Introduction

Organizations make an effort to improve productivity among their employees by increasing their motivational level and decreasing employee turnover intention (Shepperd, 1993, Huselid, 1995, Grant, 2008, Tnay et al., 2013). Public organizations are not an exception. If public employees are not well motivated at work, public goods and services may not be properly delivered to citizens (Naff and Crum, 1999, Alonso and Lewis, 2001). If public organizations experience a high level of employee turnover, they are forced to repeat the long process of recruiting, hiring, and training new employees, which will result in a huge cost to organizations (Waldman et al., 2004, Meier and Hicklin, 2008). Furthermore, there is no guarantee that similar quality of public goods and services would be provided by newcomers, should good performers leave. Therefore, one of the big challenges for public management is: how we can keep employees motivated (or satisfied) at work, and how we can encourage quality employees to stay at their current organizations.

Traditionally, turnover studies assumed that job satisfaction and job dissatisfaction belong to the same dimension and are placed on the opposite ends of the same continuum (Dunnette et al., 1967, Oshagbemi, 1997, McNeese‐Smith, 1999, Klassen and Anderson, 2009). According to this assumption, if the level of job satisfaction decreases among employees, employee turnover—a major indicator of job dissatisfaction—would increase as a result of the low job satisfaction. If this assumption is true, both job satisfaction and turnover issues can be addressed by focusing on only the ways to improve job satisfaction among employees. In other words, we can deal with two challenges such as job satisfaction and turnover with one solution.

However, some scholars proposed an alternative argument that job satisfaction and job dissatisfaction may not belong to the same dimension (Herzberg et al., 1959, Herzberg and Hamlin, 1961). According to this argument, job dissatisfaction issues cannot be properly remedied with the same approach that is used for addressing job satisfaction issues because the opposite of job satisfaction may not job dissatisfaction. Therefore, they suggested that two different approaches should be adopted to deal with two different organizational issues such as job satisfaction and job dissatisfaction.

This study examines whether or not the traditional assumptions for job satisfaction and dissatisfaction can be applicable to public sector settings. The findings in this study would help develop proper human resource management policies that can motivate and retain quality public employees.

Job Satisfaction and Motivation

Job satisfaction can be understood as an overall affective orientation that individuals would have toward their current work (Kalleberg, 1977) or whether or not employees like their jobs (Spector, 1997). As a more comprehensive definition, job satisfaction is a pleasurable emotional state resulting from the appraisal of one’s job as achieving or facilitating the achievement of one’s job values (Locke, 1969).

Since the first published investigation of job satisfaction (Hoppock, 1935), researchers have explored diverse determinants of job satisfaction and reached a conclusion that characteristics of individuals and work environments (job itself included) are two major sources of job satisfaction (Connolly and Viswesvaran, 2000, Judge and Larsen, 2001, Ilies and Judge, 2003). Major individual characteristics, as job satisfaction determinants, include race (Bartel, 1981, Tuch and Martin, 1991), gender (Clark, 1997, Sousa‐Poza and Sousa‐Poza, 2000), and educational backgrounds (Glenn and Weaver, 1982, Lee and Wilbur, 1985). With regards to work environments, major determinants of job satisfaction are job characteristics (Lee et al., 1983, Loher et al., 1985), rewards (Janssen, 2001, Danish and Usman, 2010), organizational size (Regoli et al., 1989, Artz, 2008), organizational structure (Idson, 1990, Willem et al., 2007), and organizational climate (Downey et al., 1975, Fu and Deshpande, 2014).

Job satisfaction is one of the major factors that can increase work motivation (Wright, 2001). If employees are more satisfied at work, they would be better motivated and ultimately could show higher work performance (e.g., Brown and Peterson, 1994, Ololube, 2006). Wright (2001) explained that employees’ satisfaction with types of rewards or working conditions they receive identifies “necessary conditions for their optimal employee motivation”. Since motivation plays an important role in producing better performance at work, organizations in which more employees are satisfied at work are more likely to show a better work performance.

Employees would be more likely to be satisfied at work if their work orientations accommodate what they prefer (Rose, 2003, Hakim, 2000, Clark, 1997). Work orientations refers to what rewards employees prefer to receive and what feelings employees would like to experience at work (Rose, 2003). While extrinsic rewards and working conditions such as pay, benefits, and job security have been highlighted to increase a job satisfaction level among employees (Amabile, 1993, Vallerand, 1997, Benabou and Tirole, 2003, Kuvaas et al., 2017), substantial attention has been given to the role of employees’ feelings at work such as achievement, recognition, and responsibility (Charbonneau et al., 2001, Grant, 2008, Chen and Law, 2016, Wulf and Lewthwaite, 2016, Mekler et al., 2017). Researchers found that, compared to private managers, job satisfaction among public manager tend to be more positively affected by intrinsic rewards such as recognition and self-development opportunity than by extrinsic rewards such as pay and benefits (Khojasteh, 1993, Rainey and Bozeman, 2000, Bright, 2005, Buelens and Van den Broeck, 2007).

Job Dissatisfaction, Absenteeism, and Turnover

Job dissatisfaction is defined as an unpleasant emotional state, based on appraising one’s job as frustrating and blocking the attainment of one’s job values (Lock, 1969). Unlike the assumption of the traditional approach regarding job satisfaction, job dissatisfaction may have a different dimension, compared to job satisfaction, and thus it would be affected by a different set of factors. Herzberg and his colleagues first proposed the Two-Factor Theory of Motivation, and argued that both satisfaction and dissatisfaction were not placed on the same dimension (Herzberg et al., 1959, Herzberg, 1968). According to this theory, the opposite of satisfaction is not dissatisfaction, but no satisfaction. In other words, satisfaction and no satisfaction are placed at the opposite ends of the same continuum. Likewise, the opposite of dissatisfaction is not satisfaction, but no dissatisfaction. Therefore, it is argued that dissatisfaction and no dissatisfaction are located at the opposite ends of the same continuum. Herzberg and his colleagues called the factors “motivators” if they were related to the feeling of job satisfaction, and argued that they were connected to work itself such as recognition, responsibility, achievement, and self-development opportunity. They also argued that a set of factors, called “hygiene factors”, were related to the feeling of dissatisfaction, and were related to working conditions and environments such as salary, benefits, interpersonal relationships, and company policies. Overall, Herzberg’s motivators are associated with higher-order needs, while hygiene factors are more like lower-order needs in Maslow’s need hierarchy (1954). For example, when employees are provided with a secure job status, employees would be less dissatisfied, instead of being more satisfied because job security is one of hygiene factors that are related to job dissatisfaction.

According to the turnover studies, employees tend to resign from their current organizations when they experience a low level of job involvement, autonomy, or pay (e.g., Price, 1977, Marsh and Mannari, 1977, Michaels and Spector, 1982, Harris and Adams, 2007, Grotto et al., 2017, Hom et al., 2019). Therefore, from the traditional perspective, resignation (i.e., voluntary turnover) can be a consequence of a low level of job satisfaction, or it can be a consequence of a high level of job dissatisfaction from the alternative perspective. However, the alternative approach to job satisfaction and dissatisfaction was proposed later and has not been shared by as many researchers as for the traditional approach, most turnover studies took the traditional assumption that job satisfaction and dissatisfaction is on the same continuum, and concluded that there was a negative correlation between job satisfaction and resignation (e.g., Mobley, 1977, Carsten and Spector, 1987, Tett and Meyer, 1993, Irvine and Evans, 1995, Lambert, 2010, Azeez et al., 2016). Researchers also confirmed that organizational performance would decrease when employee turnover increase (e.g., Jackofsky, 1984, Dess and Shaw, 2001, Shaw et al., 2005, Hur, 2013, Li et al., 2019). In sum, organizations are more likely to experience decreased performance when more employees resign, regardless of whether researchers took the traditional or the alternative assumption about job satisfaction and dissatisfaction.

Absenteeism is one of the major behavioral variables that turnover researchers have paid attention to (Parasuraman, 1982). A large body of literature supported the argument that absenteeism is an earlier behavioral indicator of eventual turnover (e.g., Lyons, 1972, Muchinsky, 1977, Mitra et al., 1992, Dalton and Todor, 1993, Harrison and Martocchio, 1998, Harrison et al., 2006, Cohen and Golan, 2007, Harrison and Newman, 2012, Carpenter and Berry, 2017). Researchers confirmed a significant and positive correlation between absenteeism and voluntary turnover (e.g., Patchen, 1960, Block et al., 1964, Lyons, 1972, Porter and Steers, 1973, Steers and Rhodes, 1978, Somers, 1995, Cohen and Golan, 2007). Therefore, we can say that employee turnover is likely to occur in the near future, if absenteeism is not properly addressed.

Based on the traditional assumption about job satisfaction and dissatisfaction, two tentative hypotheses are proposed as follows.

Hypothesis 1: Both job satisfaction and job dissatisfaction would be affected by the same aspects of a job.

Hypothesis 2: If job satisfaction increases, job dissatisfaction would decrease as a result.

Methods

Data

Given that managers play a central role in achieving goals and objectives of public organizations (Lynn et al., 2000, Meier and O’Toole Jr, 2002, Moynihan and Pandey, 2005), it is important to make them satisfied at work. To understand how public mangers can be better satisfied and how they can be encouraged to stay at their current organizations, this study used National Administrative Studies Projects (NASP III) data that surveyed public managers in Georgia (GA) and Illinois (IL) in the U.S., conducted by Bozeman and colleagues in 2005. National Administrative Studies Projects (NASP) data was first collected in 1992 and NASP III is a final attempt at a NASP database. Surveys were sent to public managers in GA and IL, and 790 public managers responded, which led to overall response rate 43 %. Among survey participants, 56 % were male and 44 % were female. This survey was asking public managers to evaluate how important each aspect of a job was for them. If an aspect of a job is evaluated as important or very important by a public manager and his/her expectation for this aspect of a job is met, we can imagine that their job satisfaction level would increase as a result.

Measures for intrinsic aspects of a job

Opportunity for advancement (IA 1). Survey participants were asked how important it was to have an opportunity for advancement within the organization’s hierarchy (4 being very important and 1 being not important).

Opportunity for training and career development (IA 2). Survey participants were asked how important it was to have an opportunity for training and career development at work (4 being very important and 1 being not important).

Pride (IA 3). Survey participants were asked how much they agreed to the statement, “I feel a sense of pride working for this organization” (4 being strongly agree and 1 being strongly disagree).

Recognition (IA 4)¹. Survey participants were asked how much they agreed to the statement, “There are recognition for me to work hard in my job” (4 being strongly agree and 1 being strongly disagree).

Measures for extrinsic aspects of a job

Job security (EA 1). Survey participants were asked how important it was to be offered job security at work (4 being very important and 1 being not important).

Pension and retirement plan (EA 2). Survey participants were asked how important it was to get good pension or retirement plan from the work (4 being very important and 1 being not important).

Less conflict (EA 3). Survey participants were asked how important it was to meet the desire for low conflict work environments (4 being very important and 1 being not important).

Salary (EA 4). Survey participants were asked how important it was to get a good salary from the work (4 being very important and 1 being not important).

Measures for dependent variables

Two dependent variables were used in the regression analyses (see Tables 2 and 3). For the first dependent variable, overall job satisfaction was used to measure job satisfaction. More specifically, public managers were asked how much they agreed to the statement, “Overall, I am satisfied with my job” (4 being strongly agree and 1 being strongly disagree). For the second dependent variable, absenteeism was adopted as a proxy indicator for turnover that measures the level of their job dissatisfaction. More specifically, the number of absent days due to personal leave was used to measure absenteeism. While turnover is a good measure for job dissatisfaction, absenteeism was used as a proxy measure due to unavailable information about turnover. As previously mentioned, researchers confirmed a positive correlation between absenteeism and voluntary turnover, and absenteeism is regarded as an earlier behavioral indicator of eventual turnover.

Findings

Before running regression analyses, Pearson correlation analysis was conducted. The correlation analysis showed that public managers’ job satisfaction was significantly correlated with most intrinsic aspects/rewards of a job such as IA 2 (.09, p<.05), IA3 (.58, p<.01), and IA4 (.43, p <.01), but neither of extrinsic aspects/rewards had significant correlation with job satisfaction (see Table 1). With regards to public managers’ absenteeism, no factors were significantly related to it except one extrinsic aspects/rewards of a job (EA 2, -.14, p <.001). Overall, these results are not inconsistent with what Herzberg et al (1959) predicted in their study. According to these results, it seems that public managers tend to be more job satisfied when they put more importance on training and development (IA 2), pride (IA 3), and recognition (IA 4). It also seems that public managers would be less frequently absent when they put more importance on pension and retirement plan (EA 2).

Multicollinearity might be an issue in regression analyses when there exist high correlations among independent variables and control variables in a regression model (Schroeder et al., 1990, Grewal et al., 2004). Multicollinearity will undermine the statistical significance of independent variables in regression analysis and the results of regression analysis may not be reliable (Blalock Jr, 1963, Smith and Sasaki, 1979). In this study, high correlations existed between motivators, between hygiene factors, and between motivators and hygiene factors (see Table 1), which indicate a possible existence of milticollinearity in regression analysis. However, when Variance Inflation Factors (VIFs) were examined, no evidence of multicollinearity was found. According to the general rule of interpreting VIF (O’brien, 2007, Robinson and Schumacker, 2009), multicollinearity exists if the largest individual VIF is greater than 10 and if the mean of all VIFs is considerably larger than one such as 10 or 30. For all regression analyses in this study, individual and mean VIFs were between 1 and 2.

Table 1. Correlation matrix

* p < .05, ** p < .01, *** p < .001

Regression analysis

Regression analyses were conducted in two steps for both job satisfaction and dissatisfaction. For pubic managers’ job satisfaction (see Table 2), the first regression analysis was conducted with only extrinsic aspects/rewards of a job in Model 1, and none showed a significant effect on job satisfaction. However, when intrinsic aspects/rewards of a job were added in Model 2, significant and positive effects were found in two variables such as pride (IA3) and recognition (IA4). As indicated in the correlation analysis, this result confirmed what Herzberg et al. (1959) predicted in their study. That is, only some intrinsic aspects/rewards of a job showed a significant and positive influence on public managers’ job satisfaction.

Public managers’ absenteeism (i.e., dissatisfaction) and its related factors were explored in Models 3 and 4 (see Table 3). When regression analysis was conducted with only intrinsic aspects/rewards of a job in Model 3, absenteeism was not affected by any of these intrinsic aspects/rewards. In Model 4, however, when extrinsic aspects/rewards of a job were added, organizational pension and retirement plan (EA2) had a significant and negative coefficient, and desire for low conflict (EA3) had a significant and positive coefficient. According to this finding, public managers’ expectation for EA3 did not seem to be met whereas public managers’ expectation for EA2 seemed to be met. That is, public managers’ absenteeism would go up when their expectation for EA3 that was rated important/very important was not met.

In Model 4, the job satisfaction level was also added to Model 3 to verify whether or not absenteeism is affected by job satisfaction. If job dissatisfaction is opposite to job satisfaction on the same continuum as seen in the traditional assumption, absenteeism should go down when job satisfaction goes up. That is, a negative correlation is supposed to exist between absenteeism and job satisfaction if the traditional assumption holds true. However, Model 4 showed that job satisfaction had a positive and non-significant effect on absenteeism.

In sum, job satisfaction was significantly affected by only intrinsic aspects/rewards of a job as seen in Models 1 and 2, whereas absenteeism was significantly influenced by only extrinsic aspects/rewards of a job as seen in Models 3 and 4. Therefore, Hypothesis 1 was not supported. Furthermore, it turned out that absenteeism was not located at the opposite end of the same continuum with job satisfaction. As seen in Model 4, absenteeism was not affected by the level of job satisfaction. Therefore Hypothesis 2 was not supported, either.

Table 2. Regression analysis for job satisfaction

Notes: Unstandardized coefficients are displayed * p < .05 ** p < .01 *** p < .001

Table 3. Regression analysis for job dissatisfaction

Notes: Unstandardized coefficients are displayed * p < .05 ** p < .01 *** p < .001

Discussion

This study assessed traditional assumptions about employees’ job satisfaction and examined whether or not same aspects/rewards of a job have determined both job satisfaction and job dissatisfaction among public managers. According to the findings in this study, job satisfaction and job dissatisfaction need to be addressed by taking different approaches, respectively. More specifically, intrinsic aspects/rewards of a job such as pride (IA3) and recognition (IA4) showed a significant and positive influence on improving public managers’ job satisfaction. The findings in this study also indicate that job dissatisfaction needs to be taken care of by focusing on extrinsic aspects/rewards of a job such as pension & retirement plan (EA2) and desire for low conflicts (EA3). Again, the traditional assumption about job satisfaction and dissatisfaction was not supported with public managers’ sample. The finding of this study support other studies with employees of for-profit organizations (Koelbel et al., 1991, Lundberg et al., 2009, Sanjeev and Surya, 2016), and with public employees (Park et al., 1988, Chu and Kuo, 2015).

Generally speaking, management is more constrained in public organizations than in for-profit organizations, and this constraint is one of the major causes for a low level of motivation/job satisfaction among public managers (Rosenbloom, 1998, Pollitt and Bouckaert, 2004). Furthermore, given a tight budget for improving working conditions and a limitation to provide monetary rewards to good performers in public organizations, it becomes a big challenge to increase public managers’ motivation level. Traditionally public organizations have focused on improving extrinsic aspects/rewards of a job such as pay raise or better retirement plan, and they have failed to meet public managers’ expectations for those. Findings in this study would help increase the level of motivation among public managers because it would cost little to improve intrinsic aspects of a job or to give intrinsic rewards to public managers such as recognition and pride.

While increasing job satisfaction is important, we also know that employee turnover is costly to an organization. In particular, public managers’ resignation can be a huge loss to their organizations and citizens because public managers play a key role in running public organizations and in providing quality public services to citizens (Lynn et al., 2000, Meier and O’Toole Jr, 2002, Moynihan and Pandey, 2005). Therefore, possible causes for their resignation should be better taken care of in advance. As mentioned previously, job dissatisfaction is known as a precursor to resignation (e.g., Egan et al., 2004, Coomber and Barriball, 2007) and absenteeism is a major indicator for employee turnover (e.g., Somers, 1995, Cohen and Golan, 2007). Therefore, if public managers are dissatisfied at work, they would be frequently absent, then they may end up quitting their jobs. According to the findings in this study, public managers would be less frequently absent when their organizations meet public managers’ expectation for extrinsic aspects/rewards of a job such as pension and retirement plan (EA 2) and desire for low conflict (EA 3). Overall, it is recommended to take care of both job satisfaction and dissatisfaction among public managers by focusing on different aspects/rewards of a job, respectively, unlike traditional assumptions suggest.

This study provides practical and effective ways to motivate and retain quality public managers by focusing on intrinsic and extrinsic aspects/rewards of a job. However, this study has some limitations because some data were not available. First, individual personality was not considered in this research although it has significant influences on their job satisfaction and dissatisfaction (Francis and Rodger, 1994, Strus and Cieciuch, 2017). Secondly, although turnover is a more common measure for job dissatisfaction, absenteeism was used as a proxy measure for job dissatisfaction in this study. Overall, this study opens a new possibility to take care of job satisfaction and dissatisfaction among public managers. Hopefully, the findings in this study would be verified by more comprehensive studies in the future.