Who can do this Annotated Bibliography on Job Satisfaction?
deyxrikus
Article9.pdf
https://doi.org/10.1177/0734371X19856082
Review of Public Personnel Administration 2020, Vol. 40(4) 717 –742
© The Author(s) 2019 Article reuse guidelines:
sagepub.com/journals-permissions DOI: 10.1177/0734371X19856082
journals.sagepub.com/home/rop
Article
A Career Stage Analysis of the U.S. Federal Employees’ Job Satisfaction and Turnover Intention: A Comprehensive Overview
Hyung-Woo Lee1
Abstract This study examines how the determinants of job satisfaction and turnover intention are different across employee’s career stages, using a large-scale data set from U.S. federal agencies. A comprehensive examination of career stage differences provides a sophisticated understanding of the multifaceted nature of job satisfaction and turnover intention factors. The trajectory of career stage differences is linear, and the impact of career stages is rather minimal when it comes to the determinants of job satisfaction, whereas the differences in terms of turnover intention were more drastic and salient. Also, the patterns of career stage effect were more complicated than proposed in the existing studies. The implications of these findings are explored for future research and practitioners.
Keywords job satisfaction, turnover intention, career stage, federal employee viewpoint survey
Introduction
People change over time. Levinson (1978) developed a model to clarify the process of evolution of the psychological needs and desires through different life phases. Applying this to the workplace, Super (1980) proposed a model of career develop- ment to explain the process by which the aspirations of employees in organization change across different career stages. Based on these models, scholars have
1Hannam University, Daejeon, Republic of Korea
Corresponding Author: Hyung-Woo Lee, Hannam University, 133 Ojeong-Dong, Daeduk-Gu, Daejeon 306-791, Republic of Korea. Email: [email protected]
856082ROPXXX10.1177/0734371X19856082Review of Public Personnel AdministrationLee research-article2019
718 Review of Public Personnel Administration 40(4)
investigated how the importance of job factors differs depending on the career stage of employees. For instance, scholars have examined the career stage difference in career decisions (Kooij, de Lange, Jansen, & Dikkers, 2008; Lam, Ng, & Feldman, 2012; McElroy, Morrow, & Wardlow, 1999), work motivation (Miao, Lund, & Evans, 2009), commitment (A. Cohen, 1991; McElroy et al., 1999), job satisfaction (Bedeian, Ferris, & Kacmar, 1992; Menguc & Bhuian, 2004), and work–life balance issues (Darcy, McCarthy, Hill, & Grady, 2012; Demerouti, Peeters, & Heijden, 2012). However, we need a more comprehensive overview of the topic. Examining the career stage difference with a focus on a single factor does not provide a bigger picture of career stage difference. A remaining question, for example, is whether the career stage difference will be manifested similarly for different job factors. Another remaining question is which job factor will show more drastic career stage differ- ences than others. To address these questions, this study examines the career stage differences in a comprehensive set of job factors.
In the following sections, an overview of career stage theories is provided and the hypotheses as to the different impacts of job factors across the various career stages are suggested. Then, these hypotheses are tested using the Federal Employee Viewpoint Survey, a large nationwide data set from the U.S. federal government. The results contribute to the literature by providing a comprehensive understanding of the natures and magnitudes of career stage difference, and also may give insight to practitioners to motivate their workforce more effectively.
Theory and Hypotheses
Career Stage Model
The career stage model proposed by Super (1980) provides insight for predicting how the antecedents to employee attitudes may change through different career stages. He proposed that employees in the organization go through four distinct stages during their tenure and that in each phase, employees have different career aspirations. The first stage refers to “exploration (or trial).” Employees in this stage explore for the possible match between their interests and competence and the current job’s condi- tions and requirements. The second phase refers to “establishment” in which individu- als seek to make advancements in their status in the organization. The third stage refers to the “maintenance” phase. In this phase, the main goal of individuals is to hold onto the accomplishment achieved earlier. The last phase is “disengagement” in which indi- viduals attempt to develop a new self-image independent of career success.
Similarly, Levinson (1978) proposed a process model of the life cycle. His model explains the different natures of psychological adjustment that people generally expe- rience as they grow into advanced phases of their lives. For instance, those in early adulthood (age of 20-40) show a psychological tendency to think deeply about their personal identity as an independent being, whereas those in middle adulthood (age of 40-60) focus on developing their values based on the review of what they have done in their 30s. Although Levinson’s model is about human development over the life
Lee 719
cycle in general, its relevance to career stage is evident to the extent that people’s dif- ferent psychological orientations inevitably make differences in their career aspira- tions. Indeed, Ornstein, Cron, and Slocum (1989) found that Levinson’s life cycle model is closely related to the career decisions of employees in an organization, such as turnover intention.
Based on these theoretical models, scholars have investigated how various factors pertinent to managerial practice or HR policies have differing effects on the attitudes of employees at different career stages. For example, Darcy and her colleagues (2012) examined how the antecedents to work–life conflict differ across career stages. Similar comparative research has been conducted with a focus on many other aspects of job attitudes as well (e.g., Bedeian et al., 1992; A. Cohen, 1991; Kooij et al., 2008; Miao et al., 2009). Despite a large volume of research, most of these studies tend to provide a limited understanding of career stage difference. Many of them present the differ- ences in the mean scores across different groups of career stages (e.g., McElroy et al., 1999; Weeks, Moore, McKinney, & Longenecker, 1999). Plus, most of them are focus- ing only on a small number of job factors (e.g., Darcy et al., 2012; Hogan et al., 2017; Yu, Lou, Eng, Yang, & Lee, 2018).
While research has shown that certain job factors vary in importance across differ- ent groups, that does not necessarily imply that those factors are no longer important for particular groups. Just because a certain job factor is less important for senior employees does not imply that organizations can entirely disregard that factor for motivating them. Such interpretations can be wrong given that the importance of the given factor may still be significant for the motivation of senior employees. This will not be clarified until the effect size of that factor is compared with those of other job elements. For example, even if the career stage significantly moderates the effect of pay on motivation, pay is the most important regardless of employees’ career stage if it exhibits a greater effect size than any other job factors in all career stage groups. To address this issue in the research on career stages, this study takes a comprehensive approach by demonstrating the career stage difference in terms of a multitude of job factors. Plotting the effect sizes of various factors for different career stage groups in a single graph will help readers easily compare the relative importance of these mul- tiple factors and hence take a comprehensive view of career stage differences.
Determinants of Employee Attitudes
There is a consensus among scholars that human motivation is driven by the desires of actors to satisfy their own needs. This assumption is not only shared by the early schol- ars but also stands out in more recent theories of motivation as well, such as self- determination theory (SDT) (Ryan & Deci, 2000b). The only difference of SDT compared with the traditional need theory of motivation is the type of needs that it emphasizes: SDT emphasizes the need for self-endorsement, rather than the need for existence or relatedness. Also, the idea underlying career stage differences is that peo- ple in different stages have different needs and desires. For instance, career stage dif- ference is manifested when the employees at the early stage of their career emphasize
720 Review of Public Personnel Administration 40(4)
the needs for growth, whereas those at the later stage emphasize the needs for related- ness. Thus, to discuss the career stage differences, identifying the types of human needs most appreciated in each of the career stages can be a good starting point. This study uses the following eight factors in relation to different types of needs.
Pay. The most frequently cited job element argued to influence work motivation is pay. Pay is essential for meeting basic human needs and also a primary reason that people work. The literature suggests that pay is a significant predictor of job attitudes of employees. Curral, Towler, Judge, and Kohn (2005) and Lum, Kervin, Clark, Reid, and Sirola (1998) found that pay satisfaction is negatively associated with turnover inten- tion. Given that the motivational strength of the certain type of needs languishes as the extent to which that need is currently satisfied increases (Maslow, 1943), the impor- tance of pay will be more important for junior employees because it is reasonable to assume that the pay level of junior employee is lower than that of the senior. Admit- tedly, one may argue that pay becomes more important as people get older and take more financial responsibilities (e.g., educating children). However, given the measure- ment of age in this study, the proposed hypothesis is more reasonable. The age group that is likely to be under the greatest financial needs are the lower categories measured in this study (i.e., those at the age below 40 and 40-49 compared with 50-59 or above 60). This is because people typically have children at the age of 30 who become inde- pendent at their age of 20, meaning that people get free from the financial burden to support their children around the age of 50. Besides, those younger respondents (i.e., below 40 and between 40 and 49) receive less amount of salaries. Thus, the younger groups will consider pay more significant compared with the older groups.
Hypothesis 1: The motivational importance of pay will be greater for junior than senior employees.
Workload. The second factor involves workload. Excessive workload is a well-known cause of mental distress and high blood pressure (Ilies, Dimotakis, & De Pater, 2010) and accounts for a significant portion of the variance in emotional exhaustion (Green- glass, Burke, & Fiksenbaum, 2001). Workload has relevance to work behaviors as well. Bowling, Alarcon, Bragg, and Hartman (2015) found that excessive workload increases the intention to quit and absenteeism. Workload will be of greater impor- tance for senior’s job satisfaction and turnover intention than for junior employees. The career goal of junior employees is usually to establish positive reputations in their workplaces by hardworking and excelling in their tasks (Super, 1980). Hence, junior employees may be willing to work overtime if needed. To the contrary, the goal of senior employees should be simply to maintain their status that they have already obtained, which would presumably take much fewer work efforts. Moreover, young employees are likely to be physically hardier than old seniors, as indicated in the find- ings of Guthrie, Tattan, Williams, Black, and Bacliocotti (1999) that senior employees are more likely to experience burnout from excessive job demands. Thus, the work- load issue will not be as significant for juniors’ motivation as for seniors.
Lee 721
Hypothesis 2: The motivational importance of workload will be lesser for junior than senior employees.
Relationships. The third and fourth factors involve the desire for a cooperative and friendly interpersonal relationship in the workplace. Baumeister and Leary (1995) found individuals have strong and constant desires to form and maintain deep, warm, and supportive interpersonal relationships, which provide the feelings of emotional bonds with each other. The literature on social support attests to the managerial impor- tance of satisfying this type of needs. Yoon, Baker, and Ko (1994) found that interper- sonal attachment that can be satisfied from informal relationships has a significant impact on organizational commitment. Babin and Boles (1996) also found that employee perception of coworker involvement and supervisory support can reduce work stress and increase job satisfaction. Given that the literature on social support in the workplace usually makes a distinction on the sources of social supports between colleagues and supervisors (e.g., Peeters & Le Blanc, 2001), this study tests the two relationship factors separately: relational support from colleagues and supervisors.
The importance of interpersonal relationships cannot be discounted for any age or tenure group. However, one can reasonably expect that maintaining a harmonious relationship will have greater importance for junior employees. This is because one of the biggest goals for those at the establishment phase is to obtain approval from col- leagues and supervisors and build positive reputations. Thus, the employees at the earlier stages will be more concerned about a potential interpersonal conflict that will undermine their reputation. To the contrary, senior employees must have already built their own intimate networks in the workplace that serve to buffer the aftermath of interpersonal conflicts or antagonistic office politics. Hence, they care relatively less about the relationship factors in the workplace.
Hypothesis 3: The motivational importance of relationships with colleagues will be greater for junior than senior employees. Hypothesis 4: The motivational importance of relationships with supervisors will be greater for junior than senior employees.
Intrinsic motivation. The fifth factor is related to intrinsic motivation. The enjoyment of the task is the most basic form of intrinsic motivator which has been found to have a strong influence on job attitudes and performance (Baard, Deci, & Ryan, 2004). More recently, the topic of person-job fit has appeared in the literature, which highlights the importance of intrinsic motivation. Warr and Inceoglu (2012) found the match between one’s personality and her job, which may be indicated by the enjoyment of the tasks, is significantly associated with her job engagement and satisfaction. June and Mah- mood (2011) also found that employees with a greater level of person–job fit perform better on their tasks.
Although the intrinsic factor is likely to be important for all age and tenure groups, it should be especially important in the early phase of career development. The earliest period of the career stage is the “exploration” phase whereby employees explore for
722 Review of Public Personnel Administration 40(4)
the fit between their personality or capacity and the requirement of their tasks (Super, 1980). In this phase, employees leave the organization if they fail to find their job interesting. To the contrary, those at the “maintenance” stage must have already adjusted to their tasks and developed the necessary skills. Also, unfit employees must have already screened out.
Hypothesis 5: The motivational importance of the intrinsic factor will be greater for junior than senior employees.
Career advancement opportunity. The sixth factor involves the perceived opportunity for promotion or career advancement as defined more broadly. Promotion is one of the strongest aspirations for employees in public organizations (Houston, 2000). Promo- tion is conducive to the fulfillment of one’s maximum potential, which is referred to as the growth needs in the literature (Alderfer, 1969). Getting promotion means that one will have greater job autonomy and greater opportunities for skill utilization, job diversification, and job enrichment (Harley, 1999; Porter, 1963). All of these by-prod- ucts of promotion will help increase one’s belief that their potential has been actual- ized to a greater degree (Hackman & Oldham, 1976).
This factor is expected to be more important for juniors than seniors. Admittedly, such difference may not be salient if comparing the employees at the early career stage with those at the mid-career stage. However, it is not unreasonable to expect that the strength of this factor will be significantly different between those at the earliest stage and those at the latest stage, given that the latest stage of employment is called the “disengagement” phase (Super, 1980). The priority of the employees in the disengage- ment stage is typically to plan their lives after retirement, and they are likely to have already made considerable achievements in their careers. Moreover, some in this stage may want not to get further promotions because the employees promoted too quickly have to worry about the early “peaking-out.”
Hypothesis 6: The motivational importance of career advancement opportunities will be greater for junior than senior employees.
Recognition. The seventh factor involves the recognition for good performance. Scholars have defined the nature of the relationship between organizations and their employees from the social exchange perspective (Ko & Hur, 2014), implying that employees expect that the organization will provide respect, rewards, and rec- ognition in return for their excellent work outcomes. The literature on psychologi- cal contract (Rousseau, 1989) is also predicated on the assumption that individuals believe in a reciprocal obligation between themselves and their organizations. Some may assume that monetary reward is the only aspired reward that matters for motivation. However, equally, if not more, important is nonmonetary rewards such as respect and approval from other members in an organization. Such intangible rewards increase intrinsic motivation, which has a greater influence on voluntary work effort (Weibel, Rost, & Osterloh, 2009). Many scholars have shown that
Lee 723
recognition matters for motivation and performance. DeCicco, Laschinger, and Kerr (2006) found that the respects that nurses receive in the workplace signifi- cantly predicted their commitment to their organizations. Although recognition for good work should be consistently important across career stages, its impact on job attitudes will be relatively greater for newcomers than senior employees. This is because senior employees will concern less about climbing up the career ladder by being recognized as a high performer (Super, 1980).
Hypothesis 7: The motivational importance of recognition for good performance will be greater for junior than senior employees.
Participatory decision-making. The eighth factor is the extent to which employees are allowed to participate in the decision-making that affects their work. Encouraging participation in the organizational decision process is conducive to the enhancement of self-concept and has been recognized as an important vehicle for increasing employ- ees’ intrinsic motivation (S. Kim, 2002). DeCicco and her colleagues (2006) found that organizational structures empowering employees explained a significant portion of the variance of their affective commitment. Laschinger, Finegan, and Shamian (2001) also found that nurses’ perception of empowerment significantly increases their work satisfaction.
Participation will have a greater impact on job attitudes of senior employees rather than those of juniors. This is because employees will not have a strong desire for pro- actively making influence over their work process until they gain sufficient experi- ence. The delegation of too much responsibility or discretion will leave newcomers confused and helpless. Instead, newcomers’ priority will be placed heavily on learning how to accomplish the tasks assigned to them as well as internalizing organizational cultures, norms, and atmosphere.
Hypothesis 8: The motivational importance of participation will be greater for senior than junior employees.
Positive and Negative Elements of Employee Attitudes
This study considers two important elements of employee attitudes as dependent vari- ables. The first element involves employees’ job satisfaction and the second element is their turnover intention. It is important to examine these two elements of employee attitudes simultaneously because job satisfaction is the positive aspect of employee attitude, whereas turnover intention reflects the negative aspect. The classical two- factor theory of Herzberg (1957) posited that these two are not at either end of a con- tinuum and that the antecedent conditions of the positive attitudes (e.g., job satisfaction) are different than those of the negative ones (e.g., turnover intention). This implies that managers should be prepared to both promote job satisfaction and alleviate turnover intention simultaneously. Thus, this study examines two models testing the factors for both types of job attitudes.
724 Review of Public Personnel Administration 40(4)
Method
Data and Sample
The data for this study are obtained from the Federal Employee Viewpoint Survey (FEVS), conducted annually by the U.S. Office of Personnel Management. This study used the FEVS collected in 2014. For the 2014 survey, a total of 839,788 federal employees were invited to the survey, and 392,752 respondents participated, which produced the response rate of 46.8%. An overview of the sample characteristics is provided in Table 1. The measurements for the variables in this study were selected from the pool of the FEVS questionnaire items. To calculate the variables, the sum- mated score is used.
Measurement
Dependent variables. The dependent variables in this study are twofold: job satisfac- tion and turnover intention. To measure employee’s job satisfaction, the questionnaire items related to the global aspect of satisfaction that has no specific referents to any specific job elements (e.g., pay, leadership), are selected. These items were “Consider- ing everything, how satisfied are you with your job?” “Considering everything, how satisfied are you with your organization?” “I recommend my organization as a good place to work.” This set of variables was used to measure job satisfaction in some prior studies (Cho & Lee, 2011; Cho & Perry, 2012). The reliability was .911. To measure the second dependent variable, turnover intention, the following single item was used. “Are you considering leaving your organization within the next year, and if so, why?” This question was answered by the following scale: A = No; B = Yes, to take another job within the Federal Government; C = Yes, to take another job outside the Federal Government; D = Yes, others. This variable is transformed into a binary variable (i.e., B, C, D = 1, A = 0).
Independent variables. The first independent variable is pay satisfaction. This vari- able is measured by the single question worded “Considering everything, how satis- fied are you with your pay?” The second independent variable was workload appropriateness. The single item measures this variable: “My workload is reason- able.” The third independent variable is cooperative relationships with colleagues. The following two items measured this variable: “The people I work with cooperate to get the job done,” “Employees in my work unit share job knowledge with each other.” Prior studies used this set of variables to measure a similar construct such as coworker relationships (Jung & Lee, 2015; Lewis & Pitts, 2018). The reliability score was .749. The fourth independent variable was the quality of the interpersonal relationship with supervisors. This construct is measured using the following three items: “My supervisor listens to what I have to say,” “My supervisor treats me with respect,” “I have trust and confidence in my supervisor.” These items were used in prior studies to measure a concept that is the same but labeled slightly differently as supervisory support (Harrington & Lee, 2015). The reliability was .938. The fifth
Lee 725
independent variable involves intrinsic motivation. Intrinsic motivation refers to the enjoyment of their tasks, and more broadly, the resultant feelings of self-esteem (Ryan & Deci, 2000a). While the core element of intrinsic motivators is the degree to which employees like the kind of work they do, enhanced self-esteem as a result of successful task accomplishment has been considered as a key mechanism explain- ing why people work without extrinsic reward. Hence, this variable is measured by the following three items: “My work gives me a feeling of personal accomplish- ment,” “The work I do is important,” “I like the kind of work I do.” The same items were used for intrinsic motivation in a prior study (Cho & Perry, 2012). The reli- ability was .808. The sixth independent variable involves the perceived opportunity for promotion and career advancement. The following two items measure this: “How satisfied are you with your opportunity to get a better job in your organization?” “Promotions in my work unit are based on merit.” The first item includes the word- ing of “get a better job,” which may mean being promoted to upper-level positions or being deployed or transferred to the departments they see more advantageous for promotion or suitable for self-development. This item was used in a prior study to measure career advancement (Cho & Sai, 2013). The second item captures the con- cept more indirectly, asking whether the promotion decision is based on merit. How- ever, given that merit-based promotion is generally perceived as the absence of a discriminatory barrier, this is closely related to employees’ positive expectation for career advancement. The reliability was .776. The seventh independent variable is recognition for good work outcomes. The following four items measure this con- struct: “In my work unit, differences in performance are recognized in a meaningful way,” “Employees are recognized for providing high-quality products and services,” “Creativity and innovation are rewarded,” “How satisfied are you with the recogni- tion you receive for doing a good job?” The reliability was .904. The last indepen- dent variable is participation. The following two items measure this: “How satisfied are you with your involvement in decisions that affect your work?” “Employees have a feeling of personal empowerment with respect to work processes.” This set of items was used in a prior study to measure participation (C. Kim & Schachter, 2015). The reliability for this variable was .792.
Table 1. Sample Characteristics.
Category % Category %
Gender Male 51.3 Supervisory status Senior Executive Service 1.4 Female 48.7 Supervisor 15.4
Minority Non-White 32.3 Nonsupervisor 83.2 Age Below 30 4.2 Tenure (years) Below 3 5.3
30-39 17.2 4-5 13.7 40-49 27.2 6-10 19.9 50-59 36.2 11-20 24.7 Above 60 15.2 Above 20 36.35
726 Review of Public Personnel Administration 40(4)
Control variables. Some control variables related to the demographic characteristics of respondents were included. First, gender is generally believed to affect employees’ job attitudes. The literature suggests that women are generally more satisfied with their workplaces (Clark, 1997). Also, women are different in motivating factors from men (Kovach, 1987). Thus, the effect of gender should be statistically controlled. Gender was measured by a binary variable (male = 0, female = 1). Second, the level of educa- tion is included as a control. Employees with higher education are more satisfied in organizations (Lee & Wilbur, 1985). Also, because the pay is generally correlated with education level, the variable of education may interact with the effect of that variable to have some confounding impact on work attitudes. The level of education was mea- sured as an ordinal variable (education prior to a bachelor’s degree = 1, bachelor’s degree = 2, post-bachelor’s degree = 3). Third, the attitudes of the supervisor should be determined by different factors than nonsupervisors (Kovach, 1987). Thus, the vari- able indicating supervisory status should be included to control its effects. This vari- able is measured as a binary variable (nonsupervisor/team leaders = 0, supervisor/ manager/executives = 1). Fourth, ethnicity may be an interfering factor. The literature suggests that the formation of work attitudes may be determined by cultural elements (Kirkman & Shapiro, 2001). A result is that in some cultures, certain aspects of job conditions are more emphasized than others. Thus, the variable indicating the ethnic majority/minority distinction will be included as a control variable. Finally, those with a plan to retire should automatically have a greater turnover intention. As this study focuses on turnover for other reasons than retirement, the binary variable indicating the plan to retire within 5 years will be included as a control variable.
Career stage variables. The analyses of this study are conducted for different subsam- ples, each of which represents the unique career stage. To divide the subsamples, this study uses two indicators pertinent to employees’ career stages: age (Levinson, 1978) and tenure (Super, 1980). To measure age, this study uses a variable that divides into four different age groups: A = below 40, B = 40 to 49, C = 50 to 59, D = above 60. Levinson proposed a life cycle model comprising two stages and eight substages. The first stage starts from 20 in age and lasts until 40 and the second stage spans from 40 to 60 with a 5-year interval. Given that an average adult starts her career in her mid- 20s, the first stage covers from the age of 25 to 40 (A) and is the phase in which indi- viduals find the workplace that best fits their needs and competencies and decide to settle in their career field. The next four substages are included in the next two levels of the age variable (i.e., B and C): the age range in Levinson’s model was 40 to 45, 46 to 50, 51 to 55, and 56 to 60. Thus, the divisions proposed in the theory are generally comparable with the subgroups that are obtained by the current age variable.
To measure tenure, this study uses a division into the following three different groups: A = less than 5 years, B = 6 to 14 years, C = more than 15 years. Theoretically, the career stages divided by tenure are as follows (Super, 1980): the trial stage generally refers to the first 2 years on the job; the establishment stage is between 2 and 10 years on the job; the maintenance stage is more than 10 years on the job. Given that this distinction is made in light of private sector jobs, in which
Lee 727
job security is minimal, and turnover is more common than public employees, their time span associated with the career stage will be shorter than those of public employees. Thus, the breakpoints made by the current tenure variable are reasonable in the context of federal employees.
The measurements for the latent variables in this study are validated. First of all, to examine whether the latent variables are measured with a reasonable degree of validity, the confirmatory factor analysis (CFA) is conducted, and it is found that all fit indices were acceptable.1 Second, the composite reliability (CR) and average vari- ance extracted (AVE) for each variable also were greater than the rule of thumb threshold points of .7 and .5, respectively.2 Based on the results of these preliminary statistical examinations, it can be concluded that the measurements in this study are reasonably valid.
Analytic Strategy
The main purpose of this study is to compare the relative importance of the previously discussed independent variables for job satisfaction and turnover intention across fed- eral employees at different stages of their careers. To this end, first, the respondents of the FEVS 2014 were divided into different subgroups indicating distinct career stages. As there are two legitimate indicators of career stages, namely, age and tenure, this study used both indicators, separately as well as simultaneously. As a preliminary examination of the career stage difference, a series of Chow tests were conducted on each independent variable, which shows if there is an overall significant difference among different subgroups. This preliminary step serves as a rationale for taking a closer look at how the career stage differences are manifested for each group and each dependent variable.
Second, a series of ordinary least squares (OLS) regressions on job satisfaction are conducted for different subgroups. The results are compared between (a) the subgroups of different ages, (b) the subgroups of different tenures, and (c) the sub- groups divided by combinations of the two indicators of career stage. For this last comparison, the two subgroups placed at both ends on the continuum of career stages in terms of both age and tenure (i.e., between the youngest with the least ten- ure and the oldest with the longest tenure) are selected. This analysis allows the amplification of the career stage effects. Also, another comparison is conducted between (d) young veterans (below 40 with more than 15 years of tenure), and older newcomers (above 60 with less than 5 years of experience). This comparison makes the analyses of this study complete, by covering a wide range of career stages. The same procedure is also undertaken to analyze turnover intention using the binary logit analysis.
Third, to visualize the differences among the subgroups, the graphs showing the effect sizes of each factor for job satisfaction and turnover intention will be plotted for each subgroup. Many statisticians have pointed out that the effect size conveys as important information about the significance of the causal effect as the statistical significance of the coefficients (Greenwald, Gonzalez, Harris, & Guthrie, 1996).
728 Review of Public Personnel Administration 40(4)
Reporting the effect sizes is especially meaningful for the studies using the FEVS. As the sample size of the FEVS is extraordinarily large (more than 300,000), the statisti- cal significance of regression coefficients will be meaningless: the t values for most of the coefficients are significant (J. Cohen, 1992). This is why many studies using the FEVS data present the rankings of coefficients (Trottier, Van Wart, & Wang, 2008) or marginal effects (Rubin, 2007; Wynen, Op de Beeck, & Hondeghem, 2013). Furthermore, the purpose of this study is to compare the relative importance of the independent variables. Hence, relying on the test of statistical significance would not be sufficient. Instead, this study focuses on plotting the effect sizes indicated by the semi-partial correlation (sr2) coefficient for the OLS results (job satisfaction) and the standardized coefficients for the logistic analysis (turnover intention). Tabachnick and Fidell (2007) suggested that the semi-partial correlation (sr2) coefficient repre- sents the unique contribution of each independent variable to the variance of a depen- dent variable and hence are the most proper indicators for the relative importance of independent variables in OLS model. They also proposed that standardized coeffi- cients (beta) in the binary logit analysis can be used as an alternative to the semi- partial correlation in OLS.
Results
As explained previously, this study runs a series of regression analyses for different subsamples. Before that, the Chow tests are conducted to examine the significance of overall subgroup differences as per each independent variable. As for the OLS regressions on job satisfaction, most independent variables, except for participation when grouped by a combination of age and tenure, showed significant group differ- ences in its effect on job satisfaction. As for the logistic analysis on turnover inten- tion, each of the independent variables showed significant group differences in its effect on turnover intention, which gives sufficient ground for examining the career stage differences in further detail.3
The results of OLS regressions on job satisfaction are summarized in Table 2. The model fit indices show that the proposed models explained considerable variances of job satisfaction (approximately 75%), indicating that the proposed model contains a reasonably comprehensive set of factors. The columns in the table show the standard- ized regression coefficients for each group. The first four columns in Table 2 present the coefficients for the subsamples divided by age. Examining the semi-partial corre- lation coefficients, intrinsic motivation and participation had the strongest impacts on job satisfaction, whereas the relationship with colleagues was the least important fac- tor. The next three columns in Table 2 indicate the coefficients for subgroups by ten- ure. The results are not very different from the findings from the comparison by age. The effect size for most of the variables related to the growth need (intrinsic factor and participation) was generally more important than those related to other types of needs. Overall, the results were similar to the comparison between the subsamples divided by age. However, the rise of the importance of recognition is somewhat noticeable. The last two columns in Table 2 indicate the coefficients for the subsamples divided by a combination of age and tenure.
729
T a b
le 2
. O
rd in
ar y
Le as
t Sq
ua re
s R
eg re
ss io
ns o
n Jo
b Sa
ti sf
ac ti o n.
V ar
ia bl
es
G ro
up in
g va
ri ab
le s
A ge
T en
ur e
A ge
a nd
t en
ur e
B el
o w
4 0
40 t
o 4
9 50
t o 5
9 A
bo ve
6 0
Le ss
t ha
n 5
5 to
1 4
M o re
t ha
n 15
B el
o w
4 0
w it h
le ss
t ha
n 5
ye ar
s A
bo ve
6 0
w it h
m o re
t ha
n 15
y ea
rs
Pa y
sa ti sf
ac ti o n
.0 92
** *
.0 86
** *
.1 00
** *
.1 00
** *
.0 93
** *
.0 92
** *
.0 96
** *
.0 96
** *
.0 99
** *
W o rk
lo ad
.0 77
** *
.0 69
** *
.0 72
** *
.0 67
** *
.0 80
** *
.0 70
** *
.0 67
** *
.0 80
** *
.0 68
** *
C o lle
ag ue
s .0
47 **
* .0
46 **
* .0
45 **
* .0
53 **
* .0
58 **
* .0
51 **
* .0
39 **
* .0
51 **
* .0
50 **
* Su
pe rv
is o rs
.0 79
** *
.0 84
** *
.0 82
** *
.0 87
** *
.0 91
** *
.0 83
** *
.0 74
** *
.0 85
** *
.0 78
** *
In tr
in si
c m
o ti va
ti o n
.2 55
** *
.2 50
** *
.2 42
** *
.2 37
** *
.2 67
** *
.2 49
** *
.2 42
** *
.2 71
** *
.2 35
** *
R ec
o gn
it io
n .1
27 **
* .1
17 **
* .1
21 **
* .1
35 **
* .1
22 **
* .1
18 **
* .1
31 **
* .1
19 **
* .1
38 **
* C
ar ee
r ad
va nc
em en
t .1
45 **
* .1
49 **
* .1
35 **
* .1
29 **
* .1
31 **
* .1
44 **
* .1
40 **
* .1
43 **
* .1
31 **
* Pa
rt ic
ip at
io n
.2 97
** *
.3 13
** *
.3 16
** *
.3 06
** *
.2 84
** *
.3 13
** *
.3 17
** *
.2 78
** *
.3 10
** *
M o de
l f it
A
dj us
te d
R 2
.7 50
.7 58
.7 50
.7 46
.7 47
.7 55
.7 52
.7 43
.7 45
F
9, 32
5. 43
1* **
12 ,6
20 .2
89 **
* 16
,0 71
.7 31
** *
6, 11
2. 45
0* **
9, 10
5. 98
0* **
15 ,4
39 .6
86 **
* 19
,4 41
.1 77
** *
4, 06
6. 98
3* **
4, 01
2. 70
3* **
N
40 ,3
58 52
,2 91
69 ,5
25 40
,0 96
43
,1 34
65 ,1
25 83
,4 00
18 ,3
28 17
,8 56
N ot
e. T
he n
um be
rs in
t he
c el
ls d
en o te
t he
s ta
nd ar
di ze
d co
ef fic
ie nt
s (B
et a)
. T he
c o ef
fic ie
nt s
fo r
co nt
ro l v
ar ia
bl es
w ill
b e
pr o vi
de d
up o n
re qu
es t.
*p <
.1 0.
* *p
< .0
5. *
** p <
.0 1.
730 Review of Public Personnel Administration 40(4)
Figure 1 visualizes the differences in the effect sizes for the job satisfaction factors for the subgroups. The y-axis indicates the effect size (i.e., semi-partial correlation) of each variable. The Graph c shows a clearer pattern of career stage effect than the Graph a or b. Intersecting lines indicate that the relative importance among the factors has changed. Intersections occurred for several pairs of factors: between career advancement and pay, between recognition and workload, and between recognition and supervisor relationship. The importance of participation for the oldest group with the longest tenure has increased compared with that of the youngest group with the least tenure, whereas the importance of intrinsic motivation has declined although the relative importance for these two factors did not change.
Table 3 presents the results of binary logistic regression analyses on turnover inten- tion. The first four columns in the table indicate the odds ratios for the groups of dif- ferent ages. The two pseudo-R2 measures of model fit indicate that the explanatory power of the models gradually diminishes as going to the next stages of career (age). The first model representing those employees younger than 40 showed the Cox and Snell R2 of .248, whereas the subsample of those older than 60 showed only .108. This decline shows that senior employees’ turnover intention is affected less by the factors included in the model.
The next three columns in Table 3 indicate the odds ratios for the subsamples divided by tenure. Unlike the previous comparison by age, the importance of the pros- pect for career advancement remains the most important even for those at the latest stage of career development.
The last two columns in Table 3 show the odds ratios for the subgroups divided by a combination of age and tenure. This comparison also shows that the R2 decreased from .260 (the youngest group with the least tenure) to .103 (the oldest group with the greatest tenure). The pattern of change in the effect sizes of the individual coefficients was generally similar to what was found in the comparisons by age or tenure consid- ered separately.
Figure 2 visualizes the relative strengths of the factors for turnover intention for the subgroups. One may be overwhelmed by the steeper slope of the lines than those observed in the models of job satisfaction factors and conclude that the changes in the relative importance are greater for turnover intention. However, attending to the slopes may be misleading because the R2 values have decreased when comparing those below 40 with those above 60. It may be that the effect sizes of the individual factors may have decreased naturally as a result of the decrease in the overall explanatory power of the models.
More noteworthy is the increased proportion of the factors intersecting each other, compared with those in the model of job satisfaction. For instance, the pros- pect for career advancement was the greatest suppressor of turnover intention in the early stage of career. However, it moves down to the third place for seniors, and instead, participation becomes more important. Also, pay was more critical for suppressing turnover intention in the early stages. In contrast, relationship with supervisor becomes a more important suppressor of turnover intention for seniors than pay.
Lee 731
Some notable differences compared with the result of job satisfaction analyses are observed. First, the importance of recognition for turnover intention was insignificant for most groups. Although the degree of recognition for good performance was an important condition for job satisfaction, it does not specifically suppress employees’ turnover intention. Second, the importance of pay as a suppressor of turnover intention has significantly increased, except for the oldest group, meaning that although pay may not be a factor of overall job satisfaction, the lack of pay satisfaction results in a turnover, especially for younger groups.
Additional analyses were conducted to compare between older newcomers (above 60 in age and with less than 5 years of tenure) and young veterans (below
Figure 1. Comparing the squared semi-partial correlation coefficients for the determinants of job satisfaction: (a) age groups, (b) tenure groups, and (c) age and tenure groups.
732
T a b
le 3
. Lo
gi st
ic R
eg re
ss io
n o n
T ur
no ve
r In
te nt
io n.
V ar
ia bl
es
G ro
up in
g va
ri ab
le s
A ge
T en
ur e
A ge
a nd
t en
ur e
B el
o w
4 0
40 t
o 4
9 50
t o 5
9 A
bo ve
6 0
Le ss
t ha
n 5
ye ar
s 5
to 1
4 ye
ar s
M o re
t ha
n 15
y ea
rs B
el o w
4 0
w it h
le ss
t ha
n 5
ye ar
s A
bo ve
6 0
w it h
m o re
t ha
n 15
y ea
rs
Pa y
sa ti sf
ac ti o n
.7 94
** *
.8 24
** *
.8 85
** *
.9 70
** .8
03 **
* .8
42 **
* .9
26 **
* .7
90 **
* .9
85 W
o rk
lo ad
.8 88
** *
.8 91
** *
.9 16
** *
.9 14
** *
.8 99
** *
.8 79
** *
.9 15
** *
.8 78
** *
.9 38
** *
C o lle
ag ue
s .9
33 **
* .9
48 **
* .9
38 **
* .9
98 .9
40 **
* .9
55 **
* .9
57 **
* .9
12 **
* .9
93 Su
pe rv
is o rs
.9 09
** *
.8 92
** *
.8 78
** *
.8 90
** *
.9 15
** *
.8 95
** *
.8 85
** *
.9 18
** *
.9 06
** *
In tr
in si
c m
o ti va
ti o n
.6 33
** *
.7 00
** *
.7 48
** *
.7 82
** *
.5 81
** *
.6 71
** *
.7 81
** *
.5 77
** *
.7 99
** *
R ec
o gn
it io
n 1.
02 7
1. 02
7 1.
03 7*
* .9
94 1.
01 3
1. 03
5* *
1. 00
7 1.
04 1
1. 01
2 C
ar ee
r ad
va nc
em en
t .5
80 **
* .5
91 **
* .6
45 **
* .8
44 **
* .5
70 **
* .5
97 **
* .7
21 **
* .5
78 **
* .8
90 **
* Pa
rt ic
ip at
io n
.8 20
** *
.8 56
** *
.9 05
** *
.8 82
** *
.8 22
** *
.8 44
** *
.9 09
** *
.8 01
** *
.8 72
** *
N 40
,3 58
52 ,2
91 69
,5 25
40 ,0
96 43
,1 34
65 ,1
25 83
,4 00
18 ,3
28 17
,8 56
M o de
l f it
–2
L o g
lik el
ih o o d
41 ,7
52 .7
37 52
,5 88
.6 17
72 ,7
34 .4
17 32
,3 62
.6 27
40 ,8
39 .9
94 68
,1 27
.6 68
89 ,9
60 .2
62 18
,7 35
.4 30
21 ,9
49 .5
05
C o x
an d
Sn el
l R 2
.2 48
.2 04
.1 47
.1 08
.2 43
.2 09
.1 29
.2 60
.1 03
N
ag el
ke rk
e R
2 .3
39 .2
87 .2
10 .1
48 .3
34 .2
89 .1
84 .3
54 .1
39
N ot
e. T
he n
um be
rs in
t he
c el
ls d
en o te
o dd
s ra
ti o s.
T he
c o ef
fic ie
nt s
fo r
th e
co nt
ro l v
ar ia
bl es
w ill
b e
pr o vi
de d
up o n
re qu
es t.
* p <
.1 0.
* *p
< .0
5. *
** p <
.0 1.
Lee 733
40 in age and with more than 15 years of tenure). The results of the regression analysis on job satisfaction are provided in the first two columns in Table 4. The two groups showed significant differences in terms of many factors including rec- ognition, workload, relationship with colleagues, supervisor relationships, and par- ticipation. Older newcomers place significantly less emphasis on participation, recognition, and workload than young veterans, whereas they put greater emphasis on the relationship factors.
The last two columns in Table 4 show the binary logistic analysis on turnover inten- tion for the two unique groups. Intrinsic factor (enjoyment of tasks) had a much greater effect on the intention of older newcomers to stay in their current organizations. Also, older newcomers tend to stay in organizations when they are satisfied with the per- ceived opportunities for career advancement, although the effect of career advance- ment on turnover intention was greater for young veterans.
Figure 2. Comparing the standardized coefficients for the determinants of turnover intention: (a) age groups, (b) tenure groups, and (c) age and tenure groups.
734 Review of Public Personnel Administration 40(4)
Discussion and Conclusion
Theoretical Implications
An overview. The career stage is found to significantly explain the difference in the determinants of job satisfaction and turnover intention. To name a few overall find- ings, first, the career stage differences were generally more salient for turnover inten- tion than job satisfaction, meaning that the main reasons for turnover are more diverse depending on their career stages, whereas the reasons for job satisfaction are relatively similar regardless of the different career stages. Second, the senior employees’ turn- over is not dependent on the factors included in the model, as indicated by the gradu- ally decreasing pseudo-R2. This result seems to show that the reasons for the turnover of these senior employees are not related to the factors regarding work or organization. It can be speculated that the senior employees’ turnover may be largely dependent on factors related to their personal life or age-related factors (i.e., health issues). Third,
Table 4. Comparison Between the Old Newcomers and Young Veterans.
Variables
DV: Job satisfaction (Beta)
DV: Turnover intention (Odds ratios)
Below 40 (age) More than 15
(tenure)
Above 60 (age) Less than 5
(tenure)
Below 40 (age) More than 15
(tenure)
Above 60 (age) Less than 5
(tenure)
Pay satisfaction .103*** .104*** .868*** .831*** Workload .072*** .059*** .915* .862*** Colleagues .043*** .071*** .930 .871* Supervisors .052*** .114*** .878* .925 Intrinsic motivation .234*** .235*** .676*** .883 Recognition .153*** .144*** 1.001 1.006 Career advancement .104*** .112*** .658*** .624*** Participation .349*** .284*** .854* .849*
N 1,738 2,336 1,738 2,336
Model fit
Ordinary least squares
Logistic Below 40 (age) More than 15
(tenure)
Above 60 (age) Less than 5
(tenure)
Below 40 (age) More than 15
(tenure)
Above 60 (age) Less than 5
(tenure)
R2 −2 Log likelihood
.758 .730 2,412.125 1,835.763
F value Cox and Snell R2
418.677*** 486.300*** .166 .188
Nagelkerke R2 .236 .262
*p < .10. **p < .05. ***p < .01.
Lee 735
the relative importance of many factors remains unchanged even though the career stage differences are noticeable. For instance, the importance of intrinsic motivation for job satisfaction is greater for juniors than seniors but remains the most important even for seniors. This result shows that the findings of a significant moderating effect do not warrant the ignorance of that factor for some groups.
Hypothesis testing. Hypothesis 1 suggested that the importance of pay will be greater for junior employees than seniors. This hypothesis is only partially supported: The effects of pay satisfaction alleviating turnover intention was greater for juniors than seniors, indicating that in their decisions to leave their organizations, junior employ- ees give greater weight to pay than seniors do. To the contrary, pay significantly affects the job satisfaction level of seniors, whereas its effect on junior employee’s job satisfaction was less so. These two observations imply that both junior and senior employees consider the amount of salaries important, but they do so for different reasons. For junior employees, pay is closely associated with meeting their basic existential needs. Young employees tend to receive a smaller amount of salaries than seniors, while they have greater financial demands (e.g., for educating their children). Thus, they decide to leave their current organizations and to seek a better paying job when they are dissatisfied with their salaries. To the contrary, for senior employees, pay seems to be perceived as an indication of the prestige or status in the workplace (Dyer & Parker, 1975; Melguizo & Strober, 2007). Therefore, although pay satisfac- tion significantly increases their overall job satisfaction, they are not as desperate for money as juniors. Hence, senior employees do not decide to leave their organizations for a financial reason.
Hypothesis 2 predicted that the motivational importance of workload would be greater for senior than junior. However, the result shows otherwise: The effect sizes of the workload variable on both job satisfaction and turnover intention were greater for junior than senior. This may be because junior employees are more likely to suffer excessive job demands on the top of greater family-related responsibility (e.g., taking care of children) than senior employees. A post hoc examination of the mean differ- ence in the felt workload between junior (3.47) and senior employees (3.30) is statisti- cally significant (t = 26.052, p = .000), which supports this speculation. Such conditions of junior employees may have made them consider the workload issue more important than senior employees do.
Hypotheses 3 and 4 predicted that interpersonal relationships in the workplace would affect job attitudes of junior employees to a greater degree than those of seniors. However, this hypothesis is only partially supported: The importance of the relation- ship with colleagues for job satisfaction was greater for those with shorter tenure than those with longer tenure, yet the relationship with supervisors affected senior employ- ees’ turnover intention to a greater extent. The following can explain this inconsis- tency. For those with shorter tenures, their colleagues, presumably those at similar hierarchical rank, are not only their collaborators but also competitors. Thus, their action, whether benevolent or hostile, may significantly affect the status of a focal person within an organization. And more importantly, they do not have the authority
736 Review of Public Personnel Administration 40(4)
to correct their colleagues’ actions to which they are vulnerable. To the contrary, for senior employees, those whom they perceive as a colleague—“the people they work with”—are likely their subordinates working in the work unit they are leading. Hence, they have a greater influential power on the people they work with. This may be why senior employees are less concerned about whether they cooperate. However, the neg- ative impact of the experience of interpersonal conflicts with supervisors is much more severe for senior employees, so severe that they may want to leave the organiza- tion. This is because a higher status of the senior employee makes them feel greater damage to her pride when, for example, she is scolded by her supervisor in front of her subordinates.
Hypothesis 5 proposed that intrinsic factors are more important for junior than senior. This hypothesis is supported (see Figure 1), indicating that junior employees, typically in the “exploration” phase, are concerned about the person–job fit to a stron- ger degree than senior employees. However, this does not imply that the enjoyment of tasks is unimportant for senior employees: It is still important for seniors albeit to a lesser degree (see Figure 2).
Hypothesis 6 suggested that career advancement opportunity will have a greater impact on the job attitudes of junior employees than seniors. This hypothesis is also supported: Its effect sizes were consistently higher for junior groups. These results imply that, as hypothesized, senior employees have fewer desires for getting a pro- motion, because they already achieved career success in an organization, or they do not believe that they will have an opportunity to advance their careers any longer. Also, as Super (1980) argued, senior employees’ priority is to plan their life after retirement.
This study predicted that the recognition would have considerable impacts both on job satisfaction and turnover intention because it is an important element of the intrinsic type of motivation (Ryan & Deci, 2000a). However, the importance of recognition was only the fifth greatest factor at best for explaining job satisfaction. Furthermore, this factor was generally less important when it comes to turnover intention. To the contrary, other intrinsic factors such as enjoyment of the tasks or participative management were much more important than recognition. Also, the career stage difference of recognition also contradicts with the proposed hypothe- sis. Hypothesis 7 proposed that recognition for good work would have a greater impact on the job attitudes of junior employees than seniors. However, the career stage difference of this variable was minimal: It has a similar level of importance for employees’ job satisfaction and turnover intention regardless of their career stages. Taking all these together, the results seem to indicate that public employees do not appreciate the opportunity for getting recognition from external parties so much as for enhanced self-evaluation by making meaningful impacts to policy pro- cesses, or dedicating to public values (Perry & Wise, 1990), regardless of their career stages.
Hypothesis 8 suggested that the impact of participation will be greater for seniors. This hypothesis is also supported: The effect size of this variable was generally greater for senior employees than for juniors. This pattern is observed for both job satisfaction
Lee 737
and turnover intention. Newcomers do not likely have sufficient knowledge and skills to make a meaningful contribution to collective decision-makings, and thus, encourag- ing participation from junior employees may only increase their workload (Appelbaum & Batt, 1994; Osterman, 2000). Hence, a directive, rather than participative leadership is more appropriate for junior employees (Hersey & Blanchard, 1969). To the contrary, encouraging participation from senior employees will enhance their perception of self- worth and their intrinsic motivation.
The motivation of unique groups. This study also examined the models of job satisfac- tion and turnover intention for a number of unique groups, namely, older newcomers (those who started their careers in the federal agency later than others) and young veterans (those who started their federal careers earlier than others). These people constitute rather unique groups in an organization: the numbers of these federal employees were relatively small: 1,738 and 2,336, respectively. As they showed a salient difference in their motivational tendencies, these groups in organizations may need to be managed differently.
When compared with standard senior employees (above 60), older newcomers put greater emphasis on the relational factors. The importance of supervisor relationships for job satisfaction is greater for them. The importance of the relationship with col- leagues for turnover decisions was also higher than ordinary senior employees. They seem to be concern more about whether they can assimilate into organizational culture and get along well with the existing members, which make them prioritize relationship factors. Also, career advancement is the most important factor for suppressing their turnover intention, which may be striking given their age. The intrinsic factor is also essential for them, compared with ordinary senior employees (above 60). In contrast, when compared with the young newcomers, both their job satisfaction and turnover intention are determined by the factors generally similar, except for the effect of career advancement on job satisfaction being less important and the effect of the relational factors being more important.
Young veterans tend to put greater emphasis on participation and recognition for job satisfaction, compared with ordinary young employees (below 40 of age). The effect of supervisor relationship on job satisfaction was less significant for young vet- erans. When compared with employees with similar tenure, they place less emphasis on career advancement opportunities. This may be because longer tenure means that they are at higher ranks in the hierarchy. Hence, they are likely to have already achieved career success and worry about the early “peaking-out.” Comparing the determinants of turnover intention with the same reference group (more than 15 years of tenure), pay satisfaction shows a greater effect on their turnover intention. This may indicate that age, not tenure, determines the importance of pay.
Limitations and Future Research
Despite its significant implications, this study is not free from a number of weak- nesses. Future researchers may have to address these shortcomings when
738 Review of Public Personnel Administration 40(4)
examining similar research questions. First, the different characteristics of the agencies have not been taken into account. It is reasonable to speculate that the fac- tors of job satisfaction and turnover intention may differ across the different federal agencies because the work environments and organizational situations (e.g., possi- bility for promotion) are quite different from one another. For instance, the agen- cies, where employees are exposed to a higher level of risk, such as the Federal Bureau of Investigation (FBI), will show a greater effect of workload on employee attitudes than those in the agencies of a safer work environment. This is because, even if the workload is increased by the same degree, the aftermath will be greater for employees when working in a riskier condition. It may be worthwhile for future researchers to examine how the factors of job satisfaction and turnover intention differ in these organizations.
Second, the data used in this study were collected at a single time point, namely, in the year 2014. The cross-sectional nature of the data makes it difficult to tell whether the results of this study reflect the patterns of career stage difference that can be gen- eralized to most employees in most organizations. One cannot preclude the possibility that the observed patterns simply reflect the unique tendency of the employees in this generation. Hence, it would be worthwhile for future researchers to cross-validate the results of this study after several decades if the federal government continues to pro- duce the same nationwide data set.
Third, this study focused on the antecedent to general turnover intention but did not provide an examination of the antecedents to the intention to seek a job outside the federal government. An analysis of the FEVS 2014 (with the whole sample) reveals that the biggest antecedents of leaving the federal government are related to pay and recognition for good performance, while the effects of these two variables on the intention to seek another job within the government are expected to be minimal. However, this study did not specifically examine the career stage differences regard- ing these two different types of turnover. It may be worthwhile for future researchers to address this issue more thoroughly.
Conclusion
In conclusion, this article provides a comprehensive understanding of the career stage differences. This article demonstrates that career stage differences are manifested more drastically for certain job factors than others. Also, the career stage differences are more salient in relation to turnover intention than job satisfaction. Managers and the executives in public organization can refer to the summary of the results presented in this study if they want to tailor their job/organizational conditions to better fit the needs of employees at different career stages.
Acknowledgments
This paper was written during the author’s sabbatical leave spent at the School of Public Policy & Administration, York University. The assistance provided by the faculty and staff of the school is gratefully acknowledged.
Lee 739
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.
Notes
1. The result of confirmatory factor analysis (CFA) showed that all the fit indices except for chi-square were acceptable judging from the rule of thumb threshold points (chi-square/df = 1,135.953, standardized root mean square residua [SRMR] = .034, goodness of fit index [GFI] = .926, normed fit index [NFI] = .957, relative fit index [RFI] = .943, incremental fit index [IFI] = .956, Tucker–Lewis index [TLI] = .943, comparative fit index [CFI] = .956, root mean square error of approximation [RMSEA] = .072).
2. For job satisfaction (dependent variable), CR was .912, and AVE was .775. For participa- tion, CR was .792, whereas AVE was .656. For intrinsic motivation, CR was .812, whereas AVE was .594. For recognition, CR was .907, whereas AVE was .709. For coworker rela- tionship, CR was .750, whereas AVE was .600. For supervisor relationship, CR was .943, whereas AVE was .845.
3. The results of Chow tests will be provided upon request ([email protected]).
ORCID iD
Hyung-Woo Lee https://orcid.org/0000-0001-8520-5088
References
Alderfer, C. P. (1969). An empirical test of a new theory of human needs. Organizational Behavior and Human Performance, 4, 142-175.
Appelbaum, E., & Batt, R. (1994). The new American workplace: Transforming work systems in the United States. Ithaca, NY: Cornell University Press.
Baard, P. P., Deci, E. L., & Ryan, R. M. (2004). Intrinsic need satisfaction: A motivational basis of performance and well-being in two work settings. Journal of Applied Social Psychology, 34, 2045-2068.
Babin, B. J., & Boles, J. S. (1996). The effects of perceived co-worker involvement and super- visor support on service provider role stress, performance and job satisfaction. Journal of Retailing, 72, 57-75.
Baumeister, R. F., & Leary, M. R. (1995). The need to belong: Desire for interpersonal attach- ments as a fundamental human motivation. Psychological Bulletin, 117, 497-529.
Bedeian, A. G., Ferris, G. R., & Kacmar, K. M. (1992). Age, tenure, and job satisfaction: A tale of two perspectives. Journal of Vocational Behavior, 40, 33-48.
Bowling, N. A., Alarcon, G. M., Bragg, C. B., & Hartman, M. J. (2015). A meta-analytic exami- nation of the potential correlates and consequences of workload. Work & Stress, 29, 95-113.
Cho, Y. J., & Lee, J. W. (2011). Perceived trustworthiness of supervisors, employee satisfaction and cooperation. Public Management Review, 13, 941-965.
740 Review of Public Personnel Administration 40(4)
Cho, Y. J., & Perry, J. L. (2012). Intrinsic motivation and employee attitudes: Role of manage- rial trustworthiness, goal directedness, and extrinsic reward expectancy. Review of Public Personnel Administration, 32, 382-406.
Cho, Y. J., & Sai, N. (2013). Does organizational justice matter in the federal workplace? Review of Public Personnel Administration, 33, 227-251.
Clark, A. E. (1997). Job satisfaction and gender: Why are women so happy at work? Labor Economics, 4, 341-372.
Cohen, A. (1991). Career stage as a moderator of the relationships between organizational com- mitment and its outcomes: A meta-analysis. Journal of Occupational and Organizational Psychology, 64, 253-268.
Cohen, J. (1992). A power primer. Psychological Bulletin, 112, 155-159. Currall, S. C., Towler, A. J., Judge, T. A., & Kohn, L. (2005). Pay satisfaction and organiza-
tional outcomes. Personnel Psychology, 58, 613-640. Darcy, C., McCarthy, A., Hill, J., & Grady, G. (2012). Work–life balance: One size fits all?
An exploratory analysis of the differential effects of career stage. European Management Journal, 30, 111-120.
DeCicco, J., Laschinger, H., & Kerr, M. (2006). Perceptions of empowerment and respect: Effect on nurses’ organizational commitment in nursing homes. Journal of Gerontological Nursing, 32(5), 49-56.
Demerouti, E., Peeters, M. C., & Heijden, B. I. (2012). Work–family interface from a life and career stage perspective: The role of demands and resources. International Journal of Psychology, 47, 241-258.
Dyer, L., & Parker, D. F. (1975). Classifying outcomes in work motivation research: An exami- nation of the intrinsic-extrinsic dichotomy. Journal of Applied Psychology, 60, 455-458.
Greenglass, E. R., Burke, R. J., & Fiksenbaum, L. (2001). Workload and burnout in nurses. Journal of Community & Applied Social Psychology, 11, 211-215.
Greenwald, A., Gonzalez, R., Harris, R., & Guthrie, D. (1996). Effect sizes and p values: What should be reported and what should be replicated? Psychophysiology, 33, 175-183.
Guthrie, E., Tattan, T., Williams, E., Black, D., & Bacliocotti, H. (1999). Sources of stress, psychological distress and burnout in psychiatrists: Comparison of junior doctors, senior registrars and consultants. Psychiatric Bulletin, 23, 207-212.
Hackman, J. R., & Oldham, G. R. (1976). Motivation through the design of work: Test of a theory. Organizational Behavior and Human Performance, 16, 250-279.
Harley, B. (1999). The myth of empowerment: Work organisation, hierarchy and employee autonomy in contemporary Australian workplaces. Work, Employment and Society, 13, 41-66.
Harrington, J. R., & Lee, J. H. (2015). What drives perceived fairness of performance appraisal? Exploring the effects of psychological contract fulfillment on employees’ perceived fair- ness of performance appraisal in US federal agencies. Public Personnel Management, 44, 214-238.
Hersey, P., & Blanchard, K. H. (1969). Life cycle theory of leadership. Training & Development Journal, 23(5), 26-34.
Herzberg, F. (1957). Two factor theory. New York, NY: McGraw Hill. Hogan, N. L., Lambert, E. G., Kim, B., Mendenhall, M., Cheeseman, K., & Griffin, M. (2017).
Research note: Career stage and job satisfaction among southern correctional officers. Criminal Justice Studies, 30, 421-432.
Houston, D. J. (2000). Public-service motivation: A multivariate test. Journal of Public Administration Research and Theory, 10, 713-728.
Lee 741
Ilies, R., Dimotakis, N., & De Pater, I. E. (2010). Psychological and physiological reactions to high workloads: Implications for well-being. Personnel Psychology, 63, 407-436.
June, S., & Mahmood, R. (2011). The relationship between role ambiguity, competency and per- son-job fit with the job performance of employees in the service sector SMEs in Malaysia. Business Management Dynamics, 1, 79-98.
Jung, C. S., & Lee, S. Y. (2015). The Hawthorne studies revisited: Evidence from the US federal workforce. Administration & Society, 47, 507-531.
Kim, C., & Schachter, H. L. (2015). Exploring followership in a public setting: Is it a miss- ing link between participative leadership and organizational performance? The American Review of Public Administration, 45, 436-457.
Kim, S. (2002). Participative management and job satisfaction: Lessons for management leader- ship. Public Administration Review, 62, 231-241.
Kirkman, B. L., & Shapiro, D. L. (2001). The impact of cultural values on job satisfaction and organizational commitment in self-managing work teams: The mediating role of employee resistance. Academy of Management Journal, 44, 557-569.
Ko, J., & Hur, S. (2014). The impacts of employee benefits, procedural justice, and manage- rial trustworthiness on work attitudes: Integrated understanding based on social exchange theory. Public Administration Review, 74, 176-187.
Kooij, D., de Lange, A., Jansen, P., & Dikkers, J. (2008). Older workers’ motivation to continue to work: Five meanings of age: A conceptual review. Journal of Managerial Psychology, 23(4), 364-394.
Kovach, K. A. (1987). What motivates employees? Workers and supervisors give different answers. Business Horizons, 30(5), 58-65.
Lam, S. S., Ng, T. W., & Feldman, D. C. (2012). The relationship between external job mobility and salary attainment across career stages. Journal of Vocational Behavior, 80, 129-136.
Laschinger, H. K. S., Finegan, J., & Shamian, J. (2001). Promoting nurses’ health: Effect of empowerment on job strain and work satisfaction. Nursing Economics, 19, 42-52.
Lee, R., & Wilbur, E. R. (1985). Age, education, job tenure, salary, job characteristics, and job satisfaction: A multivariate analysis. Human Relations, 38(8), 781-791.
Levinson, D. J. (1978). The seasons of a man’s life. New York, NY: Random House Digital. Lewis, G. B., & Pitts, D. (2018). Deciding to retire from the federal service. Review of Public
Personnel Administration, 38, 49-82. Lum, L., Kervin, J., Clark, K., Reid, F., & Sirola, W. (1998). Explaining nursing turnover intent:
Job satisfaction, pay satisfaction, or organizational commitment? Journal of Organizational Behavior, 19, 305-320.
Maslow, A. H. (1943). A theory of human motivation. Psychological Review, 50, 370-396. McElroy, J. C., Morrow, P. C., & Wardlow, T. R. (1999). A career stage analysis of police
officer work commitment. Journal of Criminal Justice, 27, 507-516. Melguizo, T., & Strober, M. H. (2007). Faculty salaries and the maximization of prestige.
Research in Higher Education, 48, 633-668. Menguc, B., & Bhuian, S. N. (2004). Career stage effects on job characteristic–job satisfac-
tion relationships among guest worker salespersons. Journal of Personal Selling & Sales Management, 24, 215-227.
Miao, C. F., Lund, D. J., & Evans, K. R. (2009). Reexamining the influence of career stages on salesperson motivation: A cognitive and affective perspective. Journal of Personal Selling & Sales Management, 29, 243-255.
Ornstein, S., Cron, W. L., & Slocum, J. W. (1989). Life stage versus career stage: A comparative test of the theories of Levinson and Super. Journal of Organizational Behavior, 10, 117-133.
742 Review of Public Personnel Administration 40(4)
Osterman, P. (2000). Work reorganization in an era of restructuring: Trends in diffusion and effects on employee welfare. ILR Review, 53, 179-196.
Peeters, M. C., & Le Blanc, P. M. (2001). Towards a match between job demands and sources of social support: A study among oncology care providers. European Journal of Work & Organizational Psychology, 10, 53-72.
Perry, J. L., & Wise, L. R. (1990). The motivational bases of public service. Public Administration Review, 50, 367-373.
Porter, L. W. (1963). Job attitudes in management: II. Perceived importance of needs as a func- tion of job level. Journal of Applied Psychology, 47, 141-148.
Rousseau, D. M. (1989). Psychological and implied contracts in organizations. Employee Responsibilities and Rights Journal, 2, 121-139.
Rubin, E. V. (2007). The role of procedural justice in public personnel management: Empirical results from the department of defense. Journal of Public Administration Research and Theory, 19, 125-143.
Ryan, R. M., & Deci, E. L. (2000a). Intrinsic and extrinsic motivations: Classic definitions and new directions. Contemporary Educational Psychology, 25, 54-67.
Ryan, R. M., & Deci, E. L. (2000b). Self-determination theory and the facilitation of intrinsic motivation, social development, and well-being. American Psychologist, 55, 68-78.
Super, D. E. (1980). A life-span, life-space approach to career development. Journal of Vocational Behavior, 16, 282-298.
Tabachnick, B. G., & Fidell, L. S. (2007). Using multivariate statistics. Boston, MA: Allyn & Bacon.
Trottier, T., Van Wart, M., & Wang, X. (2008). Examining the nature and significance of lead- ership in government organizations. Public Administration Review, 68, 319-333.
Warr, P., & Inceoglu, I. (2012). Job engagement, job satisfaction, and contrasting associations with person-job fit. Journal of Occupational Health Psychology, 17, 129-138.
Weeks, W. A., Moore, C. W., McKinney, J. A., & Longenecker, J. G. (1999). The effects of gender and career stage on ethical judgment. Journal of Business Ethics, 20, 301-313.
Weibel, A., Rost, K., & Osterloh, M. (2009). Pay for performance in the public sector—Benefits and (hidden) costs. Journal of Public Administration Research and Theory, 20, 387-412.
Wynen, J., Op de Beeck, S., & Hondeghem, A. (2013). Interorganizational mobility within the US federal government: Examining the effect of individual and organizational factors. Public Administration Review, 73, 869-881.
Yoon, J., Baker, M. R., & Ko, J. W. (1994). Interpersonal attachment and organizational com- mitment: Subgroup hypothesis revisited. Human Relations, 47, 329-351.
Yu, H. Y., Lou, J. H., Eng, C. J., Yang, C. I., & Lee, L. H. (2018). Organizational citizen- ship behaviour of men in nursing professions: Career stage perspectives. Collegian, 25, 19-26.
Author Biography
Hyung-Woo Lee is an associate professor at the Department of Public Administration, Hannam University, South Korea. He received a PhD in public management from Sol Price School of Public Policy, University of Southern California. His major area of academic interest is performance management, trust, leadership, and employee motivation in the pub- lic sector. His works have appeared in Nonprofit and Voluntary Sector Quarterly, Review of Public Personnel Administration, Public Personnel Management, and International Journal of Manpower.
