and operational definition of the new OCB instrument were theoretically driven. Theory was the source of the proposed dimensions. Thus, although the results of the confirmatory factor analysis suggested minor modifications, and the goodness-of-fit results indicated that the measure could be further improved, it is unlikely that the basic conceptualization and dimensionalization will expand to include additional major subdimensions. This conceptualization of OCB is in sharp contrast to some of the earlier work on citizenship, which, as Moorman and Blakely (1992) noted, expanded the number of citizenship dimensions beyond the initial conceptualization on an ad hoc basis without explicit theoretical grounding. In contrast, we based our instrument on political philosophy and initially designed it to sample all key aspects of the theoretically defined domain (Carmines & Zeller, 1979; DeVellis, 1991). Our use of substantive theory to determine what to measure should minimize expansion of the dimensions of citizenship.

A second major strength of the study is the inclusion of both primary, developmental data from 538 employees and cross-validation data from 412 employees. Both Schwab (1980) and DeVellis (1991) stressed the importance of testing the reliability and factor analytic structure of a newly developed instrument on new samples to assess construct validity. Analysis of the homogeneity of scale items, internal consistency reliability, test-retest reliability, and convergent validity characteristics of the measure all support its construct validity.

A third major strength of the study is that it goes beyond existing conceptualizations, providing a more diverse set of behaviors representing a richer, more comprehensive conceptualization of organizational citizenship

1994 793


than that found in other empirical research on citizenship. More specifically, most prior research has focused on noncontroversial citizenship behaviors with primary emphasis on altruism and conscientiousness (Bateman & Organ, 1983; Organ, 1988; Smith et al., 1983). Although these two behaviors are important and can represent positive contributions to organizational effectiveness, they represent a restrictive conceptualization of citizenship. More recent measures of organizational citizenship (e.g., MacKenzie et al., 1991; Podsakoff et al., 1990) have included the five dimensions proposed by Organ (1988) and have moved beyond the initial two dimensions. Although inclusion of these additional dimensions has expanded the conceptualization of citizenship, there has been no theoretical justification for their inclusion and the exclusion of other dimensions with potential relevance to organizational citizenship.

A more detailed conceptual comparison of the measure proposed here with the measures developed by Podsakoff and associates (MacKenzie et al., 1991; Podsakoff et al., 1990) illustrates the importance of both measures and the contributions of our measure. First, as mentioned above, the new measure is the only conceptualization of citizenship based explicitly on theory. Thus, ours is the only approach that can a priori specify the dimensionality of the construct and simultaneously limit its number of dimensions. Second, and more specifically relevant to the comparison, our measure enriches understanding of citizenship, going beyond simple, helpful, and cooperative acts to include change-oriented and challenging behaviors. Although Pod- sakoff and colleagues included the dimension of civic virtue, courageous communications that challenge norms or support unpopular views were not included; thus, their measurement is incomplete and demonstrates construct deficiency (Schwab, 1980). For example, Podsakoff and colleagues' (1990) and MacKenzie and colleagues' (1991) scales include items such as attending meetings and keeping informed but no items about expressing opinions about issues. In addition, none of their items suggest any attempt to change the status quo through active participation such as that represented by the advocacy participation dimension of the new measure. Only half the content domain Organ (1988) specified in his definition of civic virtue is represented.

Third, there is strong evidence of value in both approaches to conceptualizing organizational citizenship. The Podsakoff and associates approach includes a dimension of organizational citizenship, altruism, that we do not include. Political philosophy does not suggest altruism, or helping, as a dimension of citizenship. Closer examination of the civic citizenship focus of political philosophy provides insight into this difference. Civic citizenship concerns the behaviors of individual citizens that have ramifications for the state, not interpersonal behaviors that have consequences for interpersonal relationships. The research on organizational citizenship has typically conceived of altruism as interpersonal helping directed at co-workers and has made no direct links to organizational or group effectiveness. Because altruism may or may not have consequences for the larger entity of an organ-

794 August


ization, it is not included in the dimensionality of civic citizenship based on political philosophy or in the new measure of organization citizenship. Thus, although there is some overlap in the two instruments, they do not duplicate each other. Instead, both scales make important contributions and assess somewhat different subdimensions of organizational citizenship behavior. Podsakoff and colleagues' (1990) scale may be especially suitable for use in situations in which interpersonal relations, cooperation, and team- work are critical issues or success factors, and our instrument may be appropriate when change and innovation are salient, perhaps because of external environmental demands or competitive pressures. This intriguing discussion suggests the need for more research (with an emphasis on theory development) like that of Moorman and Blakely (1992), who compared and integrated these two streams of citizenship research.

In conclusion, our empirical analysis generally supported our theory building. The data indicate that the theory-based multidimensional organizational citizenship behavior construct provides a framework for examining specific work behaviors that are often excluded from traditional measures of job performance. In addition, the data generally support the important mediating role of covenantal relationship. Finally, the study illustrates the benefits of a cross-disciplinary approach to theory building by demonstrating the richness that can be added to conceptualizations of constructs such as organizational citizenship by considering frameworks from other disci- plines.

REFERENCES

Achen, C. H. 1982. Interpreting and using regression. Newbury Park, CA: Sage.

Almond, G. A., & Verba, S. 1963. The civic culture: Political attitudes and democracy in five nations. Princeton, NJ: Princeton University Press.

Argyris, C. 1964. Integrating the individual and the organization. New York: Wiley.

Argyris, C. 1973. Personality and organization theory revisited. Administrative Science Quar- terly, 18: 141-167.

Barber, B. 1983. The logic and limits of trust. New Brunswick, NJ: Rutgers University Press.

Barnard, C. I. 1938. The functions of the executive. Cambridge, MA: Harvard University Press.

Baron, R. M., & Kenny, D. A. 1986. The moderator-mediator variable distinction in social psychological research: Conceptual, strategic, and statistical considerations. Journal of Per-

sonality and Social Psychology, 51: 1173-1182.

Bass, B. M. 1985. Multifactor leadership questionnaire: Form 5, revised. State University of New York at Binghamton.

Bateman, T. S., & Organ, D. W. 1983. Job satisfaction and the good soldier: The relationship between affect and "citizenship." Academy of Management Journal, 26: 587-595.

Bateman, T. S., Sakano, T., & Fujita, M. 1992. Roger, me, and my attitude: Film propaganda and public opinion toward corporate leadership. Journal of Applied Psychology, 77: 768-771.

Becker, H. S. 1960. Notes on the concept of commitment. American Journal of Sociology, 66: 32-40.

Bies, R. J., & Shapiro, D. L. 1987. Interactional fairness judgments: The influence of causal accounts. Social Justice Research, 1: 199-218.

1994 795

796 Academy of Management Journal August

Blau, P. 1964. Exchange and power in social life. New York: Wiley.

Brayfield, A. H., & Crockett, W. H. 1955. Employee attitudes and employee performance. Psy- chological Bulletin, 52: 396-424.

Brief, A. P., & Motowidlo, S. J. 1986. Prosocial organizational behaviors. Academy of Manage- ment Review, 11: 710-725.

Bromley, D. G., & Busching, B. C. 1988. Understanding the structure of contractual and covenantal social relations: Implications for the sociology of religion. Sociological Analysis, 49: 15-32.

Campbell, A., Converse, P. E., Miller, W. E., & Stokes, D. E. 1966. The American voter. New York: Wiley.

Carmines, E., & McIver, J. 1981. Analyzing models with unobserved variables: Analysis of covariance structures. In G. Bohrnstedt & E. Borgatta (Eds.), Social measurement: Current issues: 65-115. Beverly Hills, CA: Sage.

Carmines, E. G., & Zeller, R. A. 1979. Reliability and validity assessment. Newbury Park, CA: Sage.

Cary, C. D. 1977. The goals of citizenship training in American and Soviet schools. Studies in Comparative Communism, 10: 281-297.

Crank, J. P., Culbertson, R. G., Poole, E. D., & Regoli, R. M. 1987. The measure of cynicism among police chiefs. Journal of Criminal Justice, 15: 37-48.

Dahl, R. A. 1961. Who governs? Democracy and power in an American city. New Haven, CT: Yale University Press.

Davis, J. A. (Principal investigator). 1980. General social surveys, 1972-1980: Cumulative codebook. Chicago, IL: National Opinion Research Center, University of Chicago.

DePree, M. 1989. Leadership is an art. New York: Doubleday.

DeVellis, R. F. 1991. Scale development: Theory and applications. Newbury Park, CA: Sage.

Dunham, R. B., Smith, F. J., & Blackburn, R. S. 1977. Validation of the index of organizational reactions with the JDI, the MSQ, and Faces scales. Academy of Management Journal, 20: 420-432.

Eisenberger, R., Fasolo, P., & Davis-LaMastro, V. 1990. Perceived organizational support and employee diligence, commitment, and innovation. Journal of Applied Psychology, 75: 51-59.

Eisenberger, R., Huntington, R., Hutchison, S., & Sowa, D. 1986. Perceived organizational support. Journal of Applied Psychology, 71: 500-507.

Elazar, D. J. 1980. The political theory of covenant: Biblical origins and modern developments. Publius, 10(4): 3-30.

Etzioni, A. 1988. The moral dimension: Toward a new economics. New York: Free Press.

Fairholm, G. W. 1991. Values leadership: Toward a newphilosophy of leadership. New York: Praeger.

Festinger, L. A. 1964. Conflict, decision, and dissonance. Stanford, CA: Stanford University Press.

George, J. M. 1991. State or trait: Effects of positive mood on prosocial behaviors at work. Journal of Applied Psychology, 76: 299-307.

George, J. M., & Brief, A. P. 1992. Feeling good-doing good: A conceptual analysis of the mood at work-organizational spontaneity relationship. Psychological Bulletin, 112: 310-329.

Gordon, W. I., Anderson, C. M., & Bruning, S. D. 1992. Employee perceptions of corporate partnership: An affective-moral quid pro quo. Employee Responsibilities and Rights Jour- nal, 5: 75-85.

1994 Van Dyne, Graham, and Dienesch 797

Graham, J. W. 1986. Principled organizational dissent: A theoretical essay. In B. M. Staw & L. L. Cummings (Eds.), Research in organizational behavior, vol. 8: 1-52. Greenwich, CT: JAI Press.

Graham, J. W. 1991. An essay on organizational citizenship behavior. Employee Responsibil- ities and Rights Journal, 4: 249-270.

Graham, J. W., & Organ, D. W. 1993. Commitment and the covenantal organization. Journal of Managerial Issues, 5: 483-502.

Graham, J. W., & Verma, A. 1991. Predictors and moderators of employee responses to employee participation programs. Human Relations, 44: 551-568.

Granovetter, M. 1985. Economic action and social structure: The problem of embeddedness. American Journal of Sociology, 91: 481-510.

Greenberger, D. B., & Strasser, S. 1986. Development and application of a model of personal control in organizations. Academy of Management Review, 11: 164-177.

Greenberger, D. B., Strasser, S., Cummings, L. L., & Dunham, R. B. 1989. The impact of personal control on performance. Organizational Behavior and Human Decision Processes, 43: 29-51.

Grover, N. L. 1982. Contracts, covenants, and creative interchange. In J. A. Broyer & W. S. Minor (Eds.), Creative interchange: 291-306. Carbondale: Southern Illinois University Press.

Hackman, J. R., & Oldham, G. R. 1975. Development of the job diagnostic survey. Journal of Applied Psychology, 60: 159-170.

Hackman, J. R., & Oldham, G. R. 1976. Motivation through the design of work: Test of a theory. Organizational Behavior and Human Performance, 16: 250-279.

Harris, M. M., & Schaubroeck, J. 1990. Confirmatory modeling in organizational behavior/ human resource management: Issues and applications. Journal of Management, 16: 337- 360.

Hartwig, F., & Dearing, B. E. 1979. Exploratory data analysis. Newbury Park, CA: Sage.

Hayduk, L. A. 1987. Structural equation modeling with LISREL: Essentials and advances. Baltimore: Johns Hopkins University Press.

Hrebiniak, L. G. 1974. Effects of job level and participation on employee attitudes and perceptions of influence. Academy of Management Journal, 17: 649-662.

Iaffaldano, M. T., & Muchinsky, P. M. 1985. Job satisfaction and job performance: A meta- analysis. Psychological Bulletin, 97: 251-273.

Inkeles, A. 1969. Participant citizenship in six developing countries. American Political Sci- ence Review, 63: 1120-1141.

James, L. R., Mulaik, S. A., & Brett, J. M. 1982. Causal analysis: Assumptions, models, and data. Beverly Hills, CA: Sage.

Janis, I. L. 1982. Groupthink: Psychological studies of policy decisions and fiascoes. Boston:

Houghton Mifflin.

Janowitz, M. 1984. The reconstruction of patriotism: Education for civic consciousness. Chi- cago: University of Chicago Press.

Joreskog, K. G. 1970. A general approach for analysis of covariance structures. Biometrika, 57: 239-251.

Joreskog, K. G., & Sorbom, D. 1989. LISREL 7: A guide to the program and applications (2d ed.). Chicago: SPSS.

Kanter, D. L., & Mirvis, P. H. 1989. The cynical Americans: Living and working in an age of discontent and disillusion. San Francisco: Jossey-Bass.


Kanter, R. M. 1968. Commitment and social organization: A study of commitment mechanisms in utopian communities. American Sociological Review, 33: 499-517.

Kerlinger, F. N. 1986. Foundations of behavioral research (3d ed.). New York: Holt, Rinehart & Winston.

Kiesler, C. A. 1971. The psychology of commitment: Experiments linking behavior to belief. New York: Academic Press.

Kim, J., & Mueller, C. W. 1978. Factor analysis: Statistical methods and practical issues. Beverly Hills, CA: Sage.

Konovsky, M. A., & Pugh, S. D. 1994. Citizenship behavior and social exchange. Academy of Management Journal, 37: 656-669.

Lane, R. E. 1965. The tense citizen and the casual patriot: Role confusion in American politics. Journal of Politics, 27: 735-760.

Lawler, E. J. 1992. Affective attachments to nested groups: A choice-process theory. American Sociological Review, 57: 327-339.

Locke, E. A. 1976. The nature and causes of job satisfaction. In M. Dunnette (Ed.), Handbook of industrial and organizational psychology: 1297-1349. Chicago: Rand-McNally.

Long, J. S. 1983. Confirmatory factor analysis: A preface to LISREL. Beverly Hills, CA: Sage.

MacKenzie, S. B., Podsakoff, P. M., & Fetter, R. 1991. Organizational citizenship behavior and objective productivity as determinants of managerial evaluations of salespersons' performance. Organizational Behavior and Human Decision Processes, 50: 123-150.

Macneil, I. R. 1985. Relational contracts: What we do and do not know. Wisconsin Law Review: 483-525.

McLean Parks, J. 1992. The role of incomplete contracts and their governance in delinquency, in-role, and extra-role behaviors. Paper presented at the annual meeting of the Society of Industrial and Organizational Psychology, Montreal.

Meyer, J. P., & Allen, N. J. 1984. Testing the "side-bet" theory of organizational commitment: Some methodological considerations. Journal of Applied Psychology, 69: 372-378.

Meyer, J. P., Allen, N. J., & Gellatly, I. R. 1990. Affective and continuance commitment to the organization: Evaluation of measures and analysis of concurrent and time-lagged relations. Journal of Applied Psychology, 75: 710-720.

Michalos, A. C. 1990. The impact of trust on business, international security and the quality of life. Journal of Business Ethics, 9: 619-638.

Milbrath, L. W., & Goel, M. L. 1977. Political participation: How and why do people get involved in politics? (2d ed.). Lanham, MD: University Press of America.

Moorman, R. H. 1991. Relationship between organizational justice and organizational citizenship behaviors: Do fairness perceptions influence employee citizenship? Journal of Ap- plied Psychology, 76: 845-855.

Moorman, R. H., & Blakely, G. L. 1992. A preliminary report on a new measure of organizational citizenship behavior. Southern Management Association Proceedings: 185-187.

Motowidlo, S. J. 1984. Does job satisfaction lead to consideration and personal sensitivity? Academy of Management Journal, 27: 910-915.

Motowidlo, S. J., Brief, A. P., Atieh, J. M., & Ashworth, S. D. 1987. Correlates of prosocial behavior in sales performance. Working paper, Personnel Decisions Research Institute. Minneapolis.

Mowday, R. T., Porter, L. W., & Steers, R. M. 1982. Employee-organization linkages: The psychology of commitment, absenteeism, and turnover. New York: Academic Press.

798 August


Niehoff, B. P., & Moorman, R. H. 1993. Justice as a mediator of the relationship between methods of monitoring and organizational citizenship behavior. Academy of Management Journal, 36: 527-556.

Nunnally, J. C. 1967. Psychometric theory. New York: McGraw-Hill.

O'Reilly, C. A., & Chatman, J. 1986. Organizational commitment and psychological attachment: The effects of compliance, identification, and internalization on prosocial behavior. Jour- nal of Applied Psychology, 71: 492-499.

O'Reilly, C. A., Chatman, J., & Caldwell, D. F. 1991. People and organizational culture: A profile comparison approach to assessing person-organization fit. Academy of Management Jour- nal, 34: 487-516.

Organ, D. W. 1977. A reappraisal and reinterpretation of the satisfaction causes performance hypothesis. Academy of Management Review, 2: 46-53.

Organ, D. W. 1988. Organizational citizenship behavior: The "good soldier" syndrome. Lex- ington, MA: Lexington Books.

Organ, D. W. 1990. The motivational basis of organizational citizenship behavior. In B. M. Staw & L. L. Cummings (Eds.), Research in organizational behavior, vol. 12: 43-72. Green- wich, CT: JAI Press.

Organ, D. W., & Greene, N. 1974. The perceived purposefulness of job behavior: Antecedents and consequences. Academy of Management Journal, 17: 69-78.

Organ, D. W., & Konovsky, M. 1989. Cognitive versus affective determinants of organizational citizenship behavior. Journal of Applied Psychology, 74: 157-164.

Pateman, C. 1970. Participation and democratic theory. London: Cambridge University Press.

Pearce, J. L., & Gregersen, H. B. 1991. Task interdependence and extrarole behavior: A test of the mediating effects of felt responsibility. Journal of Applied Psychology, 76: 838-844.

Pierce, J. L., Gardner, D. G., Cummings, L. L., & Dunham, R. B. 1989. Organization-based self- esteem: Construct definition, operationalization, and validation. Academy of Manage- ment Journal, 32: 622-648.

Podsakoff, P. M., MacKenzie, S. B., Moorman, R. H., & Fetter, R. 1990. Transformational leader behaviors and their effects on followers' trust in leader, satisfaction, and organizational citizenship behaviors. Leadership Quarterly, 1: 107-142.

Posner, B. Z., Kouzes, J. M., & Schmidt, W. H. 1985. Shared values make a difference: An

empirical test of corporate culture. Human Resource Management, 24: 293-309.

Puffer, S. 1987. Prosocial behavior, non-compliant behavior, and work performance among commission salespeople. Journal of Applied Psychology, 72: 615-621.

Regoli, R. M., Crank, J. P., & Rivera, G. F. 1990. The construction and implementation of an alternative measure of police cynicism. Criminal Justice and Behavior, 17: 395-409.

Renesch, J., & associates. 1992. New traditions in business: Spirit and leadership in the 21st century. San Francisco: Berrett-Koehler.

Rossiter, C. 1950. Characteristics of the good citizen. Social Education, 14: 310-313, 319.

Rousseau, D. M. 1978. Characteristics of departments, positions, and individuals: Contexts for attitudes and behavior. Administrative Science Quarterly, 23: 521-540.

Rousseau, D. M. 1989. Psychological and implied contracts in organizations. Employee Re- sponsibilities and Rights Journal, 2: 121-139.

Rousseau, D. M., & McLean Parks, J. 1993. The contracts of individuals and organizations. In L. L. Cummings & B. M. Staw (Eds.), Research in organizational behavior, vol. 15: 1-43. Greenwich, CT: JAI Press.

Rowley, H. H. 1962. How to read the Bible with understanding: The diversity and unity of the

1994 799


scriptures. In H. G. May & B. M. Metzger (Eds.), The Oxford annotated Bible: Revised standard version: 1513-1516. New York: Oxford University Press.

Salancik, G. R. 1977. Commitment and the control of organizational behavior and belief. In B. M. Staw & G. R. Salancik (Eds.), New directions in organizational behavior: 1-54. Chicago: St. Clair Press.

Schnake, M. 1991. Organizational citizenship: A review, proposed model, and research agenda. Human Relations, 44: 735-759.

Scholl, R. W., Cooper, E. A., & McKenna, J. F. 1987. Referent selection in determining equity perceptions: Differential effects on behavioral and attitudinal outcomes. Personnel Psy- chology, 40: 113-124.

Scholl, R. W., & McKenna, J. F. 1983. Motivational effects of zero-sum and non zero-sum career lines. Paper presented at the annual meeting of the Western Academy of Manage- ment, Santa Barbara, CA.

Sheldon, M. E. 1971. Investments and involvements as mechanisms producing commitment to the organization. Administrative Science Quarterly, 16: 459-473.

Schwab, D. P. 1980. Construct validity in organizational behavior. In B. M. Staw & L. L. Cum- mings (Eds.), Research in organizational behavior, vol. 2: 3-43. Greenwich, CT: JAI Press.

Smith, C. A., Organ, D. W., & Near, J. P. 1983. Organizational citizenship behavior: Its nature and antecedents. Journal of Applied Psychology, 68: 653-663.

Smith, F. J. 1976. The index of organizational reactions. JSAS catalog of selected documents in psychology, 6: Ms. no. 1265.

Staw, B. M., & Boettger, R. D. 1990. Task revision: A neglected form of work performance. Academy of Management Journal, 33: 534-559.

Staw, B. M., & Ross, J. 1987. Behavior in escalation situations: Antecedents, prototypes and solutions. In L. L. Cummings & B. M. Staw (Eds.), Research in organizational behavior, vol. 9: 29-78. Greenwich, CT: JAI Press.

Van Dyne, L., & Cummings, L. L. 1990. Extra-role behaviors: In pursuit of construct and definitional clarity. Paper presented at the annual meeting of the Academy of Manage- ment, San Francisco.

Verba, S., & Nie, N. H. 1972. Participation in America: Social equality and political democ-

racy. New York: Harper & Row.

Vivier, J. F. 1973. Personal and organizational factors in career development. Psychologie Francaise, 18: 111-125.

Wheaton, B., Muthen, B., Alwin, D., & Summers, G. 1977. Assessing reliability and stability in panel models. In D. Heise (Ed.), Sociological methodology: 83-136. San Francisco: Jossey- Bass.

Williams, L. J., & Anderson, S. E. 1991. Job satisfaction and organizational commitment as predictors of organizational citizenship and in-role behaviors. Journal of Management, 17: 601-617.

Williams, L. J., Podsakoff, P. M., & Huber, V. 1986. Determinants of organizational citizenship behaviors: A structural equation analysis with cross-validation. Paper presented at the annual meeting of the Academy of Management, Chicago.

Wolin, S. S. 1960. Politics and vision: Continuity and innovation in western political thought. Boston: Little, Brown.


APPENDIX Original or Uncommon Questionnaire Itemsa

Covenantal Relationship (23 items) Part A: Organizational relationship with employees (8 items)

Individualized consideration My superior gives personal attention to subordinates who seem neglected. My superior delegates responsibilities to me to provide me with training opportunities. My superior treats each subordinate as an individual. My superior spends a lot of time coaching each individual subordinate who needs it. My superior gives newcomers a lot of help.

Company identification I think [organization name] considers employees much less important than sales and profits

(1) to much more important than sales and profits (5). How do you describe [organization name] as a company to work for? poor (1), just another

place to work (2), fairly good (3), very good (4), couldn't be much better (5). From my experience, I feel [organization name] probably treats its employees: poorly (1),

somewhat poorly (2), fairly well (3), quite well (4), extremely well (5). Part B: Employee relationship with organization (15 items)

Organizational Commitment Questionnaire

Organizational Values (12 items) High-quality products and services of central importance. Individual employees recognized and rewarded for superior quality. Reputation for quality surpasses major competitors. Innovative products and services of central importance. Individual employees recognized and rewarded for innovative work. Reputation for innovation surpasses major competitors. Widespread participation in decision making highly valued. Employees encouraged to express minority points of view. Procedures facilitate widespread participation in decision making. Cooperation among employees highly valued. Individual employees recognized and rewarded for helping others. Reputation as very friendly place to work compared other firms.

Personal Cynicism (3 items) la. In general people try to be helpful most of the time. lb. In general people just look out for themselves most of the time. 2a. In general most people would try to be fair. 2b. In general most people would take advantage of you. 3a. In general most people can be trusted. 3b. In general you can't be too careful in dealing with people.

a The new organizational citizenship scale (34 items) appears in Table 3.

1994 801


Linn Van Dyne received her Ph.D. degree from the University of Minnesota and is currently an assistant professor of management at Michigan State University. Her research interests focus on proactive employee behaviors involving initiative such as affiliative and challenging extra-role behavior, minority influence, and feedback- seeking behavior. She also researches group composition, employee attachment, cross- cultural issues, and human resources policy issues.

Jill W. Graham received her Ph.D. degree in organizational behavior from Northwestern University and is currently an associate professor of management at Loyola University of Chicago. Her research interests concern various forms of virtue in organizational life, including principled dissent, activist loyalty, servant leadership, and value-driven strategic management.

Richard M. Dienesch received his Ph.D. degree in organizational behavior from the Georgia Institute of Technology. His current research interests include leadership, conflict resolution, and cross-cultural issues in organizational behavior.

organizational citizenship behavior

Documents