Susan M. Stewart, University of Puget Sound, sstewart@ups.edu
Mark N. Bing, University of Mississippi, mbing@bus.olemiss.edu
Melissa L. Gruys, Wright State University, melissa.gruys@wright.edu
Michael C. Helford, Roosevelt University, mhelford@aol.com

Abstract

As the participation rate of women in the workforce has increased, there has been an emphasis in organizational research on gender issues. One overlooked area pertains to how men and women perceive their work environments and how different climate dimensions affect dedication to organizations. This study utilized 553 (285 women, 268 men) employees to investigate gender differences in (1) affective and continuance commitment, (2) turnover intentions, (3) psychological climate perceptions (i.e., autonomy, cohesion, trust, pressure, support, recognition, fairness, innovation), as well as (4) the moderating influence of gender on the relationship between psychological climate and affective commitment, continuance commitment, and turnover intentions. We controlled for critical covariates of gender, including education, age, job tenure, job level, and organizational membership. Results show that female employees had higher levels of continuance commitment than men. There were no gender differences found for the climate dimensions when examining direct relationships. However, the task-oriented climate dimension of organizational support was a significant predictor of affective commitment and turnover intentions for men, whereas the relationship-oriented climate dimension of workplace recognition was a significant predictor of affective commitment and turnover intentions for women. Limitations, future research ideas, and the practical implications of these findings are provided.

Introduction

It is well known that there has been an increase in the presence of women in the workplace over the last few decades and this growth pattern is projected to continue.  In the United States, women’s participation in the labor force increased from 40% in 1970 to 56% in 2002 while men’s participation rates decreased from 76% in 1970 to 69% in 2002 (Bureau of Labor Statistics, 2004).  The percentage of women holding positions in managerial and professional specialty areas has risen from 22% in 1983 to 34% in 2002, while the percentage of men working in these areas rose only slightly from 25% in 1983 to 29% in 2002 (Bureau of Labor Statistics, 2004).  It is estimated that by the year 2008, females will represent 48% of the labor force (Bureau of Labor Statistics, 2004). With this increase in participation rates of women in the workforce, there has been an emphasis in organizational research on examining whether gender differences exist in the workplace attitudes and behaviors of men and women (Aven, Parker, & McEvoy, 1993; Wahn, 1998).

The range of workplace variables in which gender differences have been examined is quite broad and include job satisfaction (deVaus & McAllister, 1991; Mason, 1995), job burnout (Maslach & Jackson, 1985; Pretty, McCarthy, & Catano, 1992); leadership style (Eagley & Johnson, 1990; Hutchison, Valentino, & Kirkner, 1998), ethical behavior (Barnett & Karson, 1989; Kelley, Ferrell, & Skinner, 1990), and political tactics (Schilit & Locke, 1982; Tannen, 1995). Furthermore, important outcome variables measured in terms of absenteeism (Hackett, 1989, 1990; Martocchio, 1989), turnover (Carston & Spector, 1987; Weisberg & Kirschenbaum, 1993), organizational commitment (Aven, et al., 1993; Mathieu & Zajac, 1990), and intentions to leave (Miller & Wheeler, 1992; Rosin & Korabik, 1995; Stroh, Brett, & Reilly, 1996; Weisberg & Kirschenbaum, 1993) has been a focus of gender research. The concentration on these workplace attitudes and behaviors is not surprising in light of the costs of absenteeism and turnover to organizations (Cascio, 1991).

Although gender does not consistently have a direct impact on outcome variables such as organizational commitment (Aven, et al., 1993; Mathieu & Zajac, 1990; Porter & Steers, 1973), the factors that influence one to become a committed employee may vary across gender. It has been suggested that studies that focus solely on individual subjective reactions are, at best, incomplete if they do not include an examination of the contextual factors that shape those perceptions (Carless, 2004; Ostroff, 1993). An overlooked area in gender research pertains to how men and women perceive their work environments and how different dimensions of psychological climate affect dedication to organizations.  For example, some studies have shown that females often view themselves as treated worse than males in the workplace (Graddick & Farr, 1983; Stokes, Riger, & Sullivan, 1995).  Because females often attach more importance to certain work conditions than males (Manhardt, 1972; Reitz & Jewell, 1979), dimensions of psychological climate related to those conditions may have an especially strong impact on female commitment (Chusmir, 1988). Conversely, other dimensions of psychological climate may be more related to the organizational commitment of males.

We believe that making a connection between gender and psychological climate will offer insights unlike those provided in the literature to date, specifically with regards to how psychological appraisals of environmental workplace factors may have a gender specific impact on organizational commitment and turnover intentions. Hence, the current study will begin by investigating gender differences in (1) affective and continuance commitment, (2) turnover intentions, (3) psychological climate perceptions (including autonomy, cohesion, trust, pressure, support, recognition, fairness, and innovation), and will then examine (4) the moderating influence of gender on the relationship between psychological climate and affective commitment, continuance commitment, and turnover intentions. Two criticisms of prior research on gender-related differences has been the lack of statistical control for the effects of demographic variables (Lefkowitz, 1994) as well as few comparisons made across male and female workers in essentially similar jobs (Mannheim, 1983). We agree with these noted limitations and therefore controlled for critical covariates of gender, including education, age, job tenure, job level, and organizational membership when testing all of the research questions posed in the current study. Overall, we believe that the identification and maintenance of factors that promote organizational commitment and decrease turnover intentions may be essential in promoting short- and long-term organizational success (Dodd-McCue & Wright, 1996), and therefore hope that this review of the literature and empirical investigation will be of interest to academicians and practitioners alike.  Furthermore, men and women should know more about what is happening to them in organizations, so that they can better understand the terms of their differences (Fine, Johnson, & Ryan, 1990).

Gender and Organizational Commitment

The concept of organizational commitment has attracted considerable attention as an attempt to understand the intensity and stability of employee dedication to work organizations. Two quite different definitions of commitment have been popular in the empirical literature; one provided by Porter and his associates (Mowday, Porter, & Steers, 1982; Porter, Steers, Mowday, & Boulian, 1974), and the other by Becker (1960). According to Porter et al. (1974), commitment is the “strength of an individual’s identification with and involvement in a particular organization” (p. 604). Becker, on the other hand, described commitment as the tendency to engage in “consistent lines of activity” (p. 33) because of the perceived cost of doing otherwise (i.e., leaving). Meyer and Allen (1984; Allen & Meyer, 1990) used the terms affective and continuance commitment, respectively, to characterize Porter and Becker’s discrepant views of the construct. Employees with a strong affective commitment remain with the organization because they want to; these individuals identify with the organization and, therefore, are committed to maintaining membership in order to pursue organizational goals. Those individuals with strong continuance commitment remain because they need to do so; they are bound to the organization through extraneous interests such as pensions, benefits, seniority, and the cost of leaving, rather than through a favorable affective connection with the organization.

Despite the substantial number of studies that have investigated the antecedents of organizational commitment, the literature on the relationship between gender and organizational commitment has had mixed results.  For example, there are some authors who suggest that women are less committed to their work than men (Karrasch, 2003; Schwartz, 1989; Yammarino & Dubinsky, 1988). Much of these contentions have as their roots the idea that women, as a result of their socialization, place a greater emphasis on family roles than men (Dodd-McCue & Wright, 1996; Jensen, Christensen, & Wilson, 1985; Kinnier, Katz, & Berry, 1991; Loscocco, 1990; Steffy & Jones, 1988), which in turn may result in women placing less importance on their work roles. This assertion also posits that women establish their identity through their interdependent, nurturing relations with others, whereas men’s socialization process leads them to identify themselves as independent, assertive, and goal-directed (Cook, 1993). Supportive of this assertion is the evidence that in the accounting profession (Aranya, Kushnir, & Valency, 1986), and in professional associations (Graddick & Farr, 1983), women are less affectively committed than men.

However, researchers who appear to be focused on the continuance component of commitment have often argued that women are more committed to organizations than men (Grusky, 1966; Hrebiniak & Alutto, 1972), because they must overcome more obstacles in order to gain employment (Grusky, 1966) and have less interorganizational mobility than males (Angle & Perry, 1981). This perspective is complemented by studies showing that workers who perceive limited employment options (Angle & Perry, 1981; McGee & Ford, 1987; O’Reilly & Caldwell, 1981) and higher costs associated with establishing their organizational membership (Grusky, 1966) display greater organizational commitment, perhaps specifically continuance commitment (Aven et al., 1993). An example of empirical support for this theoretical perspective regarding gender differences in continuance commitment is a study by Wahn (1998), which found women to be higher in continuance commitment than men (although the difference between the two groups was somewhat modest).

Several meta-analyses on organizational commitment have helped to elucidate the aforementioned theoretical and empirical controversy. Mathieu and Zajac’s (1990) meta-analysis of the antecedents, correlates, and consequences of organizational commitment revealed that women are more organizationally committed than men, although the difference was small. Additionally, they did not find a difference in the strength of the gender-commitment relationship across commitment type (i.e., affective and continuance commitment). Aven et al. (1993) concluded from a separate meta-analysis that there were no gender differences in affective commitment. They also stated that they were unable to address the effect of gender on continuance commitment because the published research focused almost exclusively on affective commitment. However, their meta-analysis included six studies that used the Hrebiniak and Alutto (1972) commitment instrument, which purportedly measures an employee's calculative (i.e., continuance) involvement with an organization.

More recent research (over twenty studies; e.g., Meyer, Stanley, Herscovitch, & Topolnytsky, 2002; Riketta, 2005; Thorsteinson, 2003) found that there were no gender differences in organizational commitment.  Seven additional studies found that even when there was a mean difference in organizational commitment between men and women, there was no gender effect when predicting organizational commitment (i.e., via multiple regression) when control variables such as age, job level, educational, job and organizational tenure were included in the analyses (e.g., Abdulla & Shaw, 1999; VanderVelde, Bossink, & Jansen, 2003; Ngo & Tsang, 1998).  This suggests that certain characteristics that might be correlated with gender (e.g., job level as women are more likely to have lower level jobs) may explain the difference in organizational commitment more so than gender itself.

Hence, since the literature includes mixed findings and does not always carefully separate findings for different types of organizational commitment (affective or continuance), we propose the following:

Research Question 1: Will women and men differ in terms of affective and continuance

commitment to their organizations?

Gender and Turnover Intentions

Intentions to leave the organization, or turnover intentions, have not received as much attention as organizational commitment, yet they remain an important component of employee dedication given the powerful and proximal impact that intentions have on workplace behaviors (Fishbein & Ajzen, 1975). Not only have turnover intentions been linked with actual turnover (Shore & Martin, 1989), but some researchers have recommended its use in place of actual turnover because of the numerous external factors that determine turnover behavior (Bluedorn, 1982; Price & Mueller, 1981). Additionally, when measured appropriately, turnover intentions can be placed on a quantitative, or near continuous scale, whereas actual turnover is dichotomous and subject to a base rate deflation in its relationship with other variables when the proportions in the dichotomy are values other than .50 and .50 (Cohen & Cohen, 1983, pp. 66-67).

A review of the research related to gender differences in turnover intentions and turnover rates showed mixed results. Some studies suggest that women report higher levels of turnover intentions (e.g., Miller & Wheeler, 1992; Moncrief, Babakus, Cravens, & Johnson, 2000; Schul & Wren, 1992) and actual turnover (Mano-Negrin, 2003; Stroh, et al., 1996), than men. Research has suggested these findings may be due to job dissatisfaction on the part of women (which could be related to the lower level positions commonly held by women) or because women have lower earnings and fewer opportunities for advancement (e.g., Blau & Kahn, 1981). Some may speculate that these findings are related to statistical discrimination theory which suggests that employers’ experiences may cause them to expect women to leave an organization sooner than men (see Konrad & Cannings, 1997).  Others have found that women experience a greater number of “shocks” or events that force them to decide whether to leave the organization (e.g., a pregnancy and birth of a child) (Donnelly & Quirin, 2006).  However, a study of textile workers found that while women had a higher turnover rate than men, no gender differences were found in intentions to leave the organization (Weisberg & Kirschenbaum, 1993). Still other studies of managerial and professional positions have found no gender difference in turnover intentions (Rosin & Korabik, 1995; Xu, Veloski, Hojat, & Fields, 1995).  Some studies even found that men had higher turnover intention (Smith & Calasanti, 2005).  Clearly, the results for the relationship between gender and intentions to turnover are discrepant, and we plan to reexamine this relationship by including turnover intentions in the current study as well as demographic controls.

Research Question 2: Will women and men differ in their intentions to leave their organization?

Gender and Psychological Climate

Psychological climate can be defined as sets of perceptually-based descriptions of relevant organizational features, events, and processes (James & Jones, 1974; Jones & James, 1979). These perceptions represent cognitive interpretations of the organizational context or situation, and summarize an individual’s description of their work experiences (Schneider, 1975). The appraisal is a reflection of the organizational characteristics that are important to the individual and his or her personal and organizational well-being (James, James, & Ashe, 1990). Although description cannot be completely divorced from affective evaluation (Ashford, 1985), the distinction between descriptive and evaluative reactions to organizational experiences distinguishes climate from job satisfaction (Glick, 1985; Koys & DeCotiis, 1991).  The construct of psychological climate is useful in organizational research because it aids in the prediction of work outcomes such as organizational commitment and job satisfaction, psychological well-being, motivation, and performance (Carr, Schmidt, Ford, & DeShon, 2003; Parker, Baltes, Young, Huff, Altmann, Lacost, & Roberts, 2003).

The study of climates in organizations has been difficult because it is a complex phenomenon (Glick, 1985; Joyce & Slocum, 1984; James, 1982; Johannesson, 1973). While there is a general consensus on the concept of psychological climate, there appears to be little agreement on its dimensionality, and thus its measurement (Parker et al., 2003). Koys and DeCotiis (1991) reviewed the literature in this area and assembled a list of over 80 separately labeled dimensions of climate. They established a set of criteria by which a reduction of these dimensions could take place. After the reduction process, forty-five of the original dimensions were retained and categorized into eight concepts viewed as the “universe of psychological climate” (p. 268). This categorization of climate perceptions is of interest in the current study because these dimensions are conceptually distinct, provide a means for the theoretically-meaningful and analytically-practical classification of employee perceptions of the work environment, and resemble similar types of dimensions examined in previous gender research.

These eight psychological climate dimensions include (1) autonomy: perception of self-determination with respect to work procedures, goals, and priorities, (2) cohesion: perception of togetherness or sharing within the organization setting, including the willingness of members to provide material aid, (3) trust: perception of freedom to communicate openly with members at higher organizational levels about sensitive or personal issues with the expectation that the integrity of such communications will not be violated, (4) pressure: perception of time demands with respect to task completion and performance standards, (5) support: perception of the tolerance of member behavior by superiors, including the willingness to let members learn from their mistakes without fear of reprisal, (6) recognition: perception that member contributions to the organization are acknowledged, (7) fairness: perception that organizational practices are equitable and nonarbitrary or capricious, and (8) innovation: perception that change and creativity are encouraged, including risk-taking into new areas or areas where the member has little or no prior experience.

Research that specifically addresses the issue of gender differences in psychological climate perceptions is sparse even though there appears to be a growing concern with the experiential aspects of organizational life (Mills, 1988). Studies that address climate-related variables in organizational research indicate that women’s work experiences are different than men’s work experiences (Gewertz, 1994; Repetti, Matthews, & Waldron, 1989; Rodriguez, 1993), and that men and women employees perceive and react differently to organizational components of the work environment (Brookes & Kaplan, 1972; Campbell, 1979; Kirschenbaum, 1991; Mehrabian & Russell, 1974; Nicholson & West, 1988; Peterson, Wekerle, & Morley, 1978; Sundstrom, Town, Brown, Forman, & McGee, 1982).

One study along this line of research by Pines, Aronson, and Kafry (1980) found that professional women perceive less autonomy, less freedom, less influence, less variety in their work assignments, fewer challenges, and a less positive work environment compared to professional men. Additional research has found that men assign greater importance to rewards such as self-direction or autonomy, pay, security, and promotions, while women assign greater value to social rewards such as interesting work, good relations with coworkers, and a friendly work atmosphere (Bartol, 1976a, 1976b; Chernick & Phelan, 1974; Hofstede, 1980; Manhardt, 1972; Schuler, 1975). These findings suggest that men perceive themselves to have more control over the work environment in comparison to women, which could be explained by the extant earnings gap between men and women (Gerhart, 1990), as well as by the importance assigned by men to salary (Bigoness, 1988), autonomy, and self-direction (Hochwarter, Perrewe, & Dawkins, 1995). Bigoness (1988) also found that female MBA candidates place less emphasis on salary in comparison to their male counterparts, but did not find gender differences with respect to the emphasis placed on characteristics of the work environment, such as working facilities and training programs. The above research indicates that not only are certain components of the work environment perceived differently by men and women, but it also suggests that specific components of the work environment are differentially salient to men and women. We intend to add to this body of research by examining gender differences in perceptions of the work environment, after having controlled for important demographics, with the use of Koys and DeCotiis' (1991) eight-part taxonomy of psychological climate.

Research Question 3: Will women and men differ in their perceptions of the work environment, namely the psychological climate dimensions of autonomy, cohesion, trust, pressure, support, recognition, fairness, and innovation?

Gender as a Moderator of the Relationship Between Psychological Climate and Employee Dedication

Despite the large body of evidence elucidating the role that gender plays in various organizational outcomes (Barnett & Karson, 1989; deVaus & McAllister, 1991; Kelley et al., 1990; Mason, 1995; Rosin & Korabik, 1995; Schilit & Locke, 1982; Tannen, 1995), there is still a need to investigate its role as a potential moderator of the relationship between perceptions of the work environment and organizational outcomes. As noted above, various aspects of the work environment may be differentially salient to men and women. Therefore, the perceptions of the work environment, or dimensions of psychological climate, that influence an employee to increase or decrease his or her voluntary attachment to the organization may change depending upon employee gender. Voluntary attitudes and actions are not directly controlled by the organization in comparison to overt job behaviors (Drory, 1993). Such volitional attitudes are therefore expected to change easily in response to satisfaction or disappointment with the organization. An employee's dedication to the organization, in the forms of organizational commitment and turnover intentions, are perhaps the constructs most likely to reflect this effect (Drory, 1993, Mathieu & Zajac, 1990).

An extensive literature search revealed only three studies that examined gender as a potential moderator of the relationship between climate-related and dedication-related variables. A study by Pittman and Orthner (1988) found, for both genders, that the more positively an individual views his or her sense of fit with the organization, the higher the level of job commitment, and the more likely it is that the job will be viewed as a positive contributor to a high quality of life. When career decisions have to be made, “a commitment to the organization could be seen as an exchange for the quality of life the job is believed to make possible” (p. 242). In this study, path analyses were performed separately for males and females because the relations among the study variables were expected to vary enough between the genders to justify separate models (Pittman & Orthner, 1988). The structure of the two models was not identical, which suggests gender could have interacted with the antecedents of job commitment. Therefore, Pittman and Orthner's (1988) separate path analyses serve to indirectly implicate gender as a moderator of the antecedents of job commitment. The antecedents included in their models were perceptions of organizational supportiveness of families as well as background variables, such as job tenure and marital satisfaction. Although the power of several of these antecedent variables to predict job commitment changed across gender, the focus of this study was on work-to-family interface, and it did not examine the psychological climate variables of the current study, which may differentially predict job commitment for men and women once important demographics are controlled for. Furthermore, their study does not directly test for the moderation of the relationship between job commitment and its precursors by gender. A direct test is needed.

In a second study, Weisberg and Kirschenbaum (1993) proposed that men and women perceive, evaluate, and react to their work environment differently, and that these reactions affect their turnover intentions and actual turnover. They tested this hypothesis by examining gender, turnover intentions, and actual turnover of Israeli textile industry workers. They found that gender was not significant in explaining intentions to leave the organization, while it did prove significant in explaining actual turnover. Of central importance to the current study was their finding that perceptions of the social aspects of the work environment predicted actual turnover for females, whereas they were unrelated to actual turnover for males. In contrast, perceptions of task or job-oriented aspects of the work environment, such as perceptions of the repetitiveness of work and promotion opportunity, predicted turnover intentions for males but not for females. The current study intends to replicate and extend their findings by examining turnover intentions as well as the affective and continuance components of organizational commitment.

In the third study, Witt (1989) examined gender differences in the relationships between organizational commitment and two of its antecedents, job satisfaction and psychological climate. Witt (1989) found several of the relationships between psychological climate and organizational commitment to change across gender. For instance, the relationship between organizational support and Hrebiniak and Alutto's (1972) measurement of commitment was stronger for men than for women, with the relationship being positive in both cases. Although other gender differences were found in climate-to-commitment relationships, Witt's (1989) results must be interpreted with caution for several reasons. First, when controlling for the effects of age, job tenure, and other important demographic variables, Witt (1989) included only 20 males and 71 females in his sample. This small sample size, especially for men, is of concern given the spurious effect sizes that can occur due to the use of small sample sizes (Cohen & Cohen, 1983; Keppel, 1991). Additionally, to test for the moderating effect of gender when controlling for demographics, Witt (1989) performed 32 significance tests for independent partial correlations. Even when using Cohen and Cohens' (1983) recommended alpha value of .10, one finds no significant differences between the genders in climate-to-commitment relationships when controlling for pyramiding alpha level (i.e., Bonferroni correction with .10 ¸ 32 = .003). Witt (1989) made note of this limitation in his study. Nonetheless, given the aforementioned limitations, we believe the potential moderation of climate-to-commitment relationships by gender requires further examination.

Research Question 4: Will gender moderate the relationship between psychological climate and employee dedication?

Method

Procedure and Participants

Prepared questionnaire packets were delivered in bulk to a contact person at each of three office supply organizations located in the Midwestern United States. The contact person in each organization was either the owner or reported directly to the owner or senior manager. These persons distributed the questionnaire packets to organizational members and emphasized that anonymity was guaranteed. Each questionnaire packet consisted of an outer envelope, a cover letter from the investigators, a questionnaire, and a preaddressed and stamped return envelope to return the questionnaire by mail after completion.

A total of 774 questionnaires were returned, which yielded an overall response rate of 24.9%. This response rate is in line with other published research that utilized the mail survey method of data collection (Dillman, 2000). After missing data were taken into account, 553 cases remained in the data set. Therefore, participants were 553 (285 women, 268 men) employees holding a variety of job positions within the office supply organizations. The average female respondent was between 30 and 39 years of age, had completed some college, had worked with her current employer for 4.57 years, and supervised 1.94 people. The average male respondent was between the ages of 30 and 39, with some college education, 5.74 years of tenure with his current employer, and supervised 10.81 people.

Measures

Psychological Climate. Individual perceptions of psychological climate were assessed with a thirty-five item scale based on the eight global dimensions of Koys and DeCotiis' (1991) climate categories: cohesion (a = .86), trust (a = .74), pressure (a = .61), support (a = .82), autonomy (a = .66), recognition (a = .77), fairness (a = .80), and innovation (a = .69). Past research has confirmed the psychometric properties of this measure (Helford, 1995; Helford, Tindale, Dugoni, & Posavac, 1996). The psychological climate scale used a 5-point Likert response format (strongly agree to strongly disagree) to measure an individual's perceptions of the work environment at his or her current organization.

Affective Commitment. This fifteen-item measure was developed to assess commitment characterized by positive feelings of identification with, attachment to, and involvement in, the work organization (Helford, 1995; Helford et al., 1996). Responses were provided using a 5-point Likert response format (strongly agree to strongly disagree). High scores reflect greater affective commitment (a = .93).

Continuance Commitment. This seven-item measure was developed to assess the extent to which employees feel committed to their organizations by virtue of the costs that they believe are associated with leaving the organization, such as investments and/or lack of attractive alternatives (Helford, 1995; Helford et al., 1996). Responses were provided using a 5-point Likert format (strongly agree to strongly disagree). High scores reflect greater continuance commitment (a = .79).

Turnover Intentions. This six-item measure was developed to determine the extent to which a respondent planned to remain with his or her present organization given various time frames (“3 months from now” to “until retirement”). Items were rated on a 5-point Likert format (strongly agree to strongly disagree), and responses to all items were summed such that high scores reflect stronger turnover intentions (a = .90).

Demographics. Demographic information was also collected which included the respondent's gender, education, age, job tenure, job level (i.e., number of persons supervised), as well as the organizational membership of the respondents (these variables were represented by two unique effect codes; total number of organizations -1, or 3 - 1; see Cohen & Cohen, 1983, pp. 198-204). The number of people supervised provided a measure of job level because organizations often use it to assess the location of job positions within the organizational hierarchy, and to determine the importance of those positions (Lyness & Thompson, 1997; Reskin & Ross, 1995).

Results

The items for the measures are shown in the Appendix. The means, standard deviations, and intercorrelations for all of the study variables are presented below in Table 1.

It has been suggested that gender differences are an artifact of exposure to different organizational situations (Kanter, 1977), or the result of methodological flaws where the confounding effects of factors like education, age, job tenure, job level, and occupational setting were not held constant (Aranya, et al., 1986; Lefkowitz, 1994; Schuler, 1975). We agree with this noted limitation of research on gender differences. Consequently, all of the major analyses controlled for these variables. Specifically, the variables of education, age, job tenure, job level, and organizational membership were partialled from the dependent variables of interest (i.e., specified as covariates) prior to examining the effects of gender and other study variables.

Test of Research Questions 1 and 2

A MANCOVA was used to determine whether men and women differed significantly in affective commitment, continuance commitment, and turnover intentions when these outcomes were considered in tandem and allowed to covary. Demographic variables were controlled for by specifying education, age, job tenure, job level, and organizational membership as covariates in the analysis. The results indicated that the mean centroids for men and women were significantly distal from each other, F(3, 503) = 6.55, p < .001. Univariate ANCOVAs revealed that the significant multivariate analysis was largely a result of differences in continuance commitment, F(1, 505) = 12.10, p = .001 (critical alpha of .0167 with Bonferroni correction), with women having the higher mean level of this type of organizational commitment. Men and women did not differ significantly in their affective commitment or turnover intentions.

Table 1
Means, Standard Deviations, and Intercorrelations for All Study Variables