Exit-voice and employee absenteeism: A critique of the industrial ...

Employee Responsibilitiesand RightsJournal Vol. 9, No. 2, 1996

Exit-Voice and Employee Absenteeism: A Critique of the Industrial Relations Literature A n d r e w A. L u c h a k 1 and Ian R. Gellatly 2

Industrial relations research that has examined the effects of collective voice Ol7 entployee absenteeism through traditional exit-voice ana&ses suffers from a number of important theoretical and empirical limitations. The research is limited theoretical& bz that the framework used cannot discrete& classify absenteeism as a fom, of either exit or voice. This inabiliO, reflects a larger problem with the exit-voice framework's lack of adequate attention to the conditions under which collective voice mechanisms fail and the consequences of sttch failure for the behavior of industrial relations actors. Exit-voice studies of absenteeism have important methodological problems, particularly in the way absenteeism and voice criteria have been operationalized. These defects h~ the literature can be addressed through the integration of behavioral theory and research on such topics as absenteeism, exit, voice, loyal~, and neglect, and the literature on organizational justice. KEY WORDS: exit, voice, loyalityand neglect; collectivevoice; employeeabsenteeism; organizational justice; dispute resolution.

INTRODUC~ON

One of the most influential accounts of collective bargaining as a mode of dispute resolution in mainstream industrial relations in North America is the exit-voice framework (Freeman & Medoff, 1984). Adapted from the initial work of Hirschman (1970), this framework argues that there are two principal methods by which employees may resolve employment-related disputes. First, there is the method of "exit-and-entry," by which individual employees leave less desirable for more desirable jobs. Quitting is the most visible manifestation of exit behavior. Alternatively, there is the method of "voice," by which employees can attempt to resolve an employment dispute through direct communication with the employer. 1Faculty of Business Administration, Memorial Universityof Newfoundland, St. John's, Newfoundland, Canada, A1B 3X5. 2Faculty of Management, The Universityof Lethbridge, Lethbridge, Alberta, Canada, T1K 3M4. 91 0892-7545/96,~K)04]091509.50,I) © 1996 Plenum Publishing Corporation

92

Luchak and Gellatly

Employee voice can be exercised individually or collectively. Individual voice may be expressed through such behaviors as speaking out, complaining and whistleblowing (Near & Miceli, 1986; Parmerlee, Near, & Jensen, 1982). An important vehicle for the expression of collective voice in modern societies is trade union representation through the process of collective bargaining. Voice through collective rather than individual bargaining has been argued to be more effective at resolving employment disputes because it provides employees with an important source of power that permits them to first, express discontent without fear of employer recrimination or reprisal, second, establish more favorable work conditions through the negotiation process, and third, challenge inequitable managerial decisions through a system of grievance-arbitration (Freeman & Medoff, 1984, pp. 8-11). Freeman and Medoff (1984) only applied the exit-voice framework to one behavioral response by employees to workplace disputes -- exit behavior. Consistent with expectations, collective voice was found to reduce employee exit behavior (Freeman & Medoff, 1984, p. 104). But there are a broad range of other behaviors that employees can engage in response to a workplace problem. Loyal job stayers may sit and wait things out, hoping that the employment dispute will resolve itself. Employees more frustrated with their circumstances may engage in any number of more destructive responses such as tardiness, absenteeism, poor performance, etc. Tests of the effects of collective voice on these other responses are important for establishing the broader generalizability of the exit-voice framework as a theory of dispute resolution. Industrial relations research subsequent to Freeman and Medoff (1984) extended the analysis of the exit-voice framework by examining the effects of collective voice mechanisms on other employee responses to workplace problems. This was particularly the case with respect to one potentially important and costly employee response to workplace p r o b l e m s - absenteeism. In this regard, industrial relations research has examined the effects of unions and/or collective bargaining processes on absenteeism under the exit-voice (Allen, 1984; Leigh, 1984; Markham & Scott, 1985; Wilson & Peel, 1991) and other contexts (Allen, 1981a; 1981b, Katz, Kochan, & Gobeille, 1983; Katz, Kochan, & Weber, 1985; Leigh, 1981; 1986). If the exit-voice framework is generalizable, employees having access to collective voice should be expected to make relatively less use of absenteeism as a form of temporary withdrawal or reduced effort because the collective voice mechanism should better address those sources of workplace discontent that might otherwise find expression through absenteeism behavior. But existing studies of the union impact on absenteeism do n o t bear this prediction out, thus casting doubt on the generalizability of the exit-voice framework as a theory of dispute resolution. Through a review of the literature on the effects of collective voice on employee absenteeism, it is the thesis of this article that the failure in the industrial relations literature to generalize exit-voice as a theory of dispute resolution, at least in respect of employee absenteeism, is due to fundamental limitations in the explanatory power of the framework as developed by Freeman and Medoff (1984) itself, and in contemporary efforts at operationalizing tests of the effects of

Exit-Voice and Employee Absenteeism

93

collective voice on employee absenteeism. These limitations in the present literature make any generalizations about the relationship of collective voice to acts of employee absenteeism of doubtful validity. However, rather than jettisoning the model and its application to employment disputes outright, it is contended that the model can be rejuvenated and more valid tests of its explanatory power made through the integration of behavioral theory and research on point. The article is organized into four parts. In the first and second parts, theoretical and empirical critiques of the exit-voice framework as applied to the study of employee absenteeism are provided, respectively. In the third part, recommendations for addressing some of these limitations through the integration of behavioral theory and research on absenteeism, exit, voice, loyalty and neglect, and organizational justice are provided. Special emphasis is placed on explaining how these sources of integration improve the explanatory power of the traditional exitvoice analysis. Finally, a discussion and conclusion completes the article.

THEORETICAL ISSUES An understanding of the effects of collective voice on absenteeism through the exit-voice framework is limited by at least two conceptual problems. First, absenteeism is not easily classified as a form of either exit or voice. Second, the exitvoice framework does not provide an adequate explanation of the conditions under which voice mechanisms might not work and the consequences of such failure for the behavior of industrial relations actors. Each point is considered in turn. A b s e n c e as

Exit or Voice?

An important implication of the exit-voice dichotomy is that the range of behavioral responses available to dissatisfied employees can broadly be conceptualized in terms of exit or voice. An examination of the definitions of exit, voice, and absenteeism, however, shows that absence is not easily classified as a pure form of either exit or voice. As alluded to above, "exit" refers to a set of behaviors that reflect an individual employee's attempt to permanently withdraw from an undesirable work environment, whereas voice is a direct and constructive attempt to change a dissatisfying work situation from within. Absenteeism, however, is much harder to define and understand. What is unclear is the extent to which absence behavior is determined by personal (e.g., age, sex, tenure), work (e.g., attitudes, job content, work group, socialorganizational), and nonwork (e.g., family responsibilities) factors (for reviews of this literature, see Chadwick-Jones, Nicholson, & Brown, 1982; Farrell & Stamm, 1988; Hackett & Guion, 1985; Johns & Nicholson, 1982, Marcus & Smith, 1985; Porter & Steers, 1973; Steers & Rhodes, 1978, Rhodes & Steers, 1990). Confusion as to the ideology and meaning of absence has made the task of classifying absence as exit or voice more difficult. On the one hand, absenteeism is like exit in that it represents an attempt to withdraw from and/or avoid an unpleasant work situation.

94

Luchak and Gellatl7

Absenteeism is unlike exit, however, in that it offers no readily identifiable solution to the employee's problem. On the other hand, absenteeism is like voice in that it communicates information to the employer. Absenteeism is unlike voice, however, in that the form of communication provided through absence is indirect and hard to discern given the many things that absence can mean (Allen, 1984). The failure to properly classify absenteeism reduces the predictive power of the exit-voice framework. More specifically, this failure reflects a more general limitation of the exit-voice framework to develop an adequate explanation of the conditions under which collective voice mechanisms might not work and a consideration of the consequences of such failure for the behavior of industrial relations actors. It is to this issue that the discussion now turns. When Collective Voice Fails: Determinants and Consequences

The exit-voice framework devotes limited attention to both the conditions under which collective voice mechanisms might not work and the consequences of such failure for the industrial relations actors. Rather, the effectiveness of collective voice is rooted in the untested assumption that it provides an effective alternative to exit by safeguarding employees from the fear of job loss (Freeman & Medoff, 1984, p. 9). But while the institution of collective bargaining provides unionized workers with important substantive and procedural guarantees in the area of employment security, the institution is far from being insular in this respect. This argument can be substantiated on a number of grounds. As a matter of practice, collective voice institutions would have to have full powers of co-determination over all terms and conditions of employment over which employees had any substantive or procedural interest in order to fully insulate their membership as mentioned. Absent such powers, it would be open to management to manipulate those terms and conditions of employment over which it had exclusive (or discretionary) control as a means of resisting or limiting employee's expression of collective voice. To be sure that North American unions do not have full powers of co-determination, one need only refer to the well-entrenched model of residual management rights, which vests in management the right to act unilaterally except as specifically limited by a collective agreement (Brown & Beatty, 1992). Inequities over issues not addressed under a collective agreement are not formally grievable under same in most North American jurisdictions. By the same line of reasoning, issues over which management retains a discretion are not easily challenged. The evaluation of employee performance or assessments of ability in matters of promotion are matters over which management typically maintains a high level of discretion, even in unionized establishments. The withholding of performance- and/or promotion-related benefits from employees that choose to exercise collective voice can be a powerful disincentive not to make use of their collective voice. And it is just this type of behavior that empirical studies on the consequences of use of grievance procedures in both the union (Klaas & DiNisi, 1989; Lewin & Peterson, 1989) and Nonunion (Lewin, 1987) sector are starting to

Exit-Voice and Employee Absenteeism

95

uncover. Specifically, there is a growing body of research finding that grievers face retaliatory action from their immediate supervisors in the form of lower performance ratings and promotion rates. Moreover, these effects are found to be exacerbated where employees actually win their grievances (Klaas & DiNisi, 1989; Lewin, 1987). While there is good reason to expect that dissident employees in unionized firms may be less likely to suffer retaliatory behavior from their supervisors than their nonunion counterparts, the important point to recognize is that modern day collective voice institutions do not have the power to fully protect their membership from this type of employer conduct.

METHODOLOGICAL ISSUES The uneasy application of the exit-voice framework to the study of employee absenteeism is both revealed in and compounded by empirical studies relating to absenteeism. These criticisms relate to the operationalization of absenteeism and collective voice. Each is considered in turn.

Methodological Problems with Absenteeism Criteria There has been a lack of sufficient attention give to the operational definition and measurement of absenteeism in the industrial relations research. Specifically, there has been a general failure in the literature to, first, maintain a distinction between voluntary and involuntary absenteeism, and second, recognize that absenteeism is but one behavioral response to undesirable work conditions. With respect to the first point, while it is generally recognized that it is controllable or voluntary employee absenteeism in response to perceived inequity that the exit-voice framework is attempting to predict, this distinction has not been well maintained in the empirical research. Rather, exit-voice studies of absence repeatedly use definitions of absence that intermingle or are unable to dissociate voluntary from involuntary forms of absence (Allen, 1984; Leigh, 1984; Markham & Scott, 1985; Wilson & Peel, 1991). This criticism finds support in the variety of different absence measures that have been used in the literature without any coherent rationale for selecting one over another. For example, absence has been measured in terms of: (a) a dichotomous measure indicating whether absent or not (Leigh, 1984); (b) absence duration (Leigh, 1981; 1986; Wilson & Peel, 1991); and (c) absence rates (Allen 1981a; 1981b; 1984; Markham & Scott, 1985). The lack of attention to the measurement of absenteeism is important because collective voice institutions may have differential effects on the type of absence behaviour under study. This is the case, for example, in studies of the effects of unionization on absenteeism. While exit-voice predicts union voice to reduce voluntary absenteeism, unionization has been associated with a positive effect on involuntary forms of absence (Allen 1981a; I981b; Leigh, 1981; 1986).

96

Luchak and

Gellatly

Second, the exit-voice literature fails to recognize that absence is but one potential behavioral response to perceived inequity at work. Other responses to dissatisfaction that could be equally destructive include, for example, lateness, error rates, shirking, sabotage, employee malfeasance, etc. (Farrell & Rusbult, 1992; Rusbult & Lowery, 1985; Rusbult, Farrell, Manious, & Rogers, 1988). This variation contributes "noise" to estimates of a collective voice-absence relationship because each of these behaviors is, to varying degrees, a substitutable response for sources of inequity in the workplace. The failure to control for these alternatives will attenuate the variance explained by any voice-absence relationship present.

Methodological Problems with Collective Voice Criteria Research on the effects of unions or collective bargaining processes have generally failed to provide the negative relation between collective voice and absenteeism predicted by the exit-voice framework. Rather, unionized workers have generally been found to have significantly higher average (Allen, 1984) or not significantly different (Leigh, 1984; Markham & Scott, 1985; Wilson & Peel, 1991) levels of absenteeism or absence rates, while grievance activity has been found positively associated with absenteeism (Katz, Kochan, & Gobeille, 1983; Katz, Kochan, & Weber, 1985). Both measures of union status or grievance activity have important limitations as proxies for collective employee voice. Union or collective agreement status measures of collective voice fail to capture the many alternative ways in which unionization and collective bargaining can influence workplace absenteeism (Allen, 1984; Leigh 1981; 1984; 1986). First, unions can increase or decrease absence through the offsetting income and substitution effects that accompany the well-known union impact on wages.Second, unions may influence absenteeism through their impact on the overall level and incidence of fringe benefits. For example, sick-leave benefits may be expected to increase absenteeism by reducing their opportunity costs. Third, unionization may reduce managerial flexibility in scheduling work, thus increasing the likelihood that absence is used as a substitute for individual preferences for more flexible work hours. Fourth, unionization may be associated with greater absenteeism if management response effects to unionization, such as the substitution of capital or nonunionized personnel for unionized labour, are perceived as a violation of the wage-effort bargain. Fifth, the greater due process and substantive rights of unionized employees in the area of discipline and discharge may increase absence behavior in unionized firms if these greater rights increase the difficulty of dismissing employees who, among other things, are habitually absent. FinalIy, unions may use absenteeism as a tactic of dispute resolution where the costs of absence as such are tow relative to available alternatives. Use of the volume of grievance activity as an indicator of active collective employee voice has a number of flaws (Ichniowski, 1986; Lewin, 1984; Peterson & Lewin, 1982). First, grievance rates provide no indication of the characteristics of a grievance such as the nature of the issue in dispute, time to settlement, etc. These issues can affect the degree to which voice through the grievance procedure acts

Exit.Voice and Employee Absenteeism

97

as a substitute for employee behaviors such as absenteeism. For example, grievances filed in response to unfair treatment, such as in matters of discipline and discharge, likely engender greater feelings of inequity and hence serve as a more ready substitute for absenteeism (Klaas, Heneman, & Olson, 1991). Where the settlement of a grievance is delayed, there is an increased probability of the dispute in question spilling into the workplace in various forms, including absenteeism behavior. Second, grievance rates provide no information about the effectiveness of the voice mechanism. Rather than reflect active union voice, a high level of grievance activity may simply reflect union or employee abuse of the grievance process for reasons unrelated to contract violations. For example, unions may abuse grievance procedures as a pretext for raising awareness about bargaining demands prior to the start of contract negotiations. Employees may also contribute to inflated levels of grievance activity where legal or political pressures, such as the duty of fair representation or an upcoming union election, force union officers and/or stewards to process grievances, respectively. Conversely, rather than reflect ineffective employee voice, a low level of grievance activity may simply reflect weak union representation and/or a general inability to police the collective agreement. Low grievance activity may also reflect employee unwillingness to use the grievance procedure because of fear of reprisal. Finally, grievance rates may reflect group rather than individual processes and so may be of doubtful validity as a determinant of individual behavior. For example, the grievance filing rate may not provide an indication of voice operating on behalf of employees who have not filed grievances but who have a clear interest in the outcome of a grievance. Present measures of collective employee voice through union or collective agreement status or through grievance activity introduce too much measurement error in estimates of collective voice effects on employee absenteeism. Future research must develop more refined measures of collective employee voice.

THE BEHAVIORAL RESEARCH Behavioral research on absenteeism, exit, voice, loyalty, and neglect and the literature on organizational justice can help address some of the limitations in the research on the effects of collective voice on absenteeism. First, the behavioral literature can help address problems in the industrial relations literature with respect to the measurement of absenteeism. This literature recognizes and cogently argues that it is the pattern of employee absence that may be the most relevant indicator of voluntary versus involuntary absence. Specifically, this literature concludes that measures of absence frequency (e.g., separate incidents) best reflect voluntary or culpable absence (Chadwick-Jones, Brown, Nicholson, & Sheppard, 1971; Hackett & Guion, 1985; Hammer & Landau, 1981; Muchinsky, 1977), whereas measures of absence duration (e.g., total days) best reflect involuntary or nonculpable absences (e.g., Chadwick-Jones, Nicholson, & Brown, 1982; Hackett & Guion, 1985). In the absence of more discriminating measures of voluntary and involuntary absence, future exit-voice studies should use frequency measures as the best approximation of voluntary absence.

98

Luchak and Gellatly

Second, the behavioral literature recognizes the unique character of absenteeism as a response to dissatisfaction at work by treating it as a separate response to traditional exit and voice alternatives (Farrell, 1983; Farrell & Rusbult, 1992; Rusbult & Lowery, 1985; Rusbult, Farrell, Manious, & Rogers, 1988; Withey & Cooper, 1989). Specifically, this literature conceptualizes absenteeism as part of the broader construct of neglect. This literature has demonstrated the uniqueness of the exit, voice, and neglect concepts, as well as that of loyalty, through the results of multidimensional scaling analyses (Farrell, 1983). These analyses reveal that employee reactions to dissatisfaction at work differ along two major dimensions. First, employees can respond to dissatisfaction in a manner that is either constructive or destructive of the employment relationship. Voice and loyalty responses are constructive of the employment relationship, whereas exit and neglect are not. Second, employees can respond to dissatisfaction in a manner that is either active or passive. Exit and voice concepts reflect active responses, whereas neglect and loyalty responses are more passive. This broader conception of potential employee responses to workplace disputes is an important source of integration for traditional exit-voice analysis because it makes explicit a route through which collective voice institutions might fail or perform their functions ineffectively. For example, and in the specific case of collective voice effects on absenteeism, voice institutions such as unions might be expected to reduce the use of employee absenteeism as a response to workplace problems not because they are assumed be more effective than individual voice but only where they are instrumental in this regard. If collective voice institutions are not instrumental in resolving workplace disputes, employees may choose neglectful responses, especially if exit is not a viable option and commitment (viz., loyalty) to the union-management relationship is low. Third, the behavioral literature recognizes the potential threats to the individual of engaging in alternative behavioral responses to dissatisfaction at work, including collective voice by explicitly modeling employee's instrumentality perceptions of exit, voice, and other response mechanisms (Farrell & Rusbult, 1992; Rusbult & Lowery, 1985; Rusbult, Farrell, Manious, & Rogers, 1988; Withey & Cooper, 1989). Thus for example, employees who face high exit costs or risk an adverse response from supervisors for using a formal complaint procedure are much less likely to use exit and voice responses, respectively (Withey & Cooper, 1989). Where existing mechanisms are considered ineffective or the perceived costs of using them are high, employees are more likely to respond to workplace disputes through neglectful activities such as reduced effort. The behavioral research has modeled the instrumentality of exit options through such measures as quality of job alternatives (Rusbult, Farrell, Manious, & Rogers, 1988), exit costs (Withey & Cooper, 1989), etc. The instrumentality of voice options has been modeled in this literature through such measures as voice costs (Withey & Cooper, 1989), employee attitudes toward the instrumentality of their grievance procedure (Allen & Keaveny, 1985), the approachability of supervisors (Saunders, Sheppard, Knight, & Roth, 1992), etc. Generally, the research finds the expected negative relation between the instrumentality of the voice mechanism used and absenteeism or neglectful behavior, especially where the costs of other response options are high or the perceived effectiveness of these other options is low. This approach provides a stronger test of collective voice effects on employee behavior than either union status or grievance activity measures and should be adopted in the industrial relations literature.

Exit-Voice and Employee Absenteeism

99

Fourth, behavioral research in the area of organizational justice offers a ready made source of integration with the exit-voice framework in the study of employee absenteeism. This literature makes a distinction between two broad types of organizational justice, distributive and procedural justice (Folger & Greenberg, 1985; Greenberg, 1990; Thibaut & Walker, 1975). Distributive justice refers to the fairness of outcomes that employees receive relative to their inputs, whereas procedural justice refers to the fairness of procedures that are used to distribute outcomes (Folger & Greenberg, 1985). Also, two aspects of procedural justice have been identified, formal procedures and interactional justice (Greenberg, 1990; Tyler & Bies, 1990). Formal procedures justice refers to the degree to which fair procedures are used in the organization, whereas interactional justice refers to the manner in which organizational policies and procedures are enacted by supervisors (Moorman, 1991; Tyler & Bies, 1990). This system of classification is significant to mainstream industrial relations analysis of absenteeism and other topics because it offers greater potential to operationalize voice mechanisms across establishments in ways that are more theoretically meaningful than are traditional measures of union or collective agreement status, or grievance activity. For example, procedural justice perceptions of employees can be used to assess both union and nonunion employee perceptions of the fairness of grievance procedures and the manner in which such procedures are administered by supervisors without having to assume, as does traditional exitvoice analysis, that unionized employees have access to fairer procedures by virtue of the practice of collective bargaining. Similarly, employee's distributive justice perceptions can be used to assess the fairness of negotiated outcomes without having to assume, as does traditional exit-voice analysis, that collectively negotiated outcomes are distributively more fair.

CONCLUSIONS Industrial relations research that has applied traditional exit-voice analysis to study the effects of collective voice on employee absenteeism has not found support for the hypothesis that collective voice institutions reduce the use of absenteeism as a form of temporary withdrawal behavior. This article argues that this lack of empirical support has much to do with a number of important theoretical and empirical limitations in the conception and operationalization of the framework itself. These shortcomings in the literature can be overcome by adopting a broader framework that better accounts for the range of collective and individual employee behavioral responses that may emerge in response to workplace problems. In particular, this framework contemplates the inclusion of neglect and loyalty as responses to workplace problems that are identifiably separate from traditional exit and voice alternatives. The broader framework for traditional exit-voice analysis in industrial relations that is recommended in this article is important because it creates a more solid conceptual foundation for studying the role and functions of collective voice in the workplace generally, and as a mode of dispute resolution in particular. Specifically, this broader theory increases the explanatory power of traditional exit-voice analysis by making explicit a route through which the performance of collective voice institutions

100

Luchak and Gellatly

can be evaluated. The benefits of this broader framework are evident in a number of respects. First, the fact that collective voice institutions can act in a constructive or destructive manner and engage in or elicit active or passive employee responses forces future industrial relations research to measure the effectiveness of collective voice and other response mechanisms rather than assuming collective to be more effective than individual voice. Second, the broader framework presented draws attention to a wide range of behavioral responses that employees can engage in response to workplace problems. Among other insights, the activeness-passiveness, and constructiveness-destructiveness dimensions create continuums with which to order employee responses to employment disputes. These continuums can enrich operational tests of the effectiveness of collective voice institutions as methods of dispute resolution. Third, the broader framework recommended in this article shows promise for shedding light on a wide range of theoretical and empirical debates in industrial relations. For example, this broader framework helps to understand the paradox of why unionized workers tend to have both lower job satisfaction and lower turnover propensities (Freeman & Medoff, 1984). Rather than simply assuming, as does the exit-voice framework, that dissatisfied employees will eventually exit the firm, the broader framework discussed here offers the alternative explanation that such employees may remain with the firm indefinitely because they face both high exit and voice costs. This alternative explanation is consistent with the empirical evidence reviewed earlier that unionized workers face both high exit and voice costs. Also, this broader framework can be usefully applied to the union productivity debate by suggesting constructive and destructive paths through which collective voice effects can influence organizational performance. On the one hand, collective voice can operate constructively where, for example, it is used to promote full and informed discussion between the parties, where there is a commitment (viz., loyalty) to resolve employment disputes through established collective voice institutions, and where exit and neglect costs are high. On the other hand, collective voice can operate destructively where there is little commitment to the union-management relationship generally, and where employees have no effective alternatives available to resolve disputes other than by reducing their own inputs through neglectful behaviors. Again, this range of behaviors cannot easily be explained by the traditional exit-voice dichotomy because this dichotomy does not provide an explicit behavioral route for evaluating the effectiveness of voice in the union-management relationship. There are a number of unanswered questions in the literature that should be addressed by future studies examining the effects of collective voice as a method of dispute resolution generally, and on absenteeism as a form of neglect in particular. First, future research must operationalize the neglect concept using actual measures of absenteeism behavior. This will provide an opportunity for assessing the ultimate scope of the effects of collective voice on absence as withdrawal behavior. This is especially important given the debate in the absenteeism literature on the relative importance of work and nonwork factors as determinants of absenteeism. Second, and related, future research on collective voice as a form of dispute resolution needs to control for influences that might lead employees to choose behavioral options other than exit or voice, as well as neglectful options other than absenteeism. Finally, future research must find more

Exit-Voice and Employee Absenteeism

101

theoretically meaningful ways to operationalize voice mechanisms between organized and unorganized workplaces. Measures of distributive and procedural justice offer the greatest promise in this regard, especially given the trend in both organized and/or unorganized workplaces toward varying forms of grievance procedures, risk-sharing programs such as profit-sharing and stock ownership, and employee involvement programs such as autonomous work teams. Each of these programs has elements of voice that traditional collective bargaining status measures cannot validly tap. Evaluating the components of these programs through distributive and procedural justice scales will facilitate the study of voice effects between organized and unorganized workplaces.

ACKNOWLEDGMENTS

Portions of this research were presented at the 47th Annual Meeting of the Industrial Relations Research Association, Washington, DC, January 6-8, 1995. This research was supported by grant #3-34581 under the Associates Program, Faculty of Business Administration, Memorial University of Newfoundland. The authors thank Glenda Careen for capable research assistance and an anonymous referee for helpful comments made on an earlier draft of this paper.

REFERENCES Allen, R. E., & Keaveny, T. J. (1985). Factors differentiating grievants and nongrievants. Human Relations, 38, 519-534. Allen, S. G. (1981a). An empirical model of work attendance. Review of Economics and Statistics, 63, 77-87. Allen, S. G. (1981b). Compensation, safety, and absenteeism: Evidence from the paper industry. Industrial and Labor Relations Review, 34, 207-218. Allen, S. G. (1984). Trade unions, absenteeism, and exit voice. Industrial and Labor Relations Review, 37, 331-345. Brown, D. J., & Beatty, D. M. (1992). Canadian Labour Arbitration, 3rd ed. Aurora, Ontario: Canada Law Book. Chadwick-Jones, J. K., Brown, C. A., Nicholson, N., & Sheppard, C. (1971). Absence measures: Their reliability and stability in an industrial setting. Personnel Psychology 24, 463-470. Chadwick-Jones, J. K., Nicholson, N., & Brown, C. A. (1982). Social Psychology of Absenteeism. New York: Praeger. Farrell, D. (1983). -Exit,voice, loyalty, and neglect as responses to job dissatisfaction: A multidimensional scaling study. Academy of Management Journal, 26, 596-607. Farrell, D., & Stamm, C. L. (1988). Meta-analysis of the correlates of employee absence. Human Relations, 41, 211-227. Farrell, D., & Rusbult, C. E. (1992). Exploring the exit, voice, loyalty, and neglect typology: The influence of job satisfaction, quality of alternatives, and investment size. Employee Responsibilities and Rights Jounzal, 5, 201-218. Folger, R., & Greenberg, J. (1985). Procedural justice: An interpretive analysis of personnel systems. In Rowland, K. M., and Ferris, G. R. (Eds.), Research in Personnel and Human Resource Management, Vol. 3, 141-183. Greenwich, CT; JAI Press. Freeman, R. B., & Medoff, J. L. (1984). What Do Unions Do? New York: Basic Books. Greenberg, J. (1990). Organizational justice: Yesterday, today, and tomorrow. Journal of Management, 16, 399-432. Haekett, R. D., & Guion, R. M. (1985). A reevaluation of the absenteeism-job satisfaction relationship. Organizational Behavior and Hunzan Decision Processes, 35, 340-381.

102

Luchak and Gellatly

Hammer, T. H., & Landau, J. C. (1981). Methodological issues in the use of absence data. Journal of Applied Psychology, 66, 574-581. Hirsehman, A. O. (1970). Exit, Voice, and Loyalty. Cambridge, MA: Harvard University Press. Ichniowski, C. (1986). The effects of grievance activity on productivity. Industrial and Labor Relations Review, 40, 75-90. Johns, G., & Nicholson, N. (1982). The meanings of absence: New strategies for theory and research. Research in Organizational Behavior, 4, 127-172. Katz, H. C., Kochan, T. A., & Gobeille, K. IL (1983). Industrial relations performance, economic performance, and QWL programs: An interplant analysis. Industrial and Labor Relations Re~L,'w,37, 3-17. Katz, H. C., Kochan, T. A., & Weber, M. (1985). Assessing the effects of industrial relations systems and efforts to improve the quality of working life on organizational effectiveness. Academy of Management Journal, 28, 509-526. Klaas, B. S., & De Nisi, A. (1989). Managerial reaction to employee dissent: The impact of grievance activity on performance ratings. Academy of Management Jouma~ 32, 705-717. Klaas, B. S., Heneman, H. G., & Olson, C. A. (1991). Effects of grievance activity on absenteeism. Journal of Applied Psychology, 76, 818-824. Leigh, J. P. (1981). The effects of union membership on absence from work due to illness. Journal of Labor Research, 2, 329-336. Leigh, J. P. (1984). Unionization and absenteeism. Applied Economics, 16, 147-157. Leigh, J. P. (1986). Correlates of absence from work due to illness. Human Relations, 39, 81-100. Lewin, D. (1984). Empirical measures of grievance procedure effectiveness. Proceedingsof the 1984 Spring Meeting, Industrial Relations Research Association, Madison, WI: 491-499. Lewin, D. (1987). Dispute resolution in the nonunion firm: A theoretical and empirical analysis. Journal of Conflict Resolution, 31, 465-502. Lewin, D., & Peterson, R. B. (1988). The Modem Grievance Procedure in the United States. Westport, CT: Quorum Books. Marcus, P. M., & Smith, C. B. (1985). Absenteeism in an organizational context. Work and Occupations, 12, 251-268. Markham, S., & Scott, K. D. (1985). Controlling absenteeism: Union and nonunion differences. Personnel Administrator, 30, 87-102. Moorman, R. H. (1991). Relationship between organizational justice and organizational citizenship behaviors: Do fairness perceptions influence employee citizenship? Journal of Applied Psychology, 76, 845-855. Muchinsky, P. M. (1977). Employee absenteeism: A review of the literature. Journal of Vocational Behavior, 10, 316-340. Near, J. P., & Miceli, M. P. (1986). Retaliation against whistle blowers: Predictors and effects. Journal of Applied Psychology, 71, 137-145. Parmerlee, M. A., Near, J. P., & Jensen, T. C. (1982). Correlates of whistle-blowers' perceptions of organizational retaliation. Administrative Science Quarterly, 27, 17-34. Peterson, R. B., & Lewin, D. (1982). A model for research and analysis of the grievance process. Proceedings of the 34th Annual bfeeting~ Industrial Relations ResearchAssociation, Madison, WI, 303-312. Porter, L. W., & Steers, R. M. (1973). Organizational, work, and personal factors in employee turnover and absenteeism. Psychological Bulletin, 80, 151-176. Rhodes, S. R., & Steers, R. M. (1990). Managing Employee Absenteeism. Reading, MA: Addison-Wesley. Rusbult, C. E., Farrell, D., Manious III, A. G., & Rogers, G. (1988). Impact of exchange variables on exit, voice, loyalty, and neglect: An integrative model of responses to declining job satisfaction. Academy of Management Jouma~ 31, 599-627. Rusbult, C. E., & Lower)', D. (1985). When bureaucrats get the blues: Responses to dissatisfaction among federal employees. Journal of Applied Social Psychology, 15, 80-103. Saunders, D. M., Sheppard, B. H., Knight, V., & Roth, J. (1992). Employee voice to supervisors. Empho'ee Responsibilities and Rights Jouma~ 5, 241-259. Steers, R. M., & Rhodes, S. R. (1978). Major influences on employee attendance: A process model. Journal of Applied Psycholog),, 63, 391-407. Thibaut, J., & Walker, L (1975). ProceduralJustice: A PsychologicalAnab'sis. Hillsdate, NJ: Erlbaum. Tyler, T. R., & Bies, R. J. (1990). Beyond formal procedures: The interpersonal context of procedural justice. In Carroll, J. S. (Ed.), Applied Social Psychology in Business Settings, 77-98. Hillsdale, NJ: Erlbaum. Withe),, M. J., & Cooper, W. H. (1989). Predicting exit, voice, loyalty and neglect. Administrative Science Quarterly, 34, 521-539. Wilson, N., & Peel, M. J. (1991). The impact on absenteeism and quits of profit-sharing and other forms of employee participation. Industrial and Labor Relations Review, 44, 454-468.