An Exploration of the Relationship Between Autonomy Congruence,Perceived Supervisor Individualized Consideration,and Employee Outcomes

Abstract

Human resource practices requiring employee participation or involvement in work-related decision-making have been commonly believed to be beneficial to work outcomes; however, we suggest that the effects of those practices on work outcomes can be limited by individual preferences, which influence the perceived quality of supervision. Drawing upon prior research on work structure, person–environment (P-E) fit perspective, and perceived organizational support, we examine the joint effects of autonomy and preference for autonomy on employee work outcomes (task performance and organizational citizenship behaviors) that are mediated via perceived supervision (individualized consideration) among U.S. state government employees. The results showed that the congruence between autonomy and preference for autonomy was associated with the highest level of perceived individualized consideration by supervisors, highlighting the important role played by perceptions of the supervisor behavior in promoting employees’ positive attitudes at work.

Keywords

autonomy supervision person–environment fit

Introduction

Management scholars and researchers have long accepted that autonomy, defined as “the degree to which the job provides substantial freedom, independence, and discretion to the individual in scheduling the work and in determining the procedures to be used in carrying it out” (Hackman & Oldham, 1976, p. 258), has a variety of positive effects on employee attitudes and behavior (e.g., Bandura, 2001; Conger & Kanungo, 1988; Hackman & Oldham, 1980). Self-determination theory, one of the most widely cited theories of human motivation, suggests that autonomy is a fundamental psychological need that facilitates self-motivation and healthy psychological development (Ryan & Deci, 2000). Accordingly, it is commonly assumed that workers prefer more autonomy and less centralization at work. Giving employees more control and autonomy is also presumed to facilitate more positive perceptions of supervisor and organizational supportiveness because employees attribute favorable working conditions in part to the discretion of their managers and employers (Kurtessis et al., 2017).

However, drawing from the management literature on person–environment (P-E) fit (Edwards, 1996; Kristof, 1996), we argue that the benefits of autonomy for public sector employees’ attitudes toward their supervisors may not be universal. In particular, the employees’ preference or need for an autonomous environment may be an important boundary condition for the positive effect of job autonomy on perceptions of supervisor supportiveness. Previous studies on autonomy have examined different contexts in which autonomy operates (e.g., Adler & Borys, 1996; Bozeman & Kingsley, 1998; Langer et al., 2019), and we focus on individual values as contingencies for the influence of autonomy on perceptions of supervisor individualized consideration, with implications for worker behavior. We assume that not all employees equally prefer autonomous or participatory environments and suggest that the ideal level of job autonomy and control is in part based on individual preferences. Our argument for the importance of considering both individual differences in preference and the work environment together is guided by P-E fit theory (Edwards, 1996; Edwards & Parry, 1993),), which examines how job attitudes and other individual outcomes are explained by the fit between individuals and their work situations.

In line with this perspective, we suggest that the fit between job autonomy and a public sector employee’s preference for autonomy influences the extent to which the employee perceives his or her supervisor as considerate of the employee’s individual needs. Perceptions of individual consideration have been shown to be an important factor in employee attitudes and behavior (e.g., Howell & Avolio, 1993; Stinglhamber & Vandenberghe, 2003), rendering it an important outcome of P-E fit. Previous research found that misfit between a job’s characteristics and those that would be expected for one’s occupation predicts reduced perceptions of support from the supervisor (Ford, 2012), suggesting that P-E misfit may indeed be an antecedent of the perceived quality of the supervisor’s leadership. Employees may attribute the fit between the work environment and their preferences to the quality of the supervisor, influencing perceptions about whether they are supervised in a considerate and supportive fashion. Considerate and supportive leadership is reflected in the construct of individualized consideration, one of the core dimensions of transformational leadership.

We integrate the P-E fit perspective and leadership literature in a public sector context and propose that P-E interactions impact key employee outcomes through employee perceptions of supervisor individualized consideration. P-E fit theory suggests that not all employees interpret higher levels of autonomy as indicative of the supervisor’s consideration of their individual needs. Thus, our primary research question is as follows: How does the fit between job autonomy and public employee preference for autonomy predict perceptions of the supervisor’s individualized consideration? To address the question, we draw upon previous research on leadership and propose perceived individualized consideration as a mechanism linking P-E fit and work outcomes in public sector settings.

The purpose of our study is to help improve our understanding of the process through which autonomy influences work outcomes by investigating the interaction between perceived job autonomy and individual preference for autonomy. We consider job autonomy fit as an antecedent to perceptions of leaders because such working conditions are in part attributed to the leader’s favorable intent. In other words, employees attribute autonomy fit to the supportiveness of the supervisor, influencing perceptions of supervisor individualized consideration. We test this research model with survey data from full-time professional employees working in a large state government in the United States by conducting a series of polynomial regression analyses.

To present this study, we first provide some theoretical background and build our research hypotheses (see Figure 1). This is followed by a section describing the methods used to test the study’s hypotheses and the results of the data analyses. Finally, in the article’s last section, we discuss the implications of results and the theoretical and practical contribution of this study while discussing its limitations and offering suggested directions for future research.

Figure 1.

Summary illustration.

Literature Review

P-E Fit

P-E fit refers to the match between individuals and their work environments and has received much attention in organizational behavior (Kristof-Brown et al., 2005). The P-E fit approach suggests that individual attitudes, behaviors, and other outcomes result not from the person or environment separately, but rather from the relationship between the two (Edwards, 1996), emphasizing the importance of compatibility between workers and the work environment (Kristof, 1996) in worker well-being and performance. Understanding the consequences of mismatches between employees and their work environments is valuable for public managers who seek to address human resource issues.

In recent years, public administration scholars have begun delving into P-E fit, focusing particularly on person–organization (P-O) fit. P-O fit is a specific type of P-E fit and refers to the congruence between the individual’s values and those of the organization (Kristof, 1996; Leung & Chaturvedi, 2011). Much of this research has investigated P-O fit as having a key role in the relationship between public service motivation (PSM) and outcome variables such as satisfaction (Kim, 2012; Steijn, 2008; Wright & Pandey, 2008), commitment (Kim, 2012), turnover intention (Steijn, 2008), and performance (Bright, 2007).

While focusing on the fit between the employee and the organization, this work has generally placed less attention on the fit between the employee and the characteristics of the job, which is sometimes referred to as person–job fit. Because individual job characteristics can be in part attributed to the supervisor’s discretion, this may have implications for the perceived quality of public sector supervisors. Still, to this point, the effects of P-E fit on the perceived quality of supervision remain to be examined in the public sector. Arguably, ignoring the role of employee perceptions of supervision in the relationship between P-E fit and work outcomes has hampered our understanding of the P-E interactions that are relevant to many organizational phenomena.

Autonomy and Individualized Consideration

In examining the relationship between autonomy and perceived supervision, we focus on perceptions of individualized consideration, one component of transformational leadership whose conceptual nature relates to our theoretical perspective that P-E fit is positively related to the perceived supportiveness of the supervisor. Individualized consideration is defined as an employee’s perception of the quality of interpersonal treatment from the leader in terms of the extent to which the supervisor focuses on one’s individual needs (Bass & Riggio, 2006). Supervisors who exhibit individualized consideration pay close attention to their followers’ particular needs by providing support, encouragement, care, and nurturing (Bass, 1985; Bass & Riggio, 2006), eventually resulting in the follower’s perception that he or she is being treated with respect and value by the supervisor.

The literature on perceived organizational/supervisory support suggests that autonomy is directly and positively related to perceptions of support from the organization (Eisenberger et al., 1999; Rhoades & Eisenberger, 2002). As a job condition that employees value and that meets employees’ basic psychological needs (Ryan & Deci, 2000), job autonomy has been studied as a predictor of perceived organizational support. When organizations give employees autonomy, they show that they care about the employees’ needs while also displaying trust in workers to decide how to carry out the tasks of the job (Eisenberger et al., 1999; Rhoades & Eisenberger, 2002). In their meta-analysis, Rhoades and Eisenberger (2002) found the correlation between autonomy and perceived organizational support to be .41. More importantly, job autonomy is often provided at the supervisor’s discretion. Supervisors, even those within the same organization, can vary in the extent to which they monitor the methods employees use to perform their work. Supervisors may also vary in the schedule, break, and leave flexibility they offer. Because of this discretion, public sector employees are likely to attribute at least some of their job autonomy to the supervisor’s care and supportiveness.

In keeping with this work, we suggest that autonomy is an important antecedent of the perceived quality of supervision, and in particular the individualized consideration dimension of transformational leadership. Employees who have more autonomy over their work behavior are likely to see their work situation as more favorable and may attribute their favorable working conditions to the support and consideration of the supervisor, leading to perceptions of individualized consideration. Low autonomy workers, on the other hand, may attribute their unfavorable conditions to unsupportive treatment and decisions of their supervisors. Hence, employees who have more job autonomy should hold more positive perceptions of their supervisor’s individualized consideration.

Hypothesis 1: Autonomy is positively associated with perceived supervisor individualized consideration.

Individualized Consideration and Work Outcomes

Meta-analytic reviews on leadership (Judge & Piccolo, 2004; Lowe et al., 1996; G. Wang et al., 2011) demonstrate strong support for positive correlations between dimensions of transformational leadership and work outcomes. Literature on the leadership–task performance relationship presumes that because transformational leaders who engage in individualized consideration behavior attend to the needs of their followers and treat them with respect, employees feel trust in and satisfaction with the leader (Lowe et al., 1996; Podsakoff et al., 1990). Such employees also perceive support from their supervisor to perform their formal tasks, which can help increase task performance. Prior research has shown that a transformational leader’s individualized consideration is positively associated with both individual and organizational outcomes such as satisfaction with leadership, job satisfaction, unit performance (e.g., Howell & Avolio, 1993), and organizational citizenship behaviors (OCBs; e.g., Pillai et al., 1999; Podsakoff et al., 1996; H. Wang et al., 2005).

In this study, we focus on task performance and OCB as outcomes of individualized consideration. OCB has been defined as “individual behavior that is discretionary, not directly or explicitly recognized by the formal reward system” (Organ, 1988, p. 4) and “discretionary activities that fall outside of the employee’s actual job description” (Ambrose & Kulik, 1999, p. 245). According to Williams and Anderson (1991), OCB can be categorized into two dimensions: OCBI (individuals) and OCBO (organization). Although OCBI is directed toward specific individuals, for example, coworkers or supervisors, OCBO is directed toward the broader organization.

Based on the aforementioned transformational leadership literature, which suggests that individualized consideration increases employee performance through the increased trust and respect that it generates between the subordinate and the leader, we hypothesize the following:

Hypothesis 2a: Perceived supervisor individualized consideration is positively associated with task performance.

Hypothesis 2b: Perceived supervisor individualized consideration is positively associated with OCB.

Individualized Consideration as a Mediator Between Autonomy and Work Outcomes

We propose that job autonomy is a signal that a supervisor considers his or her employees’ needs. This perception of a supervisor’s favorable intent should in turn lead to the employee’s reciprocation through greater effort and performance. To explain why autonomy may affect performance behaviors through individualized consideration, we invoke social exchange theory (Blau, 1964) and organizational support theory (Eisenberger et al., 1990; Rhoades & Eisenberger, 2002). Based on the principles of social exchange theory, reciprocal interdependence emphasizes contingent interpersonal transactions, whereby one party’s action is contingent on another (Cropanzano & Mitchell, 2005). If one party provides a favorable benefit, the receiving party should respond in kind with favorable treatment. In line with the social exchange theory, the perceived organizational/supervisor support literature notes that, as part of the social exchange process, employees form “a general perception concerning the extent to which the organization values their contributions and cares about their well-being” (Eisenberger et al., 1990, p. 51). The treatment received from the employee’s organization is viewed as an indication of whether the organization favors or disfavors employees (Stinglhamber & Vandenberghe, 2003). This perception of supportiveness is informed in part through the working conditions that employees experience (Kurtessis et al., 2017).

Along these lines, recent work on organizational support theory shows how perceived supervisor/organizational support plays a key role in mediating the relationship between a variety of work conditions and work outcomes. For example, in their longitudinal study, Stinglhamber and Vandenberghe (2003) contended that perceived supervisor supportiveness mediated the effect of favorable job conditions such as autonomy on commitment to their supervisor. Moreover, perceived organizational support has been shown to mediate the association between perceptions of supportive human resources practices (participation in decision-making, growth/career development opportunities, internal promotion, and fairness of rewards/recognition) and work outcomes including organizational commitment and job satisfaction (Allen et al., 2003) and turnover intention (Foong-ming, 2008). Previous studies have also shown that perceived situational factors (distributive/procedural justice, communication satisfaction with supervisor, and labor–management relationship climate) affected work outcomes such as affective organizational commitment (Moideenkutty et al., 2001), OCB (Moorman et al., 1998), and job satisfaction (Ohana, 2012) via their influence on perceived organizational support.

Based on these studies, it appears that perceptions of supervisor and organizational supportiveness mediate the positive relationship between favorable working conditions and affective/behavioral outcomes. Individualized consideration, as described earlier, is a strong indicator of the perceived supportiveness of a supervisor and reflects the supervisor’s effectiveness in providing support (van Knippenberg & Sitkin, 2013). In line with the principles of social exchange theory (Blau, 1964) and perceived organizational and supervisor support (Eisenberger et al., 1990; Rhoades & Eisenberger, 2002), we assume that employees’ perceptions of how their supervisor considers their individual needs play a key role in the reciprocation of the favorable treatment offered by their supervisors. Therefore, we suggest that autonomy is related to performance behaviors largely because perceptions of autonomy are attributed to the supervisor’s favorable intent and individualized care. This attribution of favorable intent is an important part of a high-quality social exchange relationship. The employee’s general perception that his or her supervisor is considering his or her individual needs is likely to prompt the employee to reciprocate with increased performance behaviors. We thus propose the following hypotheses:

Hypothesis 3a: Perceived supervisor individualized consideration mediates the effect of autonomy on task performance.

Hypothesis 3b: Perceived supervisor individualized consideration mediates the effect of autonomy on OCB.

Incongruence/Interaction Between Autonomy and Preference for Autonomy

Kalleberg (2008) contended that when there is a match or fit between individuals’ needs, preferences, values, and abilities and their jobs, individuals seem to show a high level of satisfaction with their work. This also influences workplace effectiveness (Kalleberg, 2008). By contrast, a mismatch or lack of fit may result from incongruence between an individual’s preferences and the corresponding characteristics of their jobs (Kalleberg, 2008). When mismatches exist, a wide range of challenges seem to occur for individuals and their organizations (Kalleberg, 2008).

Based on the literature, we argue that in organizations where a great deal of job autonomy and freedom are granted to employees, their supervisors are more likely to be perceived as valuing individual needs and freedom and may be positively evaluated on transformational leadership criteria. However, we suggest that this is the case only to the extent that employees prefer to work autonomously. If job autonomy exceeds that which employees prefer, then the benefits of autonomy for perceptions of individualized consideration may diminish. Thus, the fit between autonomy and preference for autonomy may facilitate the perception of a transformational leader. To the extent that there is a misfit between autonomy and preference for autonomy, it can be expected that such conditions will be associated with lower perceived supportiveness from their supervisors (Ford, 2012). It is expected that workers who perceive their work preferences to be unmet will perceive that their supervisors are less considerate of their individual needs and preferences, resulting in negative affective responses. Ford (2012) showed that misfit between autonomy and the occupational role’s norms for autonomy is related to lower perceived support and experienced frustration and dissatisfaction, pointing out,

A greater level of autonomy and freedom than typically granted in a job may be associated with a lack of clarity and direction and more ambiguity for workers who are unaccustomed to or do not value autonomy, whereas too little autonomy is expected to increase strain and dissatisfaction due to the desire for individuals have to have some control over their work. (p. 414)

To test the incongruence (or misfit, mismatch) effect, we propose the following hypothesis:

Hypothesis 4: Incongruence between autonomy and preference for autonomy is negatively associated with perceived supervisor individualized consideration.

We expect that preference for autonomy does not directly influence work outcomes; rather, it moderates the relationship between autonomy and work outcomes through perceptions of supervisor individualized consideration. We suggest that one reason why the fit between autonomy and preference for autonomy predicts employees’ performance behaviors is that when job autonomy matches employee preferences, this signals that they are valued by their supervisors. This is consistent with organizational support theory (Eisenberger et al., 1990) and the reasoning behind the hypothesized mediated relationship between autonomy and employee behavior through individualized consideration. When the autonomy fits the employee’s preference, this should result in an even stronger effect on perceptions of individualized consideration. In turn, this impression should prompt employees to reciprocate through better performance and citizenship. Therefore, the following is hypothesized:

Hypothesis 5a: The indirect effect of autonomy on task performance through individualized consideration is stronger when preference for autonomy is high.

Hypothesis 5b: The indirect effect of autonomy on OCB through individualized consideration is stronger when preference for autonomy is high.

Method

Sample and Procedure

Data for this study were obtained from full-time professional employees working in a large state government in the United States via a self-administered web-based survey. The survey assured participants that participation was voluntary and that all data would remain confidential. An email letter including a link for the survey was sent to state government employees through their union’s listserv (approximately 50,000 union employees). Of the potential participants who received the link to the survey in an email, 824 completed the survey.¹ Because of missing responses to one or more study variables, only 642 questionnaires were included in the analysis. Survey respondents represent a variety of job areas (e.g., Counseling and Social Work, General Administration). The majority of participants were White/Caucasian (86%), and the respondents were broadly balanced between male (43%) and female (57%) employees. More than 80% had at least a bachelor’s degree and 45% had supervisory/managerial status. The mean age was 51 years, and the average tenure was 18 years.

In their meta-analysis, Schlam and Kelloway (2001) demonstrate that there is no significant correlation between response rate and effect size in self-report survey and that suggests “response rate is not likely to adversely affect the validity of findings” (p. 163). Moreover, Goldberg (2003) suggests that the effect of nonresponse on the relationships between variables is limited. Nevertheless, because the response rate was still low, it may be that nonresponse bias potentially occurred (Newman, 2009; Rogelberg & Stanton, 2007). To estimate if there were any major differences between our survey sample and the nonrespondents, additional analyses were employed, following Rogelberg and Stanton’s (2007) guidelines.

First, to check for representativeness, we compared descriptive statistics for several demographic variables (i.e., age, service years, percentage of White, and percentage of female) for each of the participating agencies and for the state workforce (using the State Workforce Management Report) and found minimal differences (for the survey sample, the average age was 51, the average length of service was 18 years, females comprised 57% of the sample, and 86% of the sample were White; for the state workforce, the average age was 48, the average length of service was 15 years, females comprised 50% of the workforce, and 74% of the workforce were White).

Second, following Armstrong and Overton’s (1977) assumption that late respondents can be temporarily treated as nonrespondents, we conducted a test to determine whether there are correlations between the variables in the sample of this study and the number of days it took respondents to complete survey. We found no difference in the major variables between early returners and late ones. Our findings showed that time of completion of the survey was not significantly correlated with major variables of this study: task performance (r = −.07, n.s.), OCBI (r = −.05, n.s.), OCBO (r = −.00, n.s.), autonomy (r = .01, n.s.), preference for autonomy (r = −.02, n.s.), or individualized consideration (r = −.01, n.s.). Taken together, these results suggest that our substantive variables were unlikely to bias response rates (Rogelberg & Stanton, 2007) and nonresponse bias was unlikely to play a role in the interpretation of the results.

Measures

Participants responded to the items using a five-point Likert-type scale ranging from 1 = strongly disagree to 5 = strongly agree, except for the measure of individualized consideration, which ranged from 1 = not at all to 5 = frequently, if not always used. Items used to measure the major variables are included in the Appendix.

Autonomy measures an individual’s sense of freedom in making choices about his or her task. This three-item scale had a coefficient alpha reliability of .85. An example of the items used to measure autonomy is “I have significant autonomy in determining how I do my job.” Preference for autonomy measures the degree to which people prefer a job that gives them autonomy in processes or ways of performing their work. The measure used in this study was also adapted from Hackman and Oldham’s (1980) Job Diagnostic Survey and included items such as “I prefer a job that gives me the chance to use my personal initiative and judgment in carrying out the work.” This three-item scale had a reliability of .75.

Individualized consideration, one of four behavioral components of transformational leadership, was measured using the Multifactor Leadership Questionnaire (MLQ) developed by Bass and Avolio (1990). The MLQ is a well-validated measure of transformational leadership and has been used extensively in leadership research (Ambrose & Kulik, 1999). Employees were asked to evaluate how frequently their direct supervisor engages in transformational leadership behaviors using the three-item individualized consideration scale. This scale had a reliability of .77. An example of an individualized consideration item is “My supervisor considers me as having different needs, abilities, and aspirations from others.”

Task performance was assessed using Williams and Anderson’s (1991) measure, which asks individuals to rate their own task performance, for example, “I always complete assigned duties in a timely fashion.” Based on a factor analysis, one item (“I will sometimes procrastinate on tasks that I do not enjoy doing”) with a low factor loading (less than 0.5) was excluded from further analysis. This scale (two items) had a reliability of .59, which shows marginal reliability. Eisinga et al. (2013) argue that using Cronbach’s alpha underestimates true reliability in cases of two-item measures. Instead, they suggest using the Spearman–Brown coefficient for measures comprising only two items, as it provides a better fit for two-item measures than does Cronbach’s. Following Eisinga et al.’s (2013) recommendation, the Spearman–Brown coefficient was calculated instead of Cronbach’s alpha for task performance measures. The Spearman–Brown coefficient for the task performance scale was equal to .74.

Organizational citizenship behavior (OCB) includes employees’ activities that fall outside of their actual job description. This study measured employees’ citizenship behaviors using Williams and Anderson’s (1991) two dimensions: OCB targeted at individuals, that is, coworkers or supervisors (OCBI), and OCB directed toward the organization (OCBO). In this study, six items were included by selecting items from both Podsakoff et al.’s (1990) and Williams and Anderson’s (1991) measure. Three items were used for OCBI and three items for OCBO. Examples of items include “I willingly help others who have work-related problems” (OCBI) and “I attend functions and events that are not required, but which help me learn more about what is going on in the organization” (OCBO). The reliabilities of the OCBI and OCBO scales were .64 and .65, respectively.

In addition, job areas² and supervisory status (1 = nonsupervisor; 2 = supervisor) were included as control variables.

Data Analysis

To test the indirect effects of autonomy and preferred autonomy on the outcome variables via individualized consideration, we used a moderated mediation approach recommended by Preacher and Hayes (2008). This method allows us to assess the indirect mediated effects using the macro program PROCESS for SPSS based on bootstrapped confidence intervals (Preacher & Hayes, 2008). We tested the significance of the indirect effects using bootstrapping and estimated bias-corrected confidence intervals for the indirect effects by boot strapping 1,000 samples.

Furthermore, to test the congruence effects, we used polynomial regression and response surface modeling (Edwards, 1996; Edwards & Parry, 1993) to determine the precise nature of the relationship between autonomy and preferred autonomy and outcome variables. Following the procedure provided by Edwards (1996), the dependent variables (individualized consideration, task performance, and OCB) were regressed on five polynomial terms, that is, autonomy (X), preference for autonomy (Y), autonomy squared (X²), autonomy × preference for autonomy (X × Y), and preference for autonomy squared (Y²). After polynomial regressions were performed, additional tests were conducted to examine the slope and curvature along two critical lines of interest, that is, the congruence line (X = Y) and the incongruence line (X = −Y). We also used response surface graphs to visually present the polynomial regression results.

Treatment of Common Method Variance

All data on autonomy, individualized consideration, preference for autonomy, and work outcomes were self-reported using a single questionnaire, which could increase the possibility of common method variance (CMV; Podsakoff et al., 2003). We conducted two tests to assess the existence of method variance. The first test is the unmeasured latent methods factor technique (i.e., single-common-method-factor approach) to control for the effects of an unmeasured latent methods factor on the variables. As recommended by Podsakoff et al. (2003), the analysis began with a model that included the latent CMV factor, χ²(214) = 566.441, root mean square error of approximation (RMSEA) = .04, comparative fit index (CFI) = .96, standardized root mean square residual (SRMR) = .03, and compared this model with a nested model that excluded the latent CMV factor, χ²(239) = 788.43, RMSEA = .05, CFI = .93, SRMR = .05. Although the method factor did improve measurement model fit (chi-square difference: χ²(25) = 221.99, p < .05), loading parameters on method factor are not large (i.e., the average squared loading = 0.08). In addition, a general comparison of the correlations with latent variable not accounted for and those with the latent factor introduced indicated that there was a slight difference in the magnitudes of correlations between variables after accounting for CMV, but the differences between the two were not large. As shown in Table 1, most of the correlations among the latent factors did not change to a large degree when controlling for the CMV in their statistical significance or direction, suggesting that CMV only had a small influence on observed relationships.

Table 1.

Correlation, mean, and standard deviation.

Variable	M	SD	1	2	3	4	5	6
1. Task performance	4.18	0.70	—	.82 (.79)	.73 (.71)	.16 (.07)	.24 (.18)	.52 (.54)
2. OCBI	3.95	0.63	.53	—	.79 (.77)	.22 (.14)	.23 (.17)	.53 (.55)
3. OCBO	3.83	0.70	.42	.53	—	.19 (.01)	.24 (.11)	.47 (.52)
4. Individualized consideration	3.11	1.12	.06	.16	.15	—	.47 (.33)	.09 (.14)
5. Autonomy	3.65	0.95	.12	.12	.14	.39	—	.20 (.24)
6. Preference for autonomy	4.37	0.57	.35	.32	.30	.04	.12	—

Note. N = 642 for all variables. Below the diagonal are correlations among study variables. Above the diagonal are correlations among latent factors. In parentheses above the diagonal are correlations among latent factors controlling for a latent common method factor. Correlations in boldface type were significant at the .05 alpha level. OCBI = organizational citizenship behavior (individuals); OCBO = organizational citizenship behavior (organization).

Second, we conducted the marker variable approach to detect CMV (Eichhorn, 2014; Lindell & Whitney, 2001; Richardson et al., 2009). According to Richardson et al. (2009), ideal markers mean “marker variables with no expected theoretical relationship with substantive variables” (p. 768). Butts et al. (2009) note, “for CMV to be an issue it must be of sufficient magnitude to have systematic influence, rather than random influence, across all items sharing the same method” (p. 127). Strong CMV can be detected when a subset of survey items that are randomly selected from different construct measures shows high reliability (the alpha coefficient) and indicates a good fit from a confirmatory factor analysis (CFA; Butts et al., 2009). According to the approach, the marker variable was created from a subset of six items³ that were randomly chosen from the approximately 50 items of the original survey instrument that were not included in this study. The alpha coefficient was very low (α = .18) among the selected six items, and the fit indices indicate a low model fit (χ²/df = 54.62/9, CFI = .51, normed fit index [NFI] = .51, RMSEA = .08). From the analyses above, CMV does not seem to have systematic influence across all survey items. Although CMV exists in this study, the tests suggest that the method bias was not large enough to influence the substantive relationships of the current model.

Results

Simple Mediation Result

The simple mediation results of the hypothesized model are provided in Table 2. Consistent with our expectations, autonomy was positively related to individualized consideration (β = 0.429, p < .05) and individualized consideration was positively associated with task performance, OCBI, and OCBO (β = 0.059, 0.079, and 0.068, respectively, p < .05), providing support for Hypotheses 1, 2a, and 2b. The indirect effects of autonomy on task performance, OCBI, and OCBO through individualized consideration were positive and significant (β = 0.025, 0.034, and 0.29, respectively, p < .05), providing support for Hypotheses 3a and 3b.

Table 2.

Simple Mediation Result of Autonomy.

Predictor	Individualized consideration	Task performance	OCBI	OCBO
Supervisory status	−0.004	0.003	0.136*	0.046
Autonomy	0.429*	0.067*	0.031	0.058
Individualized consideration		0.059*	0.079*	0.068*
R ²	.141* (F = 30.06)	.023* (F = 3.24)	.045* (F = 6.57)	.030* (F = 4.29)
Bootstrap indirect effects[lower limit to upper limit 95% confidence interval]		0.025* [0.003, 0.052]	0.034* [0.012, 0.064]	0.029* [0.008, 0.058]

Note. N = 642. Controlled for job areas. Bootstrap sample size = 1,000. Unstandardized regression coefficients are reported. OCBI = organizational citizenship behavior (individuals); OCBO = organizational citizenship behavior (organization).

p < .05.

Incongruence Effect for Autonomy

We report the results of the polynomial regression in Table 3. As presented in Table 3, five terms (X, Y, X², X × Y, Y²) were used to represent the relationship between autonomy and preference for autonomy, and the R² attributed to the five key predictors related to autonomy was significant for all outcome variables, although the individual incremental contributions of the quadratic terms to explained variance individually and their combined contributions were very small or not significant.

Table 3.

Polynomial Regression Results of Autonomy.

Dependent variable	Individualized consideration	Task performance	OCBI	OCBO
Step 1
Supervisory status	0.006	0.055	0.147*	−0.063
Autonomy	0.423*	0.067*	0.045*	0.493*
Preference for autonomy	−0.068	0.317*	0.299*	0.628*
R²	.141*	.074*	.086*	.070*
Step 2
Autonomy × Preference for autonomy	0.173*	−0.149*	−0.089*	−0.093*
R²	.149*	.090*	.093*	.077*
ΔR²	.007*	.016*	.007*	.007*
Step 3
Autonomy square	−0.016	0.050*	0.011	−0.012
Preference for autonomy square	−0.018	0.004	0.073	−0.026
R²	.149*	.099*	.096*	.077*
ΔR²	.001	.009*	.003	.001
Step 4
Individualized consideration		0.027	0.085*	0.073*
R²		.103*	.113*	.088*
ΔR²		.001	.017*	.011*
Index of moderated mediation [lower limit to upper limit 95% confidence interval]		0.010* [0.002, 0.029]	0.015* [0.004, 0.036]	0.013* [0.003, 0.036]

p < .05.

In Table 3, the result shows that there was a significant interaction between autonomy and preference for autonomy in predicting task performance, OCBI, and OCBO (β = −0.149, −0.089, and −0.093, respectively, p < .05). As shown in Table 3, the predictor variables were entered into the equation in successive steps. The set of main variables (autonomy and preference for autonomy) and control variable (supervisory status) were entered first. Next, the interaction term (autonomy × preference for autonomy) was entered. The square terms (autonomy square and preference for autonomy square) were entered third. The results in Table 3 also indicate that the square terms (Step 3) added to the prediction of task performance for models examining autonomy and preference for autonomy (ΔR² = .009, p < .05), indicating that there were curvilinear effects of autonomy and preference for autonomy in predicting task performance.

In addition, as shown in Table 4, we used Edward’s method for testing the curvature along the line of congruence and incongruence. The line of incongruence refers to a line in three-dimensional space from left to right, ranging from extreme incongruence (high preference for autonomy and low autonomy), to complete congruence, to extreme incongruence (low preference for autonomy and high autonomy) again. A negative curvature along the line of incongruence provides evidence that there is a strong congruence effect because it suggests that levels increase until there is congruence before decreasing once again. For individualized consideration, the curvature along the line of incongruence was negative, although it showed a moderately significant coefficient (=−0.288, p < .10), suggesting that perceived individualized consideration was highest among individuals whose autonomy matched their preference for autonomy. For task performance and OCBI, the curvatures for autonomy along the line of incongruence were positive (coefficients = 0.221, p < .05, and 0.148, p < .10, respectively), indicating that the slopes were curving upward. This, in combination with the significant interaction, is consistent with a simple interaction between autonomy and preference for autonomy in predicting these variables. Finally, for task performance, OCBI, and OCBO, the slope coefficients along the line of congruence for autonomy were 1.211, 0.081, and 1.152, respectively (p < .05) after controlling for individualized consideration, indicating that the slopes were positive. These results once again support a simple interaction between autonomy and preference for autonomy in predicting these outcomes.

Table 4.

Coefficients for Slope and Curvature Across Lines of Congruence and Incongruence.

Dependent variable	Line of congruence		Line of incongruence
Dependent variable	Slope	Curvature	Slope	Curvature
Individualized consideration (IC)	−0.405	0.112	0.570	−0.288^†
Task performance	0.910^†	−0.058	−0.316	0.221*
Task performance with IC included	1.211*	−0.095	−0.586	0.192^†
OCBI	0.570	−0.027	−0.013	0.148^†
OCBI with IC included	0.801*	−0.065	0.023	0.176*
OCBO	1.081*	−0.089	−0.159	0.088
OCBO with IC included	1.152*	−0.105	−0.066	0.116

Note. OCBI = organizational citizenship behavior (individuals); OCBO = organizational citizenship behavior (organization).

†

p < .10. *p < .05.

Figure 2(a) to (d) shows selected surface plots for preference for autonomy in predicting work outcomes. Figure 2(a) illustrates the response surface for individualized consideration. As shown in Figure 2a, the surface along the incongruence line curved downward. Individualized consideration decreased the most when autonomy was at the highest and preference for autonomy was at the lowest, or autonomy was at the lowest and preference for autonomy was at the highest, whereas it increased the most among employees who had high levels of both autonomy and preference for autonomy. That is, the graph demonstrated that any incongruence on autonomy was associated with lower individualized consideration.

Figure 2.

Response surface modeling.

For task performance, OCBI, and OCBO, as demonstrated in Figure 2(b) to (d), the slopes of the surface along the congruence line were significantly positive, which suggests that employees’ work outcomes were higher when both autonomy and preferred autonomy were high than when both were low. However, the areas of the surfaces for these models contained a mixture of modest concave and convex features, indicating weak support for the congruence/incongruence hypotheses predicting the outcome variables. Therefore, response surface modeling suggests that only in predicting individualized consideration was there support for a congruence effect. The remaining models provide support for an interaction between autonomy and preference for autonomy. Taken together, although support for congruence effects was strong in predicting individualized consideration and weak in predicting the performance outcomes, strong interaction effects contributed meaningfully to the explanation of variance in all models, providing partial support for Hypothesis 4.

Moderated Mediation Effect for Autonomy

Table 3 also indicated moderated mediation effect for autonomy. When we entered interaction term in the model in Step 2, it incrementally predicted individualized consideration, accounting for 14.9% of the variance in individualized consideration (ΔR² = .007, p < .05). In Table 3, we found that the interaction between autonomy and preference for autonomy was positively associated with individualized consideration (β = 0.173, p < .05). The indirect effects on the three work outcomes (i.e., task performance, OCBI, and OCBO) via individualized consideration were significant (moderated mediation coefficients = 0.010, 0.015, and 0.013, respectively, p < .05), as indicated when the lower and upper levels of the confidence intervals did not include zero, providing support for Hypotheses 5a and 5b. The conditional indirect effects of autonomy on work outcomes (i.e., task performance, OCBI, and OCBO) were more positive when preference for autonomy was high (β = 0.032, 0.041, and 0.042, respectively, p < .05) than when it was low (β = 0.021, 0.033, and 0.028, respectively, p < .05), although in both conditions the effects were significantly positive.

Discussion

Drawing upon prior research on work structure and P-E fit, we proposed that autonomy and preference for autonomy interact to influence perceptions of the supervisor’s individualized concern for the employee, which in turn leads to increased effort by the employee to help the organization through better citizenship and performance. This study aims to provide a new perspective with respect to autonomy, suggesting that organizations and supervisors do not always benefit from higher levels of employee autonomy. Hence, this research provides a different angle to look at the mechanism through which the human-resource-related work environment influences employee attitudes by focusing on an important variable that has been frequently addressed by human resource management and public administration scholars, namely, autonomy. We sought to expand our understanding of the interaction or fit between individual and environmental factors that predict more positive perceptions of supervision and better work outcomes.

In working toward this goal, we make three important contributions. First, we contribute to work on structure in organizations by showing that workers vary in their preference for autonomy and that this preference interacts with autonomy in a meaningful way to influence work outcomes. Research has largely ignored individual aspects of preference for structure when examining the organizational characteristics–outcomes relationship. As suggested by the P-E fit approach, in this study we explicitly incorporated preference for autonomy into our model.

Second, although prior P-E fit research has typically examined the direct relationship between P-E fit and employee outcomes without delineating the underlying mechanisms or examined psychological states as mediators (Cable & Edwards, 2004; Fried & Ferris, 1987), we move beyond these existing perspectives by theorizing that the perceived quality of supervision is a critical mechanism through which autonomy can facilitate work outcomes. By exploring this mediation model, we provide a direct test of the potential mechanism that links P-E fit to workers’ outcomes. As such, this study extends the leadership literature by examining P-E fit as one of the antecedents of the perceived quality of leadership.

Finally, with respect to the measures of P-E fit, this investigation differs from previous studies of P-E fit in that this research uses an indirect measure, which overcomes some limitations of previous studies. Previous public administration studies on P-E fit (e.g., Bright, 2007) have used direct measurement by asking employees explicitly whether they believe that a good fit exists (e.g., “I think that I fit well . . .”; “To what degree do you feel your values ‘match’ or fit this employer?”). However, direct measures of P-E fit have been criticized (Kristof, 1996) because they prevent estimation of the independent effects of the person and environment by confounding their constructs (Edwards, 1991) and potentially influence the work-related attitude outcomes via consistency bias, that is, “I think that I fit well, so I must be satisfied with my job” (Salancik & Pfeffer, 1977). To address some drawbacks of direct measures, P-E fit scholars suggest using indirect measurement by having employees explicitly rate individual preferences and job/organizational characteristics. In this study, employees rated their personal preferences for autonomy (e.g., “I prefer a job where I can prioritize my tasks”) and also rated how much the corresponding variables were operating in the organization (e.g., autonomy measurement “I can decide on my own how to go about doing my work.”), using the indirect measure to examine P-E fit/interaction effects.

Our results from polynomial regression analyses suggested that any incongruence on autonomy was detrimental to perceived individualized consideration. This finding suggests that it is critical to examine the congruence between autonomy and preference for autonomy rather than the autonomy of the job alone in predicting positive attitudes toward supervision. Although strong congruence effects were not present in all models, our results supported strong indirect effects of the P-E interaction through perceptions of supervision. The interaction between autonomy and preference for autonomy played a significant role in determining perceptions of supervision, which in turn was related to work outcomes, representing another core finding of this study. The presence of powerful interaction effects confirms that there was an overall tendency toward higher individualized consideration perceptions when autonomy and preference for autonomy were high. This was in turn related to more positive work outcomes among employees.

Theoretical Implications

This study has several theoretical implications regarding P-E fit, transformational leadership, and work outcomes. By integrating work structure and the P-E fit framework into the study of individualized consideration and the associated employee work outcomes, we extended leadership and work design research by revealing that employee work perceptions influence the perceived quality of supervision, which in turn translates into positive work outcomes.

First, this study found support for the mediating role of individualized consideration in the P-E fit–work outcomes relationship with respect to autonomy. This finding adds to our understanding of factors important for perceptions of supervisor individualized consideration. Previous research has shown that cultural or structural contexts (e.g., Pillai & Meindl, 1998) are related to greater perceptions of transformational leadership. By investigating autonomy and preference for autonomy in the form/perspective of P-E fit, we have demonstrated that individuals with greater P-E match with respect to autonomy are likely to view their supervisors as strong on individualized consideration and in turn exhibit more positive work behaviors. This finding is important because although there has been considerable research into person, environment, or P-E fit, the importance of these factors has not been empirically tested in the context of perceptions of transformational leadership attributes. Hence, this contributes to our understanding of the mechanism/process by which job aspects and the corresponding individual preferences are translated into perceptions of supervisor leadership quality and important employee outcomes at work.

Second, our findings demonstrate that the mismatch between autonomy and preferred autonomy appeared to be most detrimental, suggesting that employees’ autonomy may not always be beneficial to their attitudes and outcomes. These findings urge us to incorporate the individualized consideration dimension into the theoretical framework for understanding the outcomes of employee perceptions of autonomy. Although in transformational leadership theory such variables as work design structure, culture, and leader personality are proposed as impacting perceived leadership, our results suggested that autonomy congruence also predicts perceived supervisor individualized consideration and work outcomes. Although the conventional wisdom holds that a high level of autonomy is beneficial and its positive effects on work outcomes have been emphasized by previous studies (e.g., Bandura, 2001; Conger & Kanungo, 1988; Hackman & Oldham, 1980), this study provides a new view on autonomy by establishing the autonomy congruence perspective and builds on existing theories to better understand the mechanism by which the beneficial effects of autonomy are translated into positive work outcomes through perceived individualized consideration.

Still, the autonomy congruence effect in predicting work performance and citizenship behavior was weak or not supported; only the interaction between autonomy and preference for autonomy was significant in predicting these outcomes, whereas the curvilinear effects were not. Because the curvilinear effects were nonsignificant in predicting the outcomes, this suggests that too much autonomy was not necessarily harmful for these outcomes; only too little autonomy was, contingent on one’s preferences. Too much autonomy only seemed to be harmful for perceptions of the supervisor’s individualized consideration. It is possible that excessively high levels of autonomy for workers who prefer less autonomy may still result in more positive perceptions of other aspects of transformational leadership, offsetting the costs of individualized consideration. Future research is needed to explore these possibilities by incorporating other mediators, such as other perceived characteristics of the supervisor’s leadership quality, in the relationship between P-E fit and work outcomes.

Finally, as indicated, there has been a research stream in the field of public administration linking PSM to work outcomes through P-E fit. Our research suggests that there may be additional mediator (i.e., individualized consideration) of the PSM and work outcomes relationship. Although previous studies on PSM suggest that P-E or P-O mediates the relationship between PSM and work-related outcomes (Bright, 2007; Kim, 2012; Steijn, 2008; Wright & Pandey, 2008), the quality of supervision might also mediate P-E fit and work outcomes in the relationship. It may be that high levels of PSM enable employees to perceive more fit (Bright, 2007), and that the perception of P-E fit is related to more positive views of supervision and enhanced work outcomes.

Managerial Implications

The P-E fit approach is reflected in a wide range of management practices designed to enhance employee motivation and performance. The findings of this study suggest that public managers should consider how developing their leadership behaviors might enhance their attempts to effectively manage employees within their work unit when implementing human resources and motivation programs. Although a variety of human-resource-related programs are employed as effective tools in public sector agencies, not all employees welcome such programs and not all work units are appropriate for these approaches. Findings from this study suggest that organizations facing difficulties with human resources programs should look at reducing the incongruence between employees’ needs and the conditions offered by their work environment. Our results also indicate that work outcomes were at their highest levels only when both autonomy and preferred autonomy ratings showed the highest levels, rather than the lowest levels.

It is also important to note that low autonomy may not always lead to the most negative perception of individualized consideration. In fact, in a situation in which both autonomy and preferred autonomy are low, a relatively high level of individualized consideration may still be obtained. Therefore, organizations should focus on preventing situations where workers’ preferences for structure go unmatched when they establish policies to encourage participation or discretion. For some workers, high levels of autonomy likely increase ambiguity and workload beyond that which they would prefer. In doing so, supervisors can seek feedback from their employees (e.g., by directly asking them). Employees can help the supervisor obtain information about their preferences so that supervisors can modify the level of discretion they give to employees based on the preferences of employees. Although similar positions have similar tasks, there is usually some discretion for supervisors to use.

Finally, it should be noted that although the amount of autonomy that a job can offer is in part constrained by the job’s context, there is also room for discretion within that job context for managers to grant or restrict employee autonomy. Many public sector jobs that are organized as functional areas and are highly specialized (e.g., jobs that require significant expertise for many projects) may need high levels of autonomy, whereas some jobs in the public sector function in a rigid and mechanical system in which employees rely strictly on a chain of command that may constrain the amount of autonomy that a job can offer. Yet, it is often possible for managers, within the constraints of the job context, to exercise some discretion in the control they allow their subordinates to have over their own work. For example, a manager may establish participatory or collaborative practices within his or her unit to deal with complex and dynamic situations, offering employees the opportunity to have influence over the informal practices that are used to make some decisions.

Limitations and Future Research

Notwithstanding these contributions, this study also has several limitations. First, because all data on autonomy, individualized consideration, preference for structure, and work outcomes were self-reported using a single questionnaire, it could increase the possibility of CMV (Podsakoff et al., 2003). Thus, the rather strong links among predictor variables and employee attitude/performance behavior variables may be partly a consequence of CMV. Although our tests to detect CMV indicated the existence of such bias as shown before, examining preference for structure variables as moderators in this study may mitigate some of these concerns because interactions are not likely to be influenced by common method bias (Jakobsen & Jensen, 2015; Siemsen et al., 2010). In addition, we use indirect measures (i.e., structure, preference for structure), which might reduce the influence of method variance, as the P-E fit literature points out (Jung, 2018; Kristof, 1996). Nevertheless, future studies might attempt to use more rigorous research designs by obtaining data from different sources (e.g., supervisor and nonsupervisor ratings; survey and archival data) or at multiple time periods using a panel data design, as Jakobsen and Jensen (2015) suggested, to reduce this potential problem of method variance.

Another limitation of this study relates to the measurement of variables. For example, some variables included in this study were measured with only a limited number of items because longer questionnaires may be costly and time-consuming for participants to complete, yielding lower response rates. Related to this, some of our measures (e.g., task performance) showed marginal reliabilities. Although we provided the Spearman–Brown coefficient on task performance (=.74), which was in this study measured using two items, following Eisinga et al.’s (2013) recommendation, OCBI and OCBO measures also showed marginal reliabilities. We note that the OCB measures could probably be viewed as formative indicators of the constructs they represent, rather than as reflective indicators, given that the items assess discrete behaviors that combine to form the OCB constructs. Some researchers (e.g., Bollen & Lennox, 1991; MacKenzie et al., 2005) note that internal consistency reliability is not appropriate for evaluating the adequacy of formative indicators. In spite of these, future research should consider measuring these variables using multi-item scales that capture more aspects of the constructs to increase their precision and the validity of the findings.

Finally, this study used a cross-sectional survey, which requires us to be cautious in drawing any conclusions regarding causality. For example, it may be that employees who believe their supervisors are engaged in individualized consideration see their work situation as more congruent and/or positive. Thus, the relationships that emerged in this study should be understood as correlational in nature. Future research should attempt to minimize this bias by aggregating employee ratings of perceptions of work design or utilize experimental and longitudinal designs to empirically test causal direction among these variables.

Conclusion

Using a sample of state government employees, this study considered P-E fit/interaction as an element of the work context that might foster or undermine perceived supervision and work outcomes. Although employees’ autonomy has been commonly believed to be beneficial to work outcomes, we sought to contribute to the P-E fit and leadership literatures by identifying an important contingency for the outcomes of work structure, that is, individual preference for structure. We then explored perceptions of leadership quality derived from transformational leadership theory as a potential mechanism through which P-E interaction relates to work outcomes. In doing so, we highlight the importance of individualized consideration in motivating employees in public organizations. Finally, our results confirm that public organizations can benefit more from the more optimal design of jobs by matching the structure those jobs offer with worker needs, values, and preferences.

Footnotes

Appendix

Measures.

Task performance	• I always complete assigned duties in a timely fashion. • I work hard to exceed formal performance requirements of the job.
OCBI	• I willingly help others who have work-related problems. • I help orient new employees so that they are able to perform their jobs more effectively. • I regularly check with my supervisor to see if there are ways that I can be helpful to others.
OCBO	• I attend functions and events that are not required, but which help me learn more about what is going on in the organization. • I attend meetings that are not mandatory, but are considered important. • When important tasks need to be done, I am willing to work extra time to complete these tasks.
Individualized consideration	• Treats me as an individual rather than just as a member of a group. • Considers me as having different needs, abilities, and aspirations from others. • Helps me to develop my strengths.
Preference for autonomy	• I prefer a job that gives me the chance to use my personal initiative and judgment in carrying out the work. • I prefer a job that gives me considerable opportunity for independence and freedom in how I get the work done. • I prefer a job where I can prioritize my tasks.
Autonomy	• I have significant autonomy in determining how I do my job. • I can decide on my own how to go about doing my work. • I have considerable opportunity for independence and freedom in how I do my job.

Note. OCBI = organizational citizenship behavior (individuals); OCBO = organizational citizenship behavior (organization).

Declaration of Conflicting Interests

The author(s) declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.

Funding

The author(s) received no financial support for the research, authorship, and/or publication of this article.

ORCID iD

Jaehee Jong

Notes

Author Biographies

Jaehee Jong is an Assistant Professor at the Department of Public Administration in Northern Illinois University. She works on a wide range of human resources and behavior issues in the workplace, including employee attitudes toward organizations, leadership behavior, and workforce diversity.

Michael Ford is an Associate Professor of Management at the University of Alabama. His research interests include employee well-being and the employee-organization relationship.

References

Adler

P. S.

Borys

(1996). Two types of bureaucracy: Enabling and coercive. Administrative Science Quarterly, 41(1), 61–89.

Allen

D. G.

Shore

L. M.

Griffeth

R. W.

(2003). The role of perceived organizational support and supportive human resource practices in the turnover process. Journal of Management, 29(1), 99–118.

Ambrose

M. L.

Kulik

C. T.

(1999). Old friends, new faces: Motivation research in the 1990s. Journal of Management, 25(3), 231–292.

Armstrong

J. S.

Overton

S. O.

(1977). Estimating nonresponse bias in mail surveys. Journal of Marketing Research, 14, 396–402.

Bandura

(2001). Social cognitive theory: An agentic perspective. Annual Review of Psychology, 52(1), 1–26.

Bass

B. M.

(1985). Leadership: Good, better, best. Organizational Dynamics, 13(3), 26–40.

Bass

B. M.

Avolio

B. J.

(1990). Transformational leadership development: Manual for the Multifactor Leadership Questionnaire. Consulting Psychologists Press.

Bass

B. M.

Riggio

R. E.

(2006). Transformational leadership. Psychology Press.

Blau

P. M.

(1964). Exchange and power in social life. Transaction Publishers.

10.

Bollen

K. A.

Lennox

(1991). Conventional wisdom on measurement: A structural equation perspective. Psychological Bulletin, 110, 305–314.

11.

Bozeman

Kingsley

(1998). Risk culture in public and private organizations. Public Administration Review, 58, 109–118.

12.

Bright

(2007). Does person-organization fit mediate the relationship between public service motivation and the job performance of public employees? Review of Public Personnel Administration, 27(4), 361–379.

13.

Butts

M. M.

Vandenberg

R. J.

DeJoy

D. M.

Schaffer

B. S.

Wilson

M. G.

(2009). Individual reactions to high involvement work processes: Investigating the role of empowerment and perceived organizational support. Journal of Occupational Health Psychology, 14(2), 122–136.

14.

Cable

D. M.

Edwards

J. R.

(2004). Complementary and supplementary fit: A theoretical and empirical integration. Journal of Applied Psychology, 89(5), 822–834.

15.

Conger

J. A.

Kanungo

R. N.

(1988). The empowerment process: Integrating theory and practice. Academy of Management Review, 13(3), 471–482.

16.

Cropanzano

Mitchell

M. S.

(2005). Social exchange theory: An interdisciplinary review. Journal of Management, 31(6), 874–900.

17.

Edwards

J. R.

(1991). Person-job fit: A conceptual integration, literature review and methodological critique. In Cooper

C. L.

Robertson

I. T.

(Eds.), International review of industrial/organizational psychology (Vol. 6, pp. 283–357). John Wiley & Sons.

18.

Edwards

J. R.

(1996). An examination of competing versions of the person-environment fit approach to stress. Academy of Management Journal, 39(2), 292–339.

19.

Edwards

J. R.

Parry

M. E.

(1993). On the use of polynomial regression equations as an alternative to difference scores in organizational research. Academy of Management Journal, 36(6), 1577–1613.

20.

Eichhorn

B. R.

(2014). Common method variance techniques. SAS Institute.

21.

Eisenberger

Fasolo

Davis-LaMastro

(1990). Perceived organizational support and employee diligence, commitment, and innovation. Journal of Applied Psychology, 75(1), 51–59.

22.

Eisenberger

Rhoades

Cameron

(1999). Does pay for performance increase or decrease perceived self-determination and intrinsic motivation? Journal of Personality and Social Psychology, 77, 1026–1040.

23.

Eisinga

Te Grotenhuis

Pelzer

(2013). The reliability of a two-item scale: Pearson, Cronbach, or Spearman-Brown? International Journal of Public Health, 58(4), 637–642.

24.

Foong-ming

(2008). Linking career development practices to turnover intention: The mediator of perceived organizational support. Journal of Business and Public Affairs, 2(1), 1–16.

25.

Ford

M. T.

(2012). Job-occupation misfit as an occupational stressor. Journal of Vocational Behavior, 80(2), 412–421.

26.

Fried

Ferris

G. R.

(1987). The validity of the Job Characteristics Model: A review and meta-analysis. Personnel Psychology, 40(2), 287–322.

27.

Goldberg

(2003). Who responds to surveys? Assessing the effects of non-response in cross-sectional dyadic research. Assessment, 10, 41–48.

28.

Hackman

J. R.

Oldham

G. R.

(1976). Motivation through the design of work: Test of a theory. Organizational Behavior and Human Performance, 16(2), 250–279.

29.

Hackman

J. R.

Oldham

G. R.

(1980). Work redesign. Addison-Wesley.

30.

Howell

J. M.

Avolio

B. J.

(1993). Transformational leadership, transactional leadership, locus of control, and support for innovation: Key predictors of consolidated-business-unit performance. Journal of Applied Psychology, 78(6), 891–902.

31.

Jakobsen

Jensen

(2015). Common method bias in public management studies. International Public Management Journal, 18(1), 3–30.

32.

Judge

T. A.

Piccolo

R. F.

(2004). Transformational and transactional leadership: A meta-analytic test of their relative validity. Journal of Applied Psychology, 89(5), 755–768.

33.

Jung

C. S.

(2018). Current-ideal culture incongruence, hierarchical position, and job satisfaction in government agencies. International Public Management Journal, 21, 432–460.

34.

Kalleberg

A. L.

(2008). The mismatched worker: When people don’t fit their jobs. The Academy of Management Perspectives, 22(1), 24–40.

35.

Kim

(2012). Does person-organization fit matter in the public-sector? Testing the mediating effect of person-organization fit in the relationship between public service motivation and work attitudes. Public Administration Review, 72(6), 830–840.

36.

Kristof

A. L.

(1996). Person-organization fit: An integrative review of its conceptualizations, measurement, and implications. Personnel Psychology, 49(1), 1–49.

37.

Kristof-Brown

A. L.

Zimmerman

R. D.

Johnson

E. C.

(2005). Consequences of individuals’ fit at work: A meta-analysis of person–job, person–organization, person–group, and person–supervisor fit. Personnel Psychology, 58(2), 281–342.

38.

Kurtessis

J. N.

Eisenberger

Ford

M. T.

Buffardi

L. C.

Stewart

K. A.

Adis

C. S.

(2017). Perceived organizational support: A meta-analytic evaluation of organizational support theory. Journal of Management, 43, 1854–1884.

39.

Langer

Feeney

M. K.

Lee

S. E.

(2019). Employee fit and job satisfaction in bureaucratic and entrepreneurial work environments. Review of Public Personnel Administration, 39, 135–155.

40.

Leung

Chaturvedi

(2011). Linking the fits, fitting the links: Connecting different types of PO fit to attitudinal outcomes. Journal of Vocational Behavior, 79(2), 391–402.

41.

Lindell

M. K.

Whitney

D. J.

(2001). Accounting for common method variance in cross sectional research designs. Journal of Applied Psychology, 86, 114–121.

42.

Lowe

K. B.

Kroeck

K. G.

Sivasubramaniam

(1996). Effectiveness correlates of transformational and transactional leadership: A meta-analytic review of the MLQ literature. The Leadership Quarterly, 7(3), 385–425.

43.

MacKenzie

S. B.

Podsakoff

P. M.

Jarvis

C. B.

(2005). The problem of measurement model misspecification in behavioral and organizational research and some recommended solutions. Journal of Applied Psychology, 90(4), 710–730.

44.

Moideenkutty

Blau

Kumar

Nalakath

(2001). Perceived organisational support as a mediator of the relationship of perceived situational factors to affective organisational commitment. Applied Psychology, 50(4), 615–634.

45.

Moorman

R. H.

Blakely

G. L.

Niehoff

B. P.

(1998). Does perceived organizational support mediate the relationship between procedural justice and organizational citizenship behavior? Academy of Management Journal, 41(3), 351–357.

46.

Newman

D. A.

(2009). Missing data techniques and low response rates: The role of systematic nonresponse parameters. In Lance

C. E.

Vandenberg

R.J.

(Eds.), Statistical and methodological myths and urban legends (pp. 7–36). Routledge.

47.

Ohana

(2012). Perceived organizational support as mediator of distributive justice and job satisfaction: The moderating role of group commitment. The Journal of Applied Business Research, 28(5), 1063–1072.

48.

Organ

D. W.

(1988). Organizational citizenship behavior: The good soldier syndrome. Lexington Books/D.C. Heath.

49.

Pillai

Meindl

J. R.

(1998). Context and charisma: A “meso” level examination of the relationship of organic structure, collectivism, and crisis to charismatic leadership. Journal of Management, 24(5), 643–671.

50.

Pillai

Schriesheim

C. A.

Williams

E. S.

(1999). Fairness perceptions and trust as mediators for transformational and transactional leadership: A two-sample study. Journal of Management, 25(6), 897–933.

51.

Podsakoff

P. M.

MacKenzie

S. B.

Bommer

W. H.

(1996). Transformational leader behaviors and substitutes for leadership as determinants of employee satisfaction, commitment, trust, and organizational citizenship behaviors. Journal of Management, 22(2), 259–298.

52.

Podsakoff

P. M.

MacKenzie

S. B.

Lee

Podsakoff

N. P.

(2003). Common method biases in behavioral research: A critical review of the literature and recommended remedies. Journal of Applied Psychology, 88(5), 879–903.

53.

Podsakoff

P. M.

MacKenzie

S. B.

Moorman

R. H.

Fetter

(1990). Transformational leader behaviors and their effects on followers’ trust in leader, satisfaction, and organizational citizenship behaviors. The Leadership Quarterly, 1(2), 107–142.

54.

Preacher

K. J.

Hayes

A. F.

(2008). Asymptotic and resampling strategies for assessing and comparing indirect effects in multiple mediator models. Behavior Research Methods, 40(3), 879–891.

55.

Rhoades

Eisenberger

(2002). Perceived organizational support: A review of the literature. Journal of Applied Psychology, 87, 698–714.

56.

Richardson

H. A.

Simmering

M. J.

Sturman

M. C.

(2009). A tale of three perspectives: Examining post hoc statistical techniques for detection and correction of common method variance. Organizational Research Methods, 12(4), 762–800.

57.

Rogelberg

S. G.

Stanton

J. M.

(2007). Introduction: Understanding and dealing with organizational survey nonresponse. Organizational Research Methods, 10, 195–209.

58.

Ryan

R. M.

Deci

E. L.

(2000). Self-determination theory and the facilitation of intrinsic motivation, social development, and well-being. American Psychologist, 55(1), 68–78.

59.

Salancik

G. R.

Pfeffer

(1977). An examination of need-satisfaction models of job attitudes. Administrative Science Quarterly, 22, 427–456.

60.

Schlam

R. L.

Kelloway

E. K.

(2001). The relationship between response rate and effect size in occupational health research. Journal of Occupational Health Psychology, 6, 160–163.

61.

Siemsen

Roth

Oliveira

(2010). Common method bias in regression models with linear, quadratic, and interaction effects. Organizational Research Methods, 13(3), 456–476.

62.

Steijn

(2008). Person-environment fit and public service motivation. International Public Management Journal, 11(1), 13–27.

63.

Stinglhamber

Vandenberghe

(2003). Organizations and supervisors as sources of support and targets of commitment: A longitudinal study. Journal of Organizational Behavior, 24(3), 251–270.

64.

van Knippenberg

Sitkin

S. B.

(2013). A critical assessment of charismatic-transformational leadership research: Back to the drawing board? Academy of Management Annals, 7, 1–60.

65.

Wang

I. S.

Courtright

S. H.

Colbert

A. E.

(2011). Transformational leadership and performance across criteria and levels: A meta-analytic review of 25 years of research. Group & Organization Management, 36(2), 223–270.

66.

Wang

Law

K. S.

Hackett

R. D.

Wang

Chen

Z. X.

(2005). Leader-member exchange as a mediator of the relationship between transformational leadership and followers’ performance and organizational citizenship behavior. Academy of Management Journal, 48(3), 420–432.

67.

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(3), 601–617.

68.

Wright

B. E.

Pandey

S. K.

(2008). Public service motivation and the assumption of person–organization Fit: Testing the mediating effect of value congruence. Administration & Society, 40(5), 502–521.