The Impact of Coproductive Taxpayers’ Supervisory Behaviors on the Job Involvement of Tax Collectors

Abstract

Citizen coproduction of public services can be a powerful force that influences employee work motivation. However, little research has been conducted to explore what kind of coproductive behaviors can motivate public employees in a positive direction. The two purposes of this article are to explore whether taxpayers would behave as tax collectors’ supervisors during their participation in the production and delivery processes of tax services, and whether such taxpayers’ supervisory behaviors would impact the job involvement of tax collectors. Drawing on the tax collectors of the tax collection agencies in Taipei, Taiwan, the authors first develop a coproductive-taxpayer-as-supervisor measure through focus group interviews and accepted psychometric testing procedures. The measure reveals that tax collectors would perceive taxpayers as their supervisors if taxpayers provide challenges and support to tax collectors. A mail survey further indicates that both taxpayers’ challenge and supportive behaviors are positively related to tax collectors’ job involvement.

Keywords

coproductive-taxpayer-as-supervisor taxpayer challenge behavior taxpayer supportive behavior job involvement tax collectors

Introduction

This article explores what kind of coproductive taxpayers’ supervisory behaviors can impact the job involvement of tax collectors during their participation in the production and delivery processes of tax services. Why do taxpayers’ supervisory behaviors matter? Denhardt (1995) maintains that the future of public-sector leadership is not just the exercise of control by those in formal positions of power. Instead, he argues that anyone can exercise public leadership if he or she can bring about the energizing effect of leadership by helping the group or organization run more smoothly, such as helping the group or organization understand what it needs. In addition, many researchers have suggested that citizens should be treated as partners or owners of government, and not just subjects, voters, or clients or customers. Citizen-owners or citizen-partners are the main beneficiaries of the output and outcome of the public sector and are sovereign for the scope and affairs of the public sector that is today more oriented toward assessing its performance (e.g., Alford, 1998; Beckett, 2000; Schachter, 1995; Vigoda, 2002). A typical example would be that disgruntled citizens yell at public employees with the phrase “I pay your salary,” or “Remember, you work for me.” Therefore, active citizen-owners or citizen-partners play an important role in effective agency performance because they examine and evaluate the public services they are receiving to meet their expectation of improvement.

Traditionally, tax collection agencies operated on a command and control mechanism based on distrust and prosecution. Today, tax administrations in many countries have voluntary compliance programs designed to increase taxpayers’ willingness to collaborate and to fulfill their tax duty, through encouraging taxpayers to participate in the production and delivery processes of tax services (Alm, Cherry, Jones, & Mckee, 2010; Braithwaite, 2009). These include simplification of the tax system, openness and transparency in the tax collection process, and active provision of counseling to taxpayers. This approach helps to establish a synergetic relationship between taxpayers and tax collection agencies, working as equal partners, which is analogous to the “service and customer” relationship in New Public Management (NPM) principles (Kirchler, 2007).

Collaboration is the core value of coproduction and the key to its success, and the willing contribution of time and effort by citizens directly adds value to public services (Alford, 2002; Lee, Chen, & Chou, 2013; Payne, Storbacka, & Frow, 2008). Coproduction transforms the role of citizens from that of service users to that of coproducers of service value, and shares joint responsibility for the quality of public services with civil servants, which helps to alleviate government employee alienation as well as their resistance to coproduction (Sharp, 1980). Although citizen coproduction can be a powerful force that influences employee work motivation (Paarlberg, 2007; Perry & Wise, 1990), little attention has been paid to explore what kind of coproductive behaviors of citizens can motivate public employees in a positive direction.

In the private sector, coproductive customers often perform as supervisors through providing task guidance and good feelings (i.e., supportive behaviors, such as positive feedback) in service encounters, influence employees more immediately than their direct superiors, and even perform as substitutes for leadership (Schneider & Bowen, 1995). Therefore, we argue that taxpayers acting as coproducers would perform some functions of supervisors, and that taxpayers’ supervisory behaviors would influence tax collectors’ work motivation, such as job involvement.

Employee job involvement is an important antecedent variable for organizational performance (Brown, 1996; Moynihan & Pandey, 2007), and the organizational performance of tax agencies has a great impact on the finances of a modern democracy (Rosen & Gayer, 2010). However, because job involvement is more affected by individual attributes and previous life experiences rather than by organizational factors, such as organizational culture and job characteristics, managers of public organizations cannot easily stimulate job involvement by encouragement (Brown, 1996; Liou & Bazemore, 1994; Moynihan & Pandey, 2007). Also, public managers have limited tools by which they may motivate and retain their employees because civil service rules and standard pay grades limit the ability to recognize and reward individuals through pay, promotion, or bonuses (Klingner & Nalbandian, 2003). Furthermore, Wright’s (2001) literature review also shows that antecedent variables of work motivation in the public sector are limited to two major streams: employee characteristics and internal organizational environment. Accordingly, it seems that most scholars of the public sector tend to concentrate heavily on the internal organizational factors on work motivation and pay less attention to the potential influence of citizens. Therefore, the search for external factors that affect the job involvement of tax collectors to improve the efficiency and productivity of tax agencies is an important issue in public management.

Two studies are conducted to investigate the relationship between taxpayers’ supervisory behaviors and the job involvement of tax collectors. The purpose of Study 1 is to develop a measure for the coproductive-taxpayer-as-supervisor (CTS) construct. An extensive search is first conducted across the public management literature to determine whether a measure of CTS or related measures already exist. Because this exercise reveals no such existing works, we turn to primary data collection approaches and conduct two focus group interviews to conceptualize and operationalize the construct. Dimensionality of the initial CTS measure obtained from the focus groups is explored through a mail survey. In Study 2, we then infer the relationship between the CTS and job involvement and conduct another mail survey to empirically test the relationship. Reliability and validity of the CTS scale are also tested. We conclude with a discussion of the theoretical and managerial implications of this research.

Study 1: Development of CTS

Measurement

In public administration literature, there are two basic approaches to assessing the internal and external environments of public organizations: One uses objective, archival measures and the other uses subjective, perceptual measures (Andrews, Boyne, Meier, O’Toole, & Walker, 2007; Bommer, Johnson, Rich, Podsakoff, & Mackenzie, 1995; Mezias & Starbuck, 2003; Yang & Pandey, 2009). Andrews et al. (2007) conclude that archival measures are more suitable for analyzing the responses of public organizations to broader features in their environment, such as levels of socio-economic deprivation; whereas perceptual measures are more useful for exploring how employees respond to internal and external circumstances that are unique to individual organizations, such as political environment. For example, many researchers use managers’ perceptions to measure policy actors’ political influence and support and use these perceptions as explanatory variable of the managers’ behaviors (e.g., De Lancer Julnes & Holzer, 2001; Pandey & Wright, 2006; Potoski & Woods, 2001; Wang & Berman, 2001; Yang & Pandey, 2009).

Perceptual measures are also especially suitable for studying adaptive patterns across organizations (Weick, 1995), because almost all internal and external environments of public organizations are socially constructed and enacted realities (Giddens, 1984; Hodgkinson, 2001). Simon (1953), the renowned social scientist, emphasizes that “environmental forces mold organizations through the mediation of human minds” (p. 236). Therefore, public employees’ perceptions of environment are necessary mediators of relationships between objective characteristics of environment and employees’ responses (Weick, 1995). For example, Whicker, Strickland, and Olshfski (1993) maintain that, “how individual public administrators define their political environment influences their behavior and operations” (p. 539). In the field of organizational behavior, Robbins (2005) stresses that, “people’s behavior is often based on their perception of what reality is, not on reality itself,” and that, “the world as it is perceived is the world that is behaviorally important” (p. 122).

The purpose of this study is to analyze the effect of individual taxpayers’ supervisory behavior on individual tax collectors’ job involvement. Taxpayers’ supervisory behavior can be classified as a unique external circumstance of tax collection agencies. In addition, many researchers (e.g., Boyd, Dess, & Rasheed, 1993; Miller, 1988) have argued that public employees’ perceptions on their environment is more crucial to organizational strategy, structure, and process than are archival measures of the environment because perceptual measures allow researchers to portray an organization’s environment from the perspective of organizational members, and suffer less from the problems associated with aggregation of data across members in an organization. For example, two tax collectors may perceive the same taxpayer’s behavior differently, and may respond differently as a result. Therefore, this study uses tax collectors’ perceptions to measure taxpayers’ supervisory behaviors and investigates the relationship between tax collectors’ perceived CTS behaviors and the job involvement of the tax collectors. Accepted psychometric procedures (Churchill, 1979; Gerbing & Anderson, 1988; Nunnally & Bernstein, 1994) were followed to develop a CTS scale.

Focus Group Participants

The tax collectors of all 36 tax collection offices of the Taipei National Tax Administration (TNTA) and the Taipei Revenue Service (TRS) in Taiwan were used as the sources of data collection. In recent years, TNTA and TRS have reformed from traditional “tax administration” agencies, one which collected taxes via legally sanctioned coercive powers, to “tax service” agencies, emphasizing and ensuring a fair and reasonable tax service environment and assisting taxpayers to fulfill their tax obligations (Hsieh, Chou, & Chen, 2002; Lee et al., 2013). At the heart of this revolutionary transformation, taxpayers are regarded as coproductive customers and encouraged to collaborate. The TNTA and the TRS have been using public relations and marketing techniques to improve taxation education and listening to the feedback from the general public, with an aim to develop a relationship of mutual trust, to increase the willingness of taxpayers to collaborate as partners, and to build a win-win taxation environment between government and the taxpayers. Collaboration between the agencies and taxpayers entails the standardization of collection procedures and service guidelines, which reduces bureaucracy and better educates the public about how to coproduce tax services (Taipei National Tax Administration, Ministry of Finance of R. O. C. [TNTA], 2012a; Taipei Revenue Service, 2012).

Participants in the focus groups were volunteers from the tax collectors of eight out of the 18 tax collection offices of TNTA and nine out of the 18 tax collection offices of TRS, with, at most, two tax collectors from each tax collection office. Twenty-two volunteers were recruited. The voluntary participants were randomly assigned to each of the groups. The tax collectors from the same tax collection office were assigned randomly to separate groups. As a result, there were 11 participants in each group (eight women and three men in one group, and seven women and four men in the other group). In terms of sample structure, 68% were female; the age ranged from 26 to 45 years and the average years of service was 9.01, with a standard deviation of 5.58 years.

The focus group interviews were conducted in two consecutive evenings, one in the evening of a Wednesday and the other the following evening. After having introduced the topic, guaranteed all participants confidentiality, and received permission for tape recording the discussion, the moderator encouraged these participants to discuss what CTS meant to them and what critical incidents identified taxpayers who were perceived by the participants to behave as supervisors during service encounters and influenced the participants’ behaviors and attitudes toward their jobs. The focus group interviews lasted 2 hr on average.

Measure and Analyses of Dimensionality

Various statements about critical incidents related to the CTS construct were extracted from the transcripts of the focus group interviews. We listed together those statements that seemed to be duplicate or similar, and analyzed the core concept that underlies each statement group. We then produced a general statement that represents the underlying concept of each statement group. This process resulted in six different statements, which are presented in Table 1. Traditional Chinese was the native language for all intended respondents, and all items shown in Table 1 were translated from traditional Chinese into English, using the back-translation strategy recommended by Brislin (1980). After the preliminary general statements were developed, the statements were reviewed by five veteran tax collectors. The purpose of this pretest was twofold: first, to ensure the overall wording is familiar to the target sample (i.e., tax collectors) and is able to fully and clearly express the intended meaning; second, to ensure the appropriateness of the questions to prevent respondents from leaving the survey incomplete after encountering questions that are too difficult, sensitive to a personal issue, or time-consuming.

Table 1.

Measure for Coproductive-Taxpayer-as-Supervisor Construct.

Sub-dimension	Statement
1. Taxpayer challenge behavior	1. Taxpayers apply appropriate tax regulations in arguing and questioning tax audit results.
	2. Taxpayers seem to intend to evade tax.
	3. Taxpayers I served before come to me for different and more difficult requests.
2. Taxpayer supportive behavior	1. Taxpayers are willing to accept my suggestions and make changes accordingly.
	2. Taxpayers talk or write to my supervisors about their satisfaction with my service.
	3. Taxpayers are interested in learning how to file returns on their taxation correctly from me.

To explore the dimensionality of the CTS measure, a self-administered mail survey was conducted by randomly contacting 12 tax collectors from each of the 36 tax collection offices of the TNTA and the TRS through intra-organizational mailing systems (excluding the participants in the focus groups). This sample size was decided with the hope of obtaining sufficient respondents for meeting Nunnally and Bernstein’s (1994) suggested standard, according to which a minimum of five respondents per scale item is recommended as a benchmark for appropriate sample size for the exploration of scale dimensionality. Based on this, our data collection required a minimum of 30 completed questionnaires. In total, 432 questionnaires were distributed. Two hundred fifty seven completed questionnaires were returned. The response rate was 60%. In terms of sample structure, 69% were female, the age ranged from 24 to 50 years, and the average years of service was 11.21, with a standard deviation of 6.83 years.

To enable a better understanding of the grouping of the six CTS items and for ease of interpreting and describing results, an exploratory factor analysis with principal axis factor extraction and varimax rotation (Hair, Black, Babin, & Anderson, 2010, Nunnally & Bernstein, 1994) was conducted to explore the underlying factors comprising the CTS measure. We selected orthogonal (i.e., varimax) rotation technique over oblique analytic rotation because the results of a trial exploratory factor analysis with oblique analytic rotation revealed that the variable delta (δ) was less than zero (−1.85), which indicated that the solution is increasingly orthogonal (Tabachnick & Fidell, 2007). Therefore, although few constructs in the real world are uncorrelated, we selected orthogonal rotation technique. Two factors with eigenvalues greater than one were extracted, explaining a total of 77% of the variance in the six items. Detailed analysis procedures and results are shown in Appendix A.

As shown in Table 1, each factor consists of three statements. We concluded that the first factor represents taxpayer challenge behavior (TCB). The second factor appears to represent “good feelings” provided by taxpayers and is similar to the concept of supportive leadership defined by House and Dessler (1974). We therefore, named the second factor taxpayer supportive behavior (TSB).

Study 2: Development and Tests of Hypotheses

Hypothesis Development

The results of Study 1 reveal that the CTS measure comprises TCB and TSB sub-dimensions. In the following, we infer the relationships between these two variables and job involvement.

Job challenge is an important mediator variable between objective job characteristics and job involvement (Brown & Leigh, 1996). For example, among the job characteristics, skill variety and task identity have both been shown to be capable of increasing job involvement significantly in that they make jobs more interesting, meaningful, and challenging (Brown, 1996; Hassan, 2012). Skill variety challenges employees to use different skills and abilities which they personally value in doing the work. Task identity leads employees to perceive that they have accomplished something of importance on their own. These significant job characteristics are the elements of job enrichment and job enlargement programs, which are aimed at providing employees with more generally meaningful and challenging jobs, instead of simple, routine, and standardized jobs, and are expected to increase employees’ positive attitudes toward work and thus, profitability of the organization (Hackman & Lawler, 1971; Hackman, Oldham, Janson, & Purdy, 1975). Existing literature has shown that simplified, nonchallenging jobs often result in employee dissatisfaction, in work alienation, and in increased absenteeism and turnover (Brown, 1996).

In this study, the TCB variable seems to be a composite measure of the sources of skill variety and task identity provided by taxpayers. The items “Taxpayers apply appropriate tax regulations in arguing and questioning tax audit results” (Item 1 in Table 1) and “Taxpayers I served before come to me for different and more difficult requests” (Item 3 in Table 1) together imply that tax collectors have the chance to, or have a need to, appropriately use different tax regulations in responding to taxpayers’ challenges and requests. Basically, skills in tax audits and expertise on taxation are essential requirements for qualified tax collectors. Thus, we argue that Items 1 and 3 are indicative of skill variety. Item 3 also seems to represent task identity. Presumably, if a taxpayer comes back to a tax collector for different and more difficult requests, it would indicate that the taxpayer is satisfied with the previous service delivered by the tax collector, which is a form of positive feedback from taxpayers. This would lead the tax collector to believe that they themselves have accomplished something important to the taxpayer. Schneider and Bowen (1995) have suggested that customer feedback can lead employees to have a deeper sense of the significance of their tasks and use more skills in doing their tasks, and thus increase their job involvement.

Item 2, “Taxpayers seem to intend to evade tax,” seems to represent task challenges to tax collectors. Finding out if taxpayers are intending to evade tax is a challenging task in itself to tax collectors. Tax evasion means to dodge tax illegally, and tax avoidance means to reduce one’s tax bill by any legal method. However, the dividing line is blurred. What is more blurred is that tax evasion can be further divided into intentional and unintentional evasion (Lewis, 1982). Under the assumption of the economics-of-crime perspective (Allingham & Sandmo, 1972) that taxpayers are rational and strategic risk-takers and maximize the expected utility of the tax evasion gamble, tax administrations generally view taxpayers as potential criminals. That is, taxpayers weigh the potential benefits of successful evasion against the risky prospect of detection and punishment. Empirical results have shown that taxpayers will choose to evade taxes if the expected return per dollar on evading the tax is positive (Allingham & Sandmo, 1972; Alm, 2012). Therefore, tax collectors are asked and rewarded by the government, particularly by the tax enforcement authorities, to find out tax savings that take advantage of technical loopholes that the government wants to close because they reduce the amount of revenue collected and, hence, the resources available for public expenditure. Using the preceding as a basis, the following hypothesis is advanced.

Hypothesis 1: TCB is positively correlated with tax collectors’ job involvement.

Supportive leadership has been shown to be positively and significantly correlated with job involvement (Brown, 1996). In House and Dessler’s (1974) perceived leadership behavior scale, supportive leadership reflects the concepts that a leader is friendly, approachable, and considerate of the needs of subordinates. In the present study, the TSB scale seems to represent these concepts in that the friendliness and approachableness concept can be measured by Item 2 (i.e., taxpayers talk or write to my supervisors about their satisfaction with my service) in the scale, and in that the consideration dimension can be measured by Item 1 (i.e., taxpayers are willing to accept my suggestions and make changes accordingly) and Item 3 (i.e., taxpayers are interested in learning how to file returns on their taxation correctly from me).

To improve many potential relevant antecedent factors of tax compliance, such as tax morale and social norm, tax administrations in many countries have adopted the “service paradigm” that values the role of enforcement and also appreciates the facilitating role of tax administration as a service provider. The paradigm emphasizes such things as assisting taxpayers in the process of filling in tax returns and paying taxes. It predicts that taxpayers would comply more voluntarily if tax administrations regarded taxpayers as collaborative partners and clients and assisted them in choosing the best tax planning strategies to reduce their tax burden legally. Voluntary compliance is collaborative citizens’ positive actions that are in agreement with the public organization’s requirements (Alford, 2002). This “service and client approach” builds a relationship of mutual trust and cooperation, thereby improving the willingness of taxpayers to file tax returns and pay taxes on time and honestly (Alm, 2012; Alm et al., 2010; Kirchler, 2007). We can predict that a high level of voluntary compliance from collaborative taxpayers helps tax collectors to complete the tasks they are responsible for more efficiently and more effectively (Alford, 2002), which would lead to higher job involvement (Brown, 1996). Thus, Hypothesis 2 is stated as follows.

Hypothesis 2: TSB is positively correlated with tax collectors’ job involvement.

Method

Data

All of the remaining tax collectors of TNTA and TRS who were not contacted in Study 1 were mailed a questionnaire through intra-organizational mailing systems. In total, 658 questionnaires were distributed and 342 completed questionnaires were returned. The response rate was 52%. In terms of sample structure, 65% were female; the age ranged from 25 to 52 years and the average years of service was 10.82, with a standard deviation of 6.27 years.

Measures

Along with the CTS measure in the questionnaire, measures for job involvement (Kanungo, 1982) and most of the significant antecedent variables of job involvement arrived at in Brown’s (1996) meta-analysis were also included. These antecedent variables are intended to be used as control variables in later analysis. These measures are Work Ethic Endorsement scale (Mirels & Garrett, 1971), Self-Esteem Inventory (Rosenberg, 1979), Job Characteristics Inventory for skill variety, task identity, and feedback (Sims, Szilagyi, & Keller, 1976), and Perceived Leader Behavior scales for leadership consideration and leadership participation (House & Dessler, 1974). This study uses a Likert-type 7-point scale for all of its variables. Respondents are asked to rate their level of agreement or disagreement with each statement (1 = strongly disagree, 4 = neither agree nor disagree, 7 = strongly agree) in scales for all the study variables but job characteristics variables. The scales for job characteristics variables ask respondents to rate the amount of their skill variety, task identity, and feedback described in each statement (1 = very little, 4 = neither much nor little, 7 = very much). Statements for each measurement are shown in Appendix B.

Reliability and Validity of Measures

In terms of reliability, Cronbach’s α is .74 for TCB and .72 for TSB, both of which exceed the minimum standard of .70, which indicates that the scales have sufficient reliability (Nunnally & Bernstein, 1994). The Cronbach’s α values of the rest of the study variables are shown in Table 2, and are above .81, indicating sufficient reliability.

Table 2.

Means, Standard Deviations, Cronbach’s Alpha (α) Coefficients, and Intercorrelations of Study Variables.

Variable	M	SD	α	1	2	3	4	5	6	7	8	9	10	11
1. Job involvement	3.80	0.94	.81	—
2. Taxpayer challenge behavior	3.93	1.22	.74	.23**	—
3. Taxpayer supportive behavior	3.72	1.13	.72	.14*	.31**	—
4. Work ethic endorsement	4.44	0.85	.84	.30**	.24**	.19**	—
5. Self-esteem	4.95	0.28	.86	−.04	−.07	−.01	.14*	—
6. Skill variety	2.74	0.89	.81	.20**	.21**	.14*	−.01	.07	—
7. Task identity	2.87	0.94	.84	.20**	.17**	.20**	.07	.08	.26**	—
8. Feedback	2.58	0.76	.89	.23**	.19**	.23**	.08	.00	.28**	.24**	—
9. Leadership consideration	4.51	1.22	.91	.26**	.08	.10	.19**	.11*	.13*	.16**	.29**	—
10. Leadership participation	4.57	1.37	.89	.21**	.07	.09	.18**	.12*	.18**	.14*	.31**	.90**	—
11. Leadership P-C^a	4.53	1.24	.94	.24**	.08	.10	.19**	.12*	.16**	.16**	.30**	.99**	.96**	—

Note. P-C = leadership participation-consideration.

An aggregate of leadership consideration and leadership participation (Variables 9 and 10).

p < .05. **p < .01.

Regarding validity, a confirmatory factor analysis shows that the two-factor measurement for the CTS construct has sufficient model fit (Goodness-of-Fit Index [GFI] = .94; Adjusted GFI [AGFI] = .90; chi-square (χ²) = 23.78, df = 8, p < .01; Normed χ² = 2.97; root mean square error of approximation [RMSEA] = .07), convergent validity, and discriminant validity. Detailed procedures and results of the confirmatory factor analysis are shown in Appendix C. The model fit indices for the rest of the study variables are shown in Appendix B. The model fit indices show that these variables have sufficient model fit.

Analyses and Results

Table 2 shows the means, standard deviations, Cronbach’s α values of, and Pearson’s simple correlations between, the study variables. The Pearson’s simple correlations show that all explanatory variables are significantly and positively correlated with job involvement as expected, except for self-esteem (r = −.04, p > .10). The insignificant correlation between job involvement and self-esteem indicates that self-esteem may not be an important predictor in predicting job involvement in this study. In addition, as shown in Table 2, the correlations between self-esteem and the other intended control variables (i.e., work ethic endorsement, skill variety, task identity, feedback, leadership consideration, leadership participation) are either insignificant at the .05 level or pretty low, which indicates that self-esteem may not be an important control variable in clarifying the relationships between TCB and job involvement, and between TSB and job involvement. The results of a multiple regression analysis including all the predictor variables showed that self-esteem is not an important predictor and control variable in predicting job involvement. Therefore, we decided to delete self-esteem from further analysis to prevent loss of degrees of freedom and overfitting (Hair et al., 2010). The simple correlation between leadership consideration and leadership participation is nearly perfect (r = .9, p < .01), which might cause a multicollinearity problem if these two variables are to be included in a multiple regression analysis at the same time for hypothesis testing. To deal with this problem, following Cohen, Cohen, West, and Aiken’s (2003) comment, we decided to combine the measures for these two perceived leader behavior variables to form a new measure and termed this new variable leadership participation-consideration (Leadership P-C). In terms of reliability, Cronbach’s α values for all of the study variables are above .72, exceeding the minimum standard of .70 (Nunnally & Bernstein, 1994), which indicates that the scales have sufficient reliability.

Before testing the hypotheses, we used a confirmatory factor-analytic approach to Harman’s one-factor test to assess whether common method bias (CMB) constituted a problem in the testing. CMB might be present when everything in a measurement instrument is measured by a single individual using the same scale for every item. It typically increases the size of the effect and the possibility of Type I error in the data analysis. The rationale for Harman’s one-factor test is that if CMB poses a serious threat to the analysis and interpretation of the data, a single latent factor would account for all manifest variables (Craighead, Ketchen, Dunn, & Hult, 2011; Podsakoff, MacKenzie, Lee, & Podsakoff, 2003). However, CMB does not show a serious threat if the fit for the one-factor model is worse than the measurement model, in which manifest variables are assigned to load onto their theoretical constructs (MacKenzie, Podsakoff, & Paine, 1999; Podsakoff & Organ, 1986). In this study, the one-factor model yielded a χ² = 8,393.68 with 1,952 degrees of freedom (compared with χ² = 3,941.73, df = 1,908 for the measurement model with the eight study variables as latent factors). The fit is considerably worse for the unidimensional model than for the eight-factor measurement model (Δχ² = 4,451.95, Δdf = 44, p < .01), indicating that CMB is not a serious threat in the study, and that the respondents perceived of the items in the scales as measuring different constructs.

A hierarchical multiple regression analysis (Cohen et al., 2003) was conducted to test the hypotheses. The analysis results are shown in Table 3.

Table 3.

Hierarchical Multiple Regression Analyses Predicting Job Involvement.

	Step 1		Step 2
Criterion variable: Job involvement	β	t-value	β	t-value	Effect size (r)
Explanatory variable
1. Work ethic endorsement	.28	5.84**	.23	4.75**	.252
2. Skill variety	.15	2.49**	.14	2.31*	.125
3. Task identity	.12	1.95*	.11	1.87*	.102
4. Feedback	.12	2.14*	.10	1.92*	.104
5. Leadership participation- consideration^a	.16	3.21**	.15	3.18**	.171
6. Taxpayer challenge behavior	—	—	.21	4.28**	.228
7. Taxpayer supportive behavior	—	—	.14	2.34**	.127
R ²	.1706		.2952
Adjusted R²	.1634		.2802
F value	23.37**		23.04**
Increase in adjusted R²	.1634**		.1168**

An aggregate of leadership participation and leadership consideration.

p < .05. **p < .01.

In the first step, job involvement was regressed on all control variables. As expected, on the basis of the empirical findings in the existing literature (Brown, 1996), this model is statistically significant (adjusted R² = .1634, F = 23.37, p < .01) and all the control variables are positively correlated with job involvement. In the second step, TCB and TSB, the two focal explanatory variables of the study, were entered into the equation. The new equation is significant at the .01 level (adjusted R² = .2802, F = 23.04, p < .01). The adjusted R² is larger than that obtained in the first step by .1168 (.2802 − .1634). This increment was shown to be significant at the .01 level using a partial F test, and is attributable to the two explanatory variables of TCB and TSB, when the effects of the control variables in the equation on job involvement are controlled for. TCB is positively and significantly related to job involvement (β = .21, t = 4.28, p < .01). Using r (Nakagawa & Cuthill, 2007), which is a partial correlation coefficient and an effect size measure, the effect size is .228, in the small-to-medium range, according to the conventions proposed by Cohen (1988) for distinguishing among small, medium, and large effects (r = .1, .3, and .5, respectively). This suggests that a higher degree of TCB results in a higher perception of job involvement, which supports Hypothesis 1. The standardized regression coefficient of TSB is positive and significant (β = .14, t = 2.34, p < .01), thus supporting Hypothesis 2. The effect size of TSB is of less magnitude with r = .127 than that of TCB, a small effect size.

We also note that the standardized regression coefficient and effect size associated with TCB are higher than those associated with all the three actual job characteristics (β for skill variety is .14; task identity, .11; and feedback, .10; r for skill variety is .125; task identity, .102; and feedback, .104) and the actual supervisory behavior (β = .15, r = .171), and that the regression coefficient and effect size associated with TSB are higher than those associated with the two job characteristics of task identity and feedback. For testing whether these regression coefficients are equal, t tests (Greene, 2011) were conducted. The values of the test statistic, t, are computed by dividing the difference between two estimated regression coefficients (β) by the standard error for the difference. That is, $t = (β_{i} - β_{j}) / \sqrt{(C_{i i} + C_{j j} + 2 C_{i j})}$ . The quantity C_ii and C_jj are the variances of β_i and β_j, respectively. The quantity C_ij is the covariance of β_i and β_j. The t statistic has (n − k − 1) degrees of freedom. The quantity n is sample size and k is the number of unknown model parameters (i.e., the regression coefficients associated with independent variables) in the regression model. The results show that the coefficient for TCB is significantly greater than those for skill variety (t = 4.14), task identity (t = 4.25), feedback (t = 6.52) and Leadership P-C (t = 4.98) at the .01 level. This indicates that TCB is significantly more important than job characteristics and leadership behavior in predicting tax collectors’ job involvement. The coefficient for TSB is significantly greater than those for task identity (t = 1.84) and feedback (t = 1.69) at the .10 level. This suggests that TSB is significantly more important than task identity and feedback in predicting tax collectors’ job involvement.

Discussion and Conclusion

Before we summarize the contributions of our article, it is worth noting its limitations. The findings of our article may be limited to our particular sample of tax service employees. Taxpayers are forced to receive tax services and are hence different from customers who voluntarily choose to receive services, such as customers of public libraries, health, or human services (Beckett, 2000). However, tax collectors, like the employees of law enforcement agencies (e.g., police departments or the judicial system), have a mission to fight crime (Alm, 2012), which makes them different from employees of regular service organizations. Therefore, this research may not be generalized to all public service organizations.

For all of these shortcomings, the results do advance our understanding of the impact of coproductive citizens’ supervisory behaviors on the job involvement of public employees in a number of ways. First, the CTS measure developed in this study suggests that tax collectors would perceive coproductive taxpayers as their supervisors if the taxpayers challenge (the TCB behavior) or support (the TSB behavior) the tax collectors during their participation in the production and delivery processes of tax services. The CTS measure can open up an avenue for the study of the type of citizens’ coproductive behaviors and the influences of such types of coproductive behaviors on the behaviors and attitudes of public-sector employees. For example, Bowen’s (1983) study shows that the good feelings provided by customers (i.e., supportive behavior) influences, and even substitutes for leaders in producing employee satisfaction, but that task guidance from customers is avoided by employees. However, the results of this study indicate that giving appropriate directions (i.e., applying appropriate tax regulations in arguing and questioning tax audit results) is one of the principal components of the TCB behavior, which is positively correlated with the job involvement of tax collectors. This suggests that it may be the appropriate directions, or that it may be when the giving of appropriate directions is perceived by employees as a part of customer challenge behavior that makes the task guidance provided by customers acceptable to employees, and thus, would stimulate employee job involvement. Other research can further inform our basic hypothesis testing the likely influences of the TCB and the TSB on other work motivation, such as job satisfaction and organizational commitment. In addition, other research can explore a typology for coproductive citizens’ behavior, which, to our knowledge, has received little attention from scholars in public management.

This study sheds some light on the relationship between job involvement and related constructs under larger and public organizations in the context of citizen coproduction, an area that hitherto has received little attention (Brown, 1996). The results of this study support the findings in existing literature that in public organizations, job characteristics and supervisory behaviors are less conductive to job involvement than other factors such as employees’ personal characteristics (e.g., public service motivation), because public organizations are not routinely subject to external market forces Thus, they are more likely structured as bureaucracies and decide employees’ rewards more on the basis of longevity and service to the organization over time, but less on performance and situational factors. This can lead public employees to believe job characteristics and supervisory behaviors do not facilitate goal achievement as a means to get valued rewards and need satisfaction (Alonso & Lewis, 2001; Ouchi, 1980). Lawler (1992) has suggested that factors perceived as conducive to valued rewards and need satisfaction are likely to be highly correlated with job involvement. The results of this study show that tax collectors’ work ethic endorsement is more important than any other explanatory variables (including three actual job characteristics, one actual supervisory behavior, and two taxpayers’ supervisory behaviors) in predicting job involvement, in terms of standardized regression coefficients shown in Table 3. What are new in this current study are both that TCB is more important than the three job characteristics and the actual supervisory behavior in predicting employees’ job involvement, and that TSB is more important than the two job characteristics of task identity and feedback (see Step 2 in Table 3). This suggests that both TCB and TSB are important citizen resources of meaningfulness and supportiveness to public employees that previous research on the antecedent variables of job involvement has neglected.

The results of this study also lend some support to the views and initiatives of many recent scholars (e.g., Alford, 1998, 2002; Bovaird, 2007; Osborne & Plastrik, 2000; Roberts, 2004) that citizens’ coproductive behaviors can be utilized as a public management resource by allowing citizens to participate in the production and delivery processes of public services. These scholars suggest that coproduction not only reduces the production cost of services, but also satisfies the citizens’ desire to become more involved and play a more important role in the service production and delivery processes, leading to increased actual and perceived accountability of the government. This study’s contribution to the field of coproduction is demonstrating that coproductive taxpayers’ supervisory behaviors can significantly increase tax collectors’ job involvement. The literature has shown that job involvement is positively related to employees’ psychological benefits and organizational competitiveness. Highly job-involved employees are more satisfied with their jobs, supervision, coworkers, and pay and promotion opportunities, and have higher organizational commitment and fewer turnover intentions, than less involved employees (Brown, 1996; Lawler, 1986; Lee, Carswell, & Allen, 2000; Tett & Meyer, 1993). Therefore, we can deduce that taxpayers’ coproductive behaviors not only increase job involvement, but also would indirectly enhance tax collectors’ performance and organizational productivity through increased job involvement. Take one of our sample sources, the TNTA, as an example. The TNTA moved from an enforcement paradigm to a new paradigm—the “service paradigm” in 1999 (TNTA, 1999). The enforcement paradigm views taxpayers as potential criminals (Allingham & Sandmo, 1972; Alm, 2012) and appreciates the use of legally sanctioned coercive powers to directly punish the dishonest behavior of tax evasion (Turner, 2005). In contrast, the service paradigm regards taxpayers as customers and provides them with tax services of high quality and efficiency. Therefore, its purpose is to collect taxes via encouraging taxpayers to participate in the production and delivery processes of tax services through assisting taxpayers in every step of their filing returns and paying taxes to fulfill their tax obligations, and ensuring a fair and reasonable tax service environment (Alm, 2012; Alm et al., 2010; Kirchler, 2007). Since the revolutionary transformation in 1999, the number of tax evasion cases enforced by the district offices of the Administrative Enforcement Agency has decreased from 12,647 in fiscal year 2006 to 7,478 in fiscal 2010, a decrease of 40.87% (TNTA, 2011). In addition, 98.2% of taxpayers are satisfied with the tax services they received from the TNTA in 2012 (TNTA, 2012b). One of the factors that may have led to this outcome is coproductive taxpayers’ supervisory behaviors, with tax collectors’ increased job involvement as a possible mediator in this causal relationship. Future research can focus on whether taxpayers’ coproductive behaviors can, through their effects on tax collectors’ job involvement, affect also the organizational performance as measured by, for example, the perceived service quality of taxpayers, the number of tax evasion cases enforced, and so on.

Bowen (1983) suggests that when supervisors fail to provide “good feelings” or when subordinates do not value the consideration of their supervisors, employees may also turn to customers as sources of “good feelings.” In the present study, although respondents’ perceived leadership P-C is high (with M = 4.53 on a 7-point Likert-type format; see Table 2) and the standardized regression coefficient of job involvement on perceived Leadership P-C is positive and significant, the positive regression coefficient of job involvement on TSB is still significant when controlling for the effects of the other explanatory variables including leadership P-C (see Table 3). This implies that even when supervisors provide high and desirable “good feelings,” employees still value the “good feelings” from customers.

According the national specificity argument presented by Child and Kieser (1979), different countries exhibit relatively distinct and persistent cultures. Culture norms and beliefs are powerful determinants of peoples’ perceptions, dispositions, and behaviors (Markus & Kitayama, 1991) and often moderate the effects of business practices (Child & Kieser, 1979). For example, Kim, Egan, and Moon (2013) found a significant difference in the effect of managerial coaching between public organizations in South Korea and the United States due to difference in power distance. Taiwan has been put into “large power distance and collectivist” group (Hofstede, 1991; Yoo, Rao, & Hong, 2006). Therefore, future research with empirical data collected from other countries (e.g., Mexico, the United States, or the Netherlands) might offer a better understanding of how diverse cultural dimensions (e.g., power distance and collectivism presented by Hofstede, 1991) moderate the effects of the CTS construct on public employees’ job involvement and can be used to assess the generalizability of the scales and hypotheses used in this study.

Finally, what does this all mean for the actual practice of tax service management? Our results offer some insights into how tax service managers might seek to foster job involvement among tax collectors through encouraging taxpayers’ coproductive behavior. According to the scales in this study, tax collectors have a higher job involvement if they perceive that taxpayers can provide appropriate directions, task identity, ability to detect tax evasion, conviction that taxpayers support their decisions, that taxpayers can provide tangible performance feedback, and that taxpayers follow their instructions in filing returns. Therefore, creating job characteristics and a working environment of this kind is helpful to increasing employees’ job involvement. In terms of the extent to which CTS behaviors explain the change in tax collectors’ job involvement, taxpayers’ challenge behavior is higher than taxpayers’ supportive behavior. This finding implies that the greatest effect of coproductive taxpayers’ supervisory behaviors is in their ability to guide tax collectors’ actions in the service process by providing challenges.

Footnotes

Appendix A

Appendix B

Appendix C

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) disclosed receipt of the following financial support for the research, authorship, and/or publication of this article: This work was supported by the National Science Council of Taiwan (Grant NSC 96-2416-H-034-007).

Author Biographies

Chin-Mei Chen is an associate professor of tax administration in the Department of Public Finance and Tax Administration, National Taipei College of Business in Taipei, Taiwan. She has published in the Journal of Public Administration Research and Theory and Total Quality Management. Her research interests include tax administration and public human resources management.

Pei-Ching Lee is an associate professor of public administration in Kinmen Learning Center, National Open University in Kinmen County, Taiwan. He recently published in the Journal of Public Administration Research and Theory. His research interest is public human resources management.

Chien-Heng Chou is a professor of marketing in the Department of International Business Administration, Chinese Culture University in Taipei, Taiwan. His research appears in the Journal of Public Administration Research and Theory, Journal of Marketing Research, Marketing Letters, and Total Quality Management. He studies consumer behavior and marketing research.

References

Alford

(1998). A public management road less travelled: Clients as co-production of public services. Australian Journal of Public Administration, 57, 128-137.

Alford

(2002). Why do public-sector clients coproduce? Toward a contingency theory. Administration & Society, 34, 32-56.

Allingham

M. G.

Sandmo

(1972). Income tax evasion: A theoretical analysis. Journal of Public Economics, 1, 323-338.

Alm

(2012). Measuring, explaining, and controlling tax evasion: Lessons from theory, experiments, and field studies. International Tax and Public Finance, 19, 54-77.

Alm

Cherry

Jones

Mckee

(2010). Taxpayer information assistance services and tax compliance behavior. Journal of Economic Psychology, 31, 577-586.

Alonso

Lewis

G. B.

(2001). Public service motivation and job performance: Evidence from the federal sector. American Review of Public Administration, 31, 363-380.

Anderson

J. C.

Gerbing

D. W.

(1988). Structural equation modeling in practice: A review and recommended two-step approach. Psychological Bulletin, 103, 411-423.

Andrews

Boyne

G. A.

Meier

K. K.

O’Toole

L. J.

Walker

(2007, August). Differences between archival and perceptual environmental measure in the public sector. Paper presented at the Academy of Management Conference, Philadelphia, PA.

Bagozzi

R. P.

Phillips

L. W.

(1991). Assessing construct validity in organizational research. Administrative Science Quarterly, 36, 421-458.

10.

Beckett

(2000). The “government should run like a business” mantra. American Review of Public Administration, 30, 185-204.

11.

Bommer

W. H.

Johnson

J. L.

Rich

G. A.

Podsakoff

P. M.

Mackenzie

S. B.

(1995). On the interchangeability of objective and subjective measures of employee performance: A meta-analysis. Personnel Psychology, 48, 587-605.

12.

Bovaird

(2007). Beyond engagement and participation: User and community coproduction of public services. Public Administration Review, 67, 846-860.

13.

Bowen

D. E.

(1983). Customers as substitutes for leadership in service organizations: Their role as non-leader sources of guidance and support (Unpublished doctoral dissertation). Michigan State University, Ann Arbor.

14.

Boyd

Dess

Rasheed

(1993). Divergence between archival and perceptual measures of the environment: Causes and consequences. Academy of Management Review, 18, 204-206.

15.

Braithwaite

(2009). Defiance in taxation and governance: Resisting and dismissing authority in a democracy. Cheltenham, UK: Edward Elgar.

16.

Brislin

(1980). Translation and content analysis of oral and written material. In Triandis

H. C.

Berry

J. W.

(Eds.), Handbook of cross-culture psychology (Vol. 2, pp. 389-444). Boston, MA: Allyn & Bacon.

17.

Brown

S. P.

(1996). A meta-analysis and review of organizational research on job involvement. Psychological Bulletin, 120, 235-255.

18.

Brown

S. P.

Leigh

T. W.

(1996). A new look at psychological climate and its relationship to job involvement, effort, and performance. Journal of Applied Psychology, 81, 358-368.

19.

Browne

M. W.

Cudeck

(1993). Alternative ways of assessing model fit. In Bollen

K. A.

Long

J. S.

(Eds.), Testing structural equation models (pp. 136-162). Newbury Park, CA: Sage.

20.

Child

Kieser

(1979). Organizational and managerial roles in British and West German. In Lammers

Hickson

(Eds.), Organizations alike and unlike: International interinstitutional studies in the sociology of organizations (pp. 251-271). London, England: Routledge.

21.

Churchill

G. A.

Jr. , (1979). A paradigm for developing better measures of marketing constructs. Journal of Marketing Research, 16, 64-73.

22.

Cohen

(1988). Statistical power analysis for the behavioral sciences. Hillsdale, NJ: Lawrence Erlbaum.

23.

Cohen

West

S. G.

Aiken

L. S.

(2003). Applied multiple regression/correlation analysis for the behavioral sciences. Mahwah, NJ: Lawrence Erlbaum.

24.

Craighead

C. W.

Ketchen

D. J.

Jr. Dunn

K. S.

Hult

G. G.

(2011). Addressing common method variance: Guidelines for survey research on information technology, operations, and supply chain management. IEEE Transactions on Engineering Management, 58, 578-588.

25.

De Lancer Julnes

Holzer

(2001). Promoting the utilization of performance measures in public organizations: An empirical study of factors affecting adoption and implementation. Public Administration Review, 61, 693-707.

26.

Denhardt

R. B.

(1995). Public administration: An action orientation. Orlando, FL: Harcourt Brace.

27.

Gerbing

D. W.

Anderson

J. C.

(1988). An updated paradigm for scale development incorporating unidimensionality and its assessment. Journal of Marketing Research, 25, 186-192.

28.

Giddens

(1984). The constitution of society: Introduction to the theory of structuration. Berkeley: University of California Press.

29.

Greene

W. H.

(2011). Econometric analysis. Upper Saddle River, NJ: Prentice Hall.

30.

Hackman

J. R.

Lawler

E. E.

III . (1971). Employee reactions to job characteristics. Journal of Applied Psychology Monograph, 55, 259-286.

31.

Hackman

J. R.

Oldham

G. R.

Janson

Purdy

(1975). A new strategy for job enrichment. California Management Review, 17, 57-71.

32.

Hair

J. F.

Black

W. C.

Babin

B. J.

Anderson

R. E.

(2010). Multivariate data analysis: A global perspective. Upper Saddle River, NJ: Prentice Hall.

33.

Hassan

(2012). Sources of professional employees’ job involvement: An empirical assessment in a government agency. Review of Public Personnel Administration. doi:10.1177/0734371X12460555

34.

Heise

D. R.

(1972). Employing nominal variables, induced variables, and block variables in path analysis. Sociological Research Methods, 1, 147-173.

35.

Hodgkinson

G. P.

(2001). Cognitive processes in strategic management: Some emerging trends and future directions. In Anderson

N. R.

Ones

Sinangil

H. K.

Viswesvaran

(Eds.), Handbook of industrial, work and organizational psychology: Organizational psychology (Vol. 2, pp. 416-440). London, England: Sage.

36.

Hofstede

(1991). Cultures and organizations: Software of the mind. New York, NY: McGraw-Hill.

37.

House

R. J.

Dessler

(1974). The path-goal theory of leadership: Some post hoc and a priori tests. In Hunt

J. G.

Larson

L. L.

(Eds.), Contingency approaches to leadership (pp. 29-55). Carbondale: Southern Illinois University Press.

38.

Hoyle

R. H.

Panter

A. T.

(1995). Writing about structural equation models. In Hoyle

R. H.

(Ed.), Structural equation modeling: Concepts, issues, and applications (pp. 76-99). Thousand Oaks, CA: Sage.

39.

Hsieh

A. T.

Chou

C. H.

Chen

C. M.

(2002). Job standardization and service quality: A closer look at the application of total quality management to the public sector. Total Quality Management, 13, 899-912.

40.

Jöreskog

K. G.

(1969). A general approach to confirmatory maximum likelihood factor analysis. Psychometrika, 34, 183-202.

41.

Jöreskog

K. G.

Sörbom

(2001). LISREL 8. Lincolnwood, IL: Scientific Software International.

42.

Kanungo

R. N.

(1982). Work alienation: An integrative approach. New York, NY: Praeger.

43.

Kim

Egan

T. M.

Moon

M. J.

(2013). Managerial coaching efficacy, work-related attitudes, and performance in public organizations: A comparative international study. Review of Public Personnel Administration. doi:10.1177/0734371X13491120

44.

Kirchler

(2007). The economic psychology of tax behavior. Cambridge, UK: Cambridge University Press.

45.

Klingner

D. E.

Nalbandian

(2003). Public personnel management: Contexts and strategies. Upper Saddle River, NJ: Prentice Hall.

46.

Lawler

E. E.

III . (1986). High-involvement management: Participative strategies for improving organizational performance. San Francisco, CA: Jossey-Bass.

47.

Lawler

E. E.

III . (1992). The ultimate advantage: Creating the high-involvement organization. San Francisco, CA: Jossey-Bass.

48.

Lee

Carswell

J. J.

Allen

N. J.

(2000). A meta-analytic review of occupational commitment: Relations with person- and work-related variables. Journal of Applied Psychology, 85, 799-811.

49.

Lee

P. C.

Chen

C. M.

Chou

C. H.

(2013). Decreasing tax collectors’ perceived social loafing through collaborative behaviors of taxpayers. Journal of Public Administration Research and Theory. doi:10.1093/jopart/mut015

50.

Lewis

(1982). The psychology of taxation. Oxford, UK: Martin Robertson.

51.

Liou

K. T.

Bazemore

(1994). Professional orientation and job involvement among detention case workers. Public Administration Quarterly, 18, 223-236.

52.

MacKenzie

S. B.

Podsakoff

P. M.

Paine

J. B.

(1999). Do citizenship behaviors matter more for managers than for salespeople? Journal of the Academy of Marketing Science, 27, 396-410.

53.

Markus

Kitayama

(1991). Cultural and the self: Implications for cognition, emotion, and motivation. Psychological Review, 98, 224-253.

54.

Mezias

J. M.

Starbuck

W. H.

(2003). The odyssey continues. British Journal of Management, 14, 45-47.

55.

Miller

(1988). Relating porter’s business strategies to environment and structure: Analysis and performance implications. Academy of Management Journal, 31, 280-308.

56.

Mirels

H. L.

Garrett

J. B.

(1971). The Protestant work ethic as a personality variable. Journal of Consulting and Clinical Psychology, 36, 40-44.

57.

Moynihan

D. P.

Pandey

S. K.

(2007). Finding workable levers over work motivation: Comparing job satisfaction, job involvement, and organizational commitment. Administration & Society, 39, 803-832.

58.

Nakagawa

Cuthill

I. C.

(2007). Effect size, confidence interval and statistical significance: A practical guide for biologists. Biological Reviews, 82, 591-605.

59.

Nunnally

J. C.

Bernstein

I. H.

(1994). Psychometric theory. New York, NY: McGraw-Hill.

60.

Osborne

Plastrik

(2000). The reinventor’s fieldbook: Tools for transforming your government. New York, NY: Jossey-Bass.

61.

Ouchi

W. G.

(1980). Markets, bureaucracies, and clans. Administrative Science Quarterly, 25, 129-141.

62.

Paarlberg

L. E.

(2007). The impact of customer orientation on government employee performance. International Public Management Journal, 10, 201-231.

63.

Pandey

S. K.

Wright

B. E.

(2006). Connecting the dots in public management: Political environment, organizational ambiguity and the public manager’s role ambiguity. Journal of Public Administration Research and Theory, 16, 511-532.

64.

Payne

A. F.

Storbacka

Frow

(2008). Managing the co-creation of value. Journal of the Academy of Marketing Science, 36, 83-96.

65.

Perry

J. L.

Wise

L. R.

(1990). The motivational bases of public service. Public Administration Review, 50, 367-373.

66.

Podsakoff

P. M.

MacKenzie

S. B.

Lee

J. Y.

Podsakoff

N. P.

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

67.

Podsakoff

P. M.

Organ

D. W.

(1986). Self-reports in organizational research: Problems and prospects. Journal of Management, 12, 531-544.

68.

Potoski

Woods

N. D.

(2001). Designing state clean air agencies: Administrative procedures and bureaucratic autonomy. Journal of Public Administration Research and Theory, 11, 203-221.

69.

Robbins

S. P.

(2005). Organizational behavior. Upper Saddle River, NJ: Pearson Prentice Hall.

70.

Roberts

(2004). Public deliberation in an age of direct citizen participation. American Review of Public Administration, 34, 315-353.

71.

Rosen

H. S.

Gayer

(2010). Public finance. New York, NY: McGraw-Hill.

72.

Rosenberg

(1979). Conceiving the self. New York, NY: Basic Books.

73.

Schachter

H. L.

(1995). Reinventing government or reinventing ourselves: Two models for improving government performance. Public Administration Review, 55, 530-537.

74.

Schneider

Bowen

D. E.

(1995). Winning the service game. Boston, MA: Harvard Business School Press.

75.

Sharp

E. B.

(1980). Toward a new understanding of urban services and citizen participation: The coproduction concept. Midwest Review of Public Administration, 14, 105-118

76.

Simon

H. A.

(1953). Birth of an organization: The economic cooperation administration. Public Administration Review, 13, 227-236.

77.

Sims

H. P.

Szilagyi

A. D.

Keller

R. T.

(1976). The measurement of job characteristics. Academy of Management Journal, 19, 195-212.

78.

Tabachnick

B. G.

Fidell

L. S.

(2007). Using multivariate statistics. Boston, MA: Allyn & Bacon.

79.

Taipei National Tax Administration, Ministry of Finance of R.O.C. (1999). National Tax Administration News, 6, 12-15.

80.

Taipei National Tax Administration, Ministry of Finance of R.O.C. (2011). Yearbook of national tax statistics. Taipei, Taiwan: Author.

81.

Taipei National Tax Administration, Ministry of Finance of R.O.C. (2012a). Public service white papers. Taipei, Taiwan: Author.

82.

Taipei National Tax Administration, Ministry of Finance of R.O.C. (2012b). 2012 annual report on taxpayers’ opinion of tax services. Taipei, Taiwan: Author.

83.

Taipei Revenue Service. (2012). The introduction of Taipei Revenue Service. Taipei, Taiwan: Author.

84.

Tett

R. P.

Meyer

J. P.

(1993). Job satisfaction, organizational commitment, turnover intention, and turnover: Path analyses based on meta-analytic findings. Personnel Psychology, 46, 259-293.

85.

Turner

J. C.

(2005). Explaining the nature of power: A three-process theory. European Journal of Social Psychology, 35, 1-22.

86.

Vigoda

(2002). From responsiveness to collaboration: Governance, citizens, and next generation of public administration. Public Administration Review, 62, 527-540.

87.

Wang

Berman

(2001). Hypotheses about performance measurement in countries: Findings from a survey. Journal of Public Administration Research and Theory, 11, 403-428.

88.

Weick

(1995). Sensemaking in organizations. Thousand Oaks, CA: Sage.

89.

Wheaton

(1987). Assessment of fit in over-identified models with latent variables. Sociological Methods & Research, 16, 118-154.

90.

Wheaton

Muthën

Alvin

D. F.

Summers

G. F.

(1977). Assessing reliability and stability in panel models. In Herse

D. R.

(Ed.), Sociological methodology (pp. 84-136). San Francisco, CA: Jossey-Bass.

91.

Whicker

M. L.

Strickland

R. A.

Olshfski

(1993). The troublesome cleft: Public administration and political science. Public Administration Review, 53, 531-541.

92.

Wright

B. E.

(2001). Public-sector work motivation: A review of the current literature and a revised conceptual model. Journal of Public Administration Research and Theory, 11, 559-586.

93.

Yang

Pandey

S. K.

(2009). How do perceived political environment and administrative reform affect employee commitment? Journal of Public Administration Research and Theory, 19, 335-360.

94.

Yoo

D. K.

Rao

S. S.

Hong

(2006). A comparative study on cultural difference and quality practices—Korea, USA, Mexico, and Taiwan. International Journal of Quality & Reliability Management, 23, 607-624.