“Giving Something Back to Society”: A Study Exploring the Role of Reciprocity Norms in Public Service Motivation

Abstract

Public service motivation (PSM) is often conceived as a predominantly other-concerned type of motivation. This study challenges this notion by investigating the role of reciprocity norms, which are rooted in self-concern, as a determinant of PSM. Taking up a debate on whether PSM may be a manifestation of reciprocity vis-à-vis society, I draw on a combination of psychological contract theory and social identity theory to substantiate previous theoretical linkages between these concepts. Using a sample of 205 Swiss police officers, I then empirically analyze the relationship between both positive and negative reciprocity norms and PSM. Thereby, I utilize societal identification, which is a prerequisite for the emergence of shared psychological contracts, as a moderator. Findings from moderated regression analyses support the theoretical arguments for the norm-based and affective but not the rational components of PSM. In sum, this study provides relevant insights into some of the self-concerned foundations of PSM.

Keywords

public service motivation reciprocity norms police psychological contracts social identity theory

Introduction

The provision of public services is heavily dependent on the people who work in government. With public employees being the central driver of government performance, it is of paramount importance that they are well motivated to advance the public weal (Rainey, 2003). Thus, it comes as no surprise that the concept of public service motivation (PSM), first proposed by Perry and Wise (1990) to better explain the specificities of motivation in public work, has been one of the most prevalent concepts in recent public administration research (Ritz, Brewer, & Neumann, 2016). Defined as “a general altruistic motivation to serve the interests of a community of people, a state, a nation, or humankind” (Rainey & Steinbauer, 1999, p. 24), PSM is a type of prosocial motivation (Bolino & Grant, 2016). As with other forms of prosocial motivation, the core of PSM is a “desire to expend effort to benefit other people” (Grant, 2008, p. 49), whereas the peculiarity of PSM is its embeddedness in the context of public institutions (Bolino & Grant, 2016). Importantly, prosocial motivations such as PSM may either be self-concerned, other-concerned, or both at the same time (De Dreu & Nauta, 2009; Kim & Vandenabeele, 2010). However, most previous PSM research has not devoted much attention to said self-oriented aspects as the roots of PSM lie in other-orientation (Perry, Hondeghem, & Wise, 2010). This empirical study sheds light on an important and predominantly self-concerned determinant of PSM by analyzing the role of reciprocity norms vis-à-vis society in PSM. As reciprocal exchanges in today’s world are usually upheld for the personal benefits each party derives from them (Godbout, 1992; Nowak & Sigmund, 2005), this study thereby questions the common notion that PSM is a mainly selfless type of motivation.

While situations of reciprocal exchange, both between two individuals and within collectives small enough to allow for the monitoring of other members, have been researched extensively and in various disciplines (see, for example, Cropanzano & Mitchell, 2005; Falk & Fischbacher, 2006; Nowak & Sigmund, 2005; Rousseau, 1995; Whatley, Webster, Smith, & Rhodes, 1999), to date, little scholarly attention has been devoted to acts of reciprocity vis-à-vis anonymous collectives such as a society. Some notable exceptions include Fehr and Fischbacher (2003), Fehr and Henrich (2003) and Gintis, Bowles, Boyd, and Fehr (2003) who termed the willingness to perform altruistic acts in anonymous, one-shot situations strong reciprocity, highlighting that such generosity is not always likely to be returned. In public administration literature, Koehler and Rainey (2008) have in a similar vein brought up the question of whether there exists a norm of reciprocity that is effective vis-à-vis a society or nation, stressing that “an individual may feel obligated to offer civic service as to compensate his or her country for benefits received as a citizen” (Koehler & Rainey, 2008, p. 43). They argue that an individual’s repayment to society may take the form of increased service-oriented motivation, or PSM. However, an empirical test of Koehler and Rainey’s (2008) theoretical argument has not yet been carried out despite the considerable relevance it may hold both for researchers who strive to understand the determinants of PSM, and practitioners seeking to motivate public employees to serve the public interest.

The present study fills this void by investigating the research question of whether internalized reciprocity norms may manifest themselves vis-à-vis society and thereby play a role in a person’s levels of PSM. Taking up the debate initiated by Koehler and Rainey (2008), I draw on theory on shared psychological contracts and on social identity theory to further underpin the notion that PSM may in part be rooted in reciprocity norms. Subsequently, I utilize survey data from a sample of state police in Switzerland and analyze it using moderated multiple regression in an attempt answer the research question.

The remainder of this study is organized as follows. The “Literature Review” section provides a review of the literature on reciprocity norms and PSM. The “Theory and Hypotheses” section outlines relevant theoretical perspectives on reciprocity vis-à-vis society and presents the hypotheses. The “Sample, Measures, and Method” section details the sample, measures, and methods used. The results from the analyses are presented in the “Results” section. In the “Discussion” section, I discuss these results in light of the hypotheses before drawing some conclusions in the final section.

Literature Review

While a few empirical studies on PSM have been published that peripherally tap into the theme of reciprocity, to date, the role of reciprocity norms in PSM has mostly been discussed theoretically. Taylor (2007) and Steen (2006), for instance, have both drawn on the work of Le Grand (2003) to categorize three different types of public employees: knaves, act-irrelevant knights, and act-relevant knights. Each of these types has a different motivation to perform public service. Knaves engage in public service based on self-interested motives, while knights do so without tangible rewards or even at personal cost. Knights can be further distinguished into act-relevant knights who have a desire to benefit others out of feelings of reciprocity or obligation, and act-irrelevant knights who act out of feelings of compassion or a sense of injustice. Taylor (2007) argues that PSM is likely based on a mix of all three motives, which suggests that it should be regarded as a not purely selfless or other-oriented prosocial motivation (Steen, 2006; see also De Dreu & Nauta, 2009). A similar attempt to discern different types of public servants has been made by Brewer, Selden, and Facer (2000). In their empirical study, these authors demonstrated that there exist various distinct conceptions of PSM in students and practitioners of public administration, using the Q-methodology. They identified four different groups based on their motives for performing public service: samaritans, patriots, humanitarians, and communitarians. While samaritans are predominantly motivated by opportunities to help others, patriots are driven by feelings of duty to work for the public good, and humanitarians are concerned with social justice and with making a difference in society. Last but not least, communitarians are motivated to perform public service to give something back to society. Communitarians additionally view citizenship as a special bond between themselves and others, reinforcing their personal wish to serve their community or country. Interestingly, the group of communitarians identified by Brewer et al. (2000) corresponds very closely to the act-relevant knights described in Taylor (2007) and Steen (2006), and the relevance that reciprocity has for both of these groups suggests that PSM is to some extent rooted in norms of reciprocity.

The notion that a relationship exists between PSM and reciprocity norms is further supported in a study by Houston (2011), who investigated whether well-developed welfare states crowd out their citizens’ norm of reciprocity, and thereby motives associated with PSM. Houston argues that comprehensive welfare programs for the underprivileged reduce the social obligations of individual citizens because they may feel that their reciprocal duties are already fulfilled by paying higher taxes to fund these programs. Coherent with this argument, he found that in well-developed welfare states such as in Scandinavia, individuals tend to value work motives related to PSM less than individuals in less-developed welfare states such as the Anglo-Saxon ones. This finding adds to the relevance of analyzing the relationship between reciprocity norms and PSM in Switzerland, as will be done in this study, since Switzerland’s social expenditures and tax rates are among the lowest in the Organisation for Economic Co-Operation and Development (OECD, 2016).

While each of the above studies provides interesting insights into certain aspects of the role of reciprocity norms in PSM, this theme was not the main focus in any of them and the associated theoretical underpinnings have only peripherally been explored. The most extensive theoretical contribution focusing on the role of reciprocity norms in PSM so far has been provided by Koehler and Rainey (2008), who discuss the interdisciplinary foundations of PSM. Taking a sociobiological perspective on reciprocity norms as an antecedent of altruistic behaviors, they point out that the concept of reciprocal altruism may explain why both animals and humans altruistically help one another, even at personal cost. The key reason is that engaging in reciprocal altruism increases the chances of survival through cooperation when compared with egoistic behavior. However, it also requires both the ability to remember acts of altruism and the capacity to develop moral obligations to return altruistic behaviors. Expanding on this idea, Koehler and Rainey (2008) stress that networks of reciprocal obligations may arise in populations (animals or humans) that improve the fitness or well-being of the entire group, such that all individuals involved in the reciprocal exchanges benefit in the long run. It logically follows that group membership, although at times costly, is more beneficial than being on one’s own, motivating individuals in a self-concerned way to benefit the group and to give back to secure continued participation.

Moving to an organizational behavior perspective and drawing on Rousseau’s (1995) concept of psychological contracts, Koehler and Rainey (2008) highlight that similar reciprocal relationships may play a role between individuals and more anonymous collectives such as society or a nation. While psychological contracts are commonly used to describe norms of reciprocity between organizations and employees, Koehler and Rainey (2008) stress that the underlying values of such contracts often reflect those of the greater society. This means that similar contracts may also emerge between individuals and society as a whole. In this line of reasoning, individuals may develop feelings of duty to serve the public, or in other words PSM, as a means to repay society for benefits that they have previously received as members thereof, creating a mutually offsetting pattern of exchanges. Following Koehler and Rainey’s (2008) theoretical argumentation, I will expand upon the theme of reciprocity vis-à-vis society in the following section to develop the hypotheses of this study.

Theory and Hypotheses

When asking public employees about their motivations to work in the public service, a frequent answer is that they want to give something back to society (Brewer et al., 2000). Interestingly, the wish to give something back implies that one has previously received some sort of benefit or gesture of goodwill from the respective counterparty. In other words, certain public servants appear to conceive their daily work as part of a long-term reciprocal exchange of altruism with society. Note that this study relies on a behavioral definition of altruism, as opposed to a psychological one (see Fehr & Fischbacher, 2003), whereas PSM as a motivation is viewed as an antecedent of altruism (see Schott, Neumann, & Baertschi, 2016). More specifically, altruism is defined as “conferring a benefit b on the recipient at a cost c to the donor” (Nowak & Sigmund, 2005, p. 1291) without referring to motivational aspects. In the following and in regard to the research question, the theoretical mechanisms and motives behind reciprocal exchanges of altruism will be delineated on the basis of the concepts of direct reciprocity, indirect reciprocity, and strong reciprocity. Thereby, the key focus will be on strong reciprocity as this concept captures an extraordinary aspect in the desire to give something back to society expressed by many public employees (Brewer et al., 2000): Here, the counterparty in the reciprocal exchange is not a clearly defined individual or small group but instead society at large.

Generally, we may say that relationships among humans are frequently characterized by cooperative altruistic exchanges. Nineteenth-century German sociologist Georg Simmel even went so far as to stating that “all contacts among men rest on the schema of giving and returning the equivalence” (Simmel, 1950, p. 387). The motivation to repay acts of altruism is thereby rooted in social norms of reciprocity that facilitate such social exchanges (Perugini, Gallucci, Presaghi, & Ercolani, 2003; Whatley et al., 1999). The most basic form of reciprocal exchange takes place between the same two individuals in a repeated fashion and is probably best illustrated by the principle I’ll-scratch-your-back-if-you’ll-scratch-mine (Binmore, 1994, p. 25). Termed direct reciprocity, this type of exchange has in more recent years been explained using game theory: Assuming that the cost of providing an act of altruism is smaller than the benefit of receiving it, repeated direct reciprocal exchanges will allow both parties to obtain a net benefit. Thus, direct reciprocity constitutes a case of the repeated Prisoner’s Dilemma game in which a strategy of continued cooperation yields a higher payoff than defecting (Nowak & Sigmund, 2005). This higher payoff is explained by the reaction to defecting by the counterparty who would punish the defector by terminating the mutually beneficial cooperation.

An extension of direct reciprocity that applies to situations of reciprocal exchange in groups is indirect reciprocity (Alexander, 1987). Here, the exchange of altruistic acts is between randomly changing dyads of group members instead of between the same two individuals in a repeated fashion. Accordingly, indirect reciprocity follows the principle I’ll-scratch-someone-else’s-back-if-you’ll-scratch-mine (Binmore, 1994, p. 25). Without repeated exchange between the same two individuals, the possibility to personally hold defectors accountable by terminating the cooperation is nonexistent. Thus, as Nowak and Sigmund (2005) stress, enforcement needs to be carried out by the community as a whole. A prerequisite for such community enforcement, in turn, is the ability to monitor all individuals within the group. Observers of acts of cheating need to inform the population of the defector’s identity, and gossip will lead to a decrease in his or her reputation. As a consequence, the defector will be excluded from future participation in reciprocal exchanges with the group, which constitutes the punishment (Nowak & Sigmund, 2005). Assuming that the cost of providing an act of altruism is smaller than the benefit of receiving it, as we did previously regarding direct reciprocity, providing acts of altruism to the group will yield a return on investment and thereby motivate individuals to comply.

What is harder to make sense of based on the principles of reciprocity, however, are acts of altruism toward other group members in situations when neither personal nor community enforcement are possible. Such situations predominantly include large and anonymous collectives such as societies, and they are limited to human populations (Fehr & Fischbacher, 2003). Nevertheless, as the literature on strong reciprocity (e.g., Fehr & Henrich, 2003; Gintis, 2000; Gintis et al., 2003) and on PSM (e.g., Anderfuhren-Biget, Varone, Giauque, & Ritz, 2010; Brewer et al., 2000; Perry et al., 2010; Perry & Wise, 1990; Rainey & Steinbauer, 1999) suggests, such forms of altruism do exist. This brings up the question of whether certain individuals also adhere to the principle I’ll-scratch-your-back-regardless-of-whether-someone-else-will-scratch-mine, knowing that reputation building is no longer a means of securing repayment as, first, anonymity hinders the identification of donors and defectors in society and, second, moral hazards (Holström, 1979) lead many members of society to exploit altruism by not reciprocating in the absence of monitoring and enforcement (see also Nowak & Sigmund, 2005). Coherent with this line of argument, Gintis et al. (2003) described strong reciprocity as “a predisposition to cooperate with others and to punish those who violate the norms of cooperation, at personal cost, even when it is implausible to expect that these costs will be repaid” (p. 153). This definition also takes up the distinction between positive and negative reciprocity: While the former concerns reciprocating positive behaviors with positive behaviors, the latter regards taking revenge for or punishing negative behaviors (Matejkowski, McCarthy, & Draine, 2011). Gintis et al. (2003) go on arguing that strong reciprocity may be rooted in coevolutionary processes among a certain group of individuals that have led to strong reciprocity becoming a stable evolutionary strategy. Several scholars have provided experimental evidence of strong reciprocity (e.g., Fehr, Fischbacher, & Gächter, 2002; Fehr & Henrich, 2003; Gintis et al., 2003; McCabe, Rassenti, & Smith, 1998), although elements of their methodology as well as the claim regarding the coevolutionary processes have been criticized (Haley & Fessler, 2005). But even these critics did “not rule out the possibility that humans possess strong reciprocity motives” (Haley & Fessler, 2005, p. 247).

This brings us back to Koehler and Rainey (2008) who provide an alternative explanation for the emergence of strong reciprocity. Their argument that reciprocal obligations may play a role between individuals and society draws on the theory of psychological contracts (Rousseau, 1995) which originally focused on the relationship between employee and employing organization. In this theory, the understanding of a contract goes beyond written terms and strict legal interpretations. Instead, informal agreements and the involved parties’ beliefs of what they have agreed to are central to the formation of mutual obligations. While the classic concept of psychological contracts relates to the dyadic relationship between a single individual and the organization, Rousseau (1995) also introduced a group level-extension of her theory named shared psychological contracts (or normative contracts). Such shared contracts “exist where the organization has many members who identify themselves in similar ways with it and each other [. . .] and these members believe themselves party to the same contract” (Rousseau, 1995, p. 9). Thus, an individual’s perceived obligations arising from shared psychological contracts are not mainly directed at the organization but more so at its other members. Furthermore, the more strongly an individual feels that he or she is bound to the shared psychological contract through identification, the more it will guide his or her course of action, making it a type of internalized social norm (Rousseau, 1995).

According to Koehler and Rainey (2008), psychological contracts may also exist between individuals and society as a whole, creating a reciprocal relationship that is beneficial to both parties. Individuals may therein feel an obligation to engage in civic service as a means to repay their country for the benefits it offers, as manifested by increased PSM. However, while Koehler and Rainey’s (2008) argument is compelling, they did not touch upon the role of identification with the group which Rousseau (1995) regarded as crucial in the emergence of shared psychological contracts. A framework to conceptualize such identification with a collective is provided by social identity theory (Tajfel, 1981, 2010). In the literature on this theory, social identity is often defined as “that part of the individuals’ self-concept which derives from their knowledge of their membership of a social group (or groups) together with the value and emotional significance attached to that membership” (Tajfel, 1981, p. 255). Thereby, in the case of a shared psychological contract among the members of a society, one likely source of common identification is a political entity such as a nation or a state (Mummendey, Klink, & Brown, 2001), as is also illustrated by the description of the communitarian public servant in Brewer et al. (2000). This assertion holds particularly within the realms of public work and policing which by their nature are associated with a certain political entity (Vandenabeele, 2007) and are subject to professional rules, norms and hierarchical structures corresponding to that entity. As such, societal identification enables individuals to categorize themselves into the social group they feel they belong to, the so-called in-group, and to distance themselves from what they perceive as out-groups (Mummendey et al., 2001). Accordingly, societal identification allows individuals to get a graspable picture of who the other members of their shared psychological contract are. However, identification is not based solely on nationality but also on cultural factors such as shared values, norms, attitudes, and goals that lead to social interaction and cohesion within a society and create a network of social interdependence (Tajfel, 2010).

Thus, addressing the theoretical void regarding societal identification in shared psychological contracts between individuals and societies, I argue that reciprocity norms alone are insufficient to explain strong reciprocity in the form of PSM in a person. What is required in addition is a strong identification with society, enabling individuals to trace out the other in-group members of their shared psychological contract. Hence, PSM should become particularly manifest in the presence of an interaction between internalized reciprocity norms and societal identification. Furthermore, recalling the dichotomous definition of strong reciprocity by Gintis et al. (2003), the described moderation should be effective regardless of whether positive or negative reciprocity is concerned. However, the interaction between negative reciprocity and societal identification should be associated with a decrease in PSM levels, as opposed to an increase. The above considerations lead us to the following three hypotheses.

Hypothesis 1: Neither positive nor negative reciprocity norms will be related to PSM.

Hypothesis 2a: The interaction between positive reciprocity norms and societal identification will be positively related to PSM.

Hypothesis 2b: The interaction between negative reciprocity norms and societal identification will be negatively related to PSM.

Finally, several authors (e.g., Andersen & Serritzlew, 2012; Taylor, 2007) have recommended analyzing the subdimensions of PSM separately since their relationships with other variables may vary given that each dimension taps into certain specific motives to benefit society. Heeding this recommendation and drawing on Perry’s (1996) distinction between the rational, norm-based, and affective motives that constitute PSM, I argue that the norm-based dimension of commitment to the public interest (CPI) and the affective dimensions of compassion (COM) and self-sacrifice (SSF) should exhibit greater associations with the interactions described in Hypotheses 2a and 2b than the rational dimension of attraction to policy making (APM). There are two reasons for this. First, CPI, COM, and SSF provide by their focus on benefiting others more relevant motives related to reciprocating vis-à-vis society than the rational APM dimension (see Ritz, 2011). Second, regarding CPI, norm-based motives arise from the wish to conform to norms (Perry, 1996), which is also central both to shared psychological contracts and strong reciprocity and thus may be of particular importance to individuals inclined to strongly internalize norms. The proposition regarding the varying effects on the PSM dimensions is captured by Hypothesis 3. In the following, Hypotheses 1 to 3 will be subjected to an empirical test.

Hypothesis 3: The interaction effects postulated by Hypotheses 2a and 2b will be stronger for the norm-based (CPI) and affective (COM, SSF) dimensions than for the rational (APM) dimension of PSM.

Sample, Measures, and Method

Sample

Data for analysis were sampled from a midsized state police corps in central Switzerland. In this predominantly rural region, both the crime rates and the police officer to population ratio are well below the national average. Studying reciprocity norms vis-à-vis society and their relationship with PSM on a police sample appears particularly interesting as most police officers have extensive street-level experience. Thus, although much of their daily work is strictly regulated by law, they also have frequent opportunities to use administrative discretion to very directly benefit members of society—or in other words to give something back (Giauque, Ritz, Varone, & Anderfuhren-Biget, 2012). Further underscoring the research question’s relevance in the context of this study, the organizational culture of the police corps from which the data were drawn is such that it allows for a relatively high level of administrative discretion when compared with similar corps in Switzerland. In July 2015, an online survey was distributed via email to all 246 police officers working at the corps. The invitation and a later reminder were sent out in cooperation with the chief of staff, which led to a response rate of 83.3% based on started surveys (205 participants) or 78.0% when counting survey completions (192 participants). The sample included officers from all departments (police command, staff department, criminal investigation, security police, legal department, and prevention and communication) and hierarchical levels within the corps. Descriptive statistics on the sample’s demographic characteristics are presented in Table 1.

Table 1.

Descriptive Statistics and Spearman Correlation Coefficients.

Variable	M	SD	1	2	3	4	5	6	7	8	9	10	11	12	13	14
PSM-APM (1)	.00	1.35	1
PSM-COM (2)	.00	.77	.40***	1
PSM-CPI/SSF (3)	.00	.57	.63***	.66***	1
Reciprocity positive (4)	.00	.68	−.13^†	−.02	.00	1
Reciprocity negative (5)	.00	1.02	.00	−.02	−.13^†	.49***	1
Societal identification (11-pt. scale) (6)	7.91	2.04	.09	−.02	−.05	.11	.33***	1
Age (7)	40.13	10.20	.29***	.27***	.22**	−.26***	−.13^†	−.05	1
Male (8)	1.86	.35	.27***	.07	.19**	−.04	.03	.13^†	.31***	1
Children (dichotomous) (9)	1.41	.49	−.18*	−.26***	−.18*	.17*	.02	−.01	−.55***	−.32***	1
Preference for urban life (7-pt. scale) (10)	2.59	1.63	.01	.03	.03	−.03	−.08	−.26***	−.13^†	.05	.04	1
Work hours per week (in % of max.) (11)	97.74	9.65	.00	−.08	.04	.03	.05	−.03	−.08	.31***	.10	.13^†	1
Tenure at police (12)	15.10	10.53	.22**	.21**	.16*	−.22**	−.07	−.01	.91***	.32***	−.51***	−.07	−.07	1
Managerial position (dichotomous) (13)	1.24	.43	.22**	.15*	.19**	−.16*	−.04	.02	.36***	.19**	−.26***	.05	.10	.42***	1
Overall job satisfaction (7-pt. scale) (14)	5.53	1.01	.14*	.09	.27***	.03	−.10	−.13^†	.09	−.02	−.09	−.16*	−.10	.01	.08	1

Note. N = 194; listwise deletion of missings; variables 1 to 5 are regression factor scores. PSM = public service motivation; APM = attraction to policy making; COM = compassion; CPI/SSF = commitment to the public interest/self-sacrifice.

†

p < .10. *p < .05. **p < .01. ***p < .001.

Measures

Reciprocity norms were measured using a shortened version of the Personal Norm of Reciprocity (PNR) scale developed by Matejkowski et al. (2011). Each four items served to measure positive and negative reciprocity norms, respectively. The PNR dimensions of positive and negative reciprocity capture the tangible, behavioral aspects of reciprocity (Perugini et al., 2003) and therefore are most central to the motivation to reciprocate vis-à-vis society. The third dimension of the PNR, beliefs in reciprocity, was not used as it refers to the belief that reciprocity is usually adhered to by others (Matejkowski et al., 2011), which is less relevant in this study. PNR items were translated into German and back-translated into English by a second person to improve translation accuracy.

PSM was measured using an abridged and adapted version of the German language scale by Giauque, Ritz, Varone, Anderfuhren-Biget, and Waldner (2011) designed to capture the specificities of the Swiss context. This PSM scale included a total of 12 items, measuring PSM on the four dimensions originally suggested by Perry (1996): APM, SSF, COM, and CPI. Thus, each dimension was measured using three items.

To assess the validity of the two above measurement constructs given the study sample, a confirmatory factor analysis (CFA) was conducted (see Table A1 in the Appendix). After removing a low-loading, reverse-coded item of the PSM dimension compassion, the initial measurement model fit the data very well (root mean square error of approximation [RMSEA] = .039; comparative fit index [CFI] = .959; Tucker–Lewis Index [TLI] = .948), χ²(135) = 175.411, p < .05), judging by a combination of fit indices recommended by Jackson, Gillaspy, and Purc-Stephenson (2009). However, while estimating this first model including both the two dimensions of reciprocity norms and the four dimensions of PSM, a warning message indicated that the covariance matrix was not positive definite. This issue could be traced back to the PSM dimensions of CPI and SSF being collinear and thus undistinguishable (Lei & Wu, 2007). Therefore, following a recommendation by Kline (2011), I decided to respecify the model by merging these two dimensions into a single dimension named CPI/SSF. In addition, the lowest-loading item from the positive reciprocity scale was deleted to improve the convergent validity of this particular scale. The respecified model had similarly good model fit as the initial one (RMSEA = .042; CFI = .956; TLI = .945), χ²(123) = 165.802, p < .01), and its parameters were estimated without any warning messages. To assess convergent validity, I calculated the average variance extracted (AVE) for each dimension of both PNR and PSM. The PSM dimension of APM and the two dimensions of PNR exhibited AVE values around .5 or higher (see bottom of Appendix Table A1), indicating good convergent validity. The PSM dimensions COM and CPI/SSF had lower values at .35 and .25, respectively, meaning that results in subsequent analyses involving these variables should be interpreted with caution. Calculating Raykov’s omega for each dimension yielded values between .92 and .65, indicating that composite reliability ranged from excellent to good. The only exception was the two-item measure of COM, whose value at .51 was still deemed acceptable. Calculating Cronbach’s alpha yielded similar results (see Table A1). To assess discriminant validity, the heterotrait-monotrait (HTMT) ratio of correlations (Henseler, Ringle, & Sarstedt, 2015) was used. A HTMT value below .850 is commonly interpreted as an indicator of good discriminant validity. The HTMT values for the correlations among all PNR and PSM subdimensions ranged between .105 and .548, indicating that discriminant validity was good. In a next step, regression factor scores (DiStefano, Zhu, & Mindrila, 2009) were estimated for each case and used in all subsequent analyses.

The moderating variable, societal identification, was measured using an 11-point semantic differential that required interviewees to indicate which one of the two provided descriptions of attitudes toward society best fit their personal views. While no direct measure of societal identification was available in the survey, the used item should still be suited to serve as a proxy variable of the basic tenets of social identity theory (Tajfel, 1981). More specifically, the item taps into the question of whether one has a strong feeling of belonging to a certain society (e.g., Swiss or Swiss-German), while time being emotionally attached to its collective culture, values, and norms. The first description, located at the left end of the semantic differential (value 1 = minimum societal identification), described a low-identification attitude toward society, emphasizing individual freedom, cultural openness, and openness to immigration. The second description, located at the right end of the scale (value 11 = maximum societal identification), described a high societal identification, highlighting the importance of preserving collective societal values, the preference of law and order policies over individual freedom, and restrictive immigration policies. The values in between the two extremes (2-10) correspondingly described varying degrees of medium societal identification. As such, the measure of societal identification pertains to the perception of being part of a cohesive societal in-group, which is a key element in shared psychological contracts. This in-group, in turn, defines itself through the prevalent cultural attributes in a nation or society, thereby distancing itself from out-groups, for instance by advocating a restrictive stance on immigration.

Control variables included age, gender, whether interviewees had children, preference for urban versus rural life, work hours per week, tenure at police, managerial position, and overall job satisfaction. Table 1 provides descriptive statistics and bivariate Spearman correlations of all variables used in this study.

Method

To test the hypotheses, moderated multiple linear regression analysis was used. This method was chosen because the response variables, the factor scores of the various PSM dimensions, were interval-scale variables. For each of the three PSM dimensions (APM, COM, and CPI/SSF), two separate regression models were estimated. While the first model for each dimension included only the main and control variables, the second model additionally included the two interaction terms involving positive reciprocity norms and societal identification, and negative reciprocity norms and societal identification, respectively. Prior to estimating the models, all variables were mean centered to obtain standardized regression coefficients and missings were removed using listwise deletion. The four assumptions of linear models—linearity, homoscedasticity, uncorrelatedness, and normality—were evaluated for each model using a global validation approach (Peña & Slate, 2006). This procedure allows for an objective assessment as opposed to subjective approaches based on graphical techniques. Per the tests, the assumptions of linearity, homoscedasticity, uncorrelatedness, and the global statistic were not violated in any of the models. However, the presence of skewness was detected for the APM and COM models, meaning that the condition of normality was not met. To alleviate this issue, I used the MM-type regression estimator described by Koller and Stahel (2011) in all of the regression models. This estimator is based on a bi-square redescending score function and on robustly computed standard errors as described in Croux, Dhaene, and Hoorelbeke (2003), returning robust and efficient results in the presence of nonnormality. Finally, a check for common method biases (CMB) was performed using two tests recommended by Podsakoff, MacKenzie, Lee, and Podsakoff (2003). Harman’s single factor test showed that a single factor in an exploratory factor analysis explained only 15.0% of the variance, which is well below the common cutoff at 50%. In addition, and based on the CFA described above, the single-common-method-factor test estimated that 25.6% of the variance was attributable to method biases, well below the cutoff at 50% (Podsakoff et al., 2003). Thus, I assumed that CMB were not a serious issue.

Results

For a preliminary analysis of the data in general and of Hypothesis 1 in particular, we turn to the bivariate correlations presented in Table 1. As the coefficients illustrate, there exist moderate positive correlations among the subdimensions of PSM, all of which are statistically significant (ρ between .40 and .66, p < .001). As for the PNR scale, the correlation between positive and negative reciprocity norms is of similar strength and also statistically significant (ρ = .49, p < .001). An additional positive and statistically significant correlation of lower magnitude is found between negative reciprocity norms and societal identification (ρ = .33, p < .001). However, no statistically significant correlations were found between any of the dimensions of PSM and both positive and negative reciprocity norms, providing preliminary support for Hypothesis 1. In addition, several statistically significant correlations of low magnitude exist between the main and certain control variables (ρ between −.26 and .29, p between <.05 and <.001). Most notably, age, tenure at police, and managerial position were all positively correlated with all three PSM dimensions, and negatively correlated with positive reciprocity norms. In addition, the variable children was correlated negatively with all PSM dimensions, and positively with positive reciprocity.

For a more thorough test of Hypotheses 1 to 3, we consult the results from the MM-type moderated regression analyses, as per Table 2. This table contains a total of six models each two of which feature one of the three PSM dimensions as the dependent variable. The first model for each dimension (models APM1, COM1, and CPI/SSF1) only includes the main and control variables as predictors, whereas the second model (APM2, COM2, and CPI/SSF2) additionally integrates the interaction terms.

Table 2.

MM-Type Regressions on Subdimensions of Public Service Motivation Using Robust Standard Errors.

Dependent variable	PSM-APM1	PSM-APM2	PSM-COM1	PSM-COM2	PSM-CPI/SSF1	PSM-CPI/SSF2
Dependent variable	Coefficient (SE)	Coefficient (SE)	Coefficient (SE)	Coefficient (SE)	Coefficient (SE)	Coefficient (SE)
Intercept	.028 (.067)	.039 (.071)	.034 (.054)	.046 (.056)	.009 (.039)	.027 (.040)
Reciprocity positive	−.120 (.085)	−.135 (.085)	.037 (.067)	.058 (.067)	.075 (.050)	.059 (.049)
Reciprocity negative	.087 (.087)	.092 (.089)	.039 (.069)	.018 (.070)	−.083 (.051)	−.078 (.050)
Societal identification	.110 (.075)	.096 (.077)	.002 (.059)	−.023 (.060)	.029 (.044)	−.004 (.044)
Age	.494* (.190)	.468* (.194)	.331* (.152)	.300* (.152)	.195^† (.111)	.144 (.111)
Male	.253** (.082)	.254** (.083)	−.054 (.065)	−.039 (.065)	.066 (.048)	.076 (.048)
Children	.080 (.084)	.066 (.085)	−.131* (.066)	−.135* (.066)	−.021 (.048)	−.026 (.048)
Preference for urban life	.112 (.071)	.104 (.072)	.074 (.057)	.062 (.056)	.049 (.041)	.040 (.041)
Work hours per week (in % of max.)	−.073 (.076)	−.072 (.077)	.006 (.061)	.001 (.061)	.024 (.044)	.017 (.044)
Tenure at police	−.334^† (.186)	−.330^† (.188)	−.165 (.149)	−.152 (.148)	−.097 (.109)	−.079 (.108)
Managerial position	.124 (.075)	.122 (.076)	.024 (.060)	.037 (.060)	.045 (.044)	.054 (.044)
Overall job satisfaction	.115 (.071)	.109 (.072)	.052 (.057)	.041 (.057)	.136** (.042)	.133** (.042)
Reciprocity positive × Societal ident.		.117 (.079)		.132* (.061)		.138** (.045)
Reciprocity negative × Societal ident.		−.084 (.084)		−.091 (.065)		−.125** (.047)
R ²	.198	.204	.130	.154	.173	.219
R² adjusted	.149	.146	.077	.092	.122	.162
Robust Wald statistic (vs. intercept only)	42.6***	43.4***	25.5**	30.7**	36.1***	47.8***
Robust Wald statistic (change)		2.4		5.1^†		12.0**

Note. N = 194; coefficients are standardized; listwise deletion of missings. PSM = public service motivation; APM = attraction to policy making; COM = compassion; CPI/SSF = commitment to the public interest/self-sacrifice.

†

p < .10. *p < .05. **p < .01. ***p < .001.

Regarding the APM dimension of PSM, there are two variables that have a statistically significant positive association with APM in APM1, both of which are control variables. First, an increase by one standard deviation in age (10.2 years) is associated with an increase by .494 standard deviations in APM (p < .05). Second, an increase by one standard deviation on the variable male is associated with an increase by .253 standard deviations in APM (p < .01), which converts to an increase by .722 standard deviations in APM for males when compared with females. Notably, neither one of the coefficients for positive reciprocity and negative reciprocity is statistically significant, corroborating Hypothesis 1. After adding the interaction terms of both positive and negative reciprocity and societal identification in APM2, only marginal changes to the coefficients of the main and control variables can be observed. As for the statistically significant relationships, the coefficient for age decreases to .468 (p < .05) and the coefficient for the variable male slightly increases to .254 (p < .01). Note that while the two interaction terms have the predicted signs, neither one of them is of statistical significance which means that no support for Hypotheses 2a and 2b is found in this model. Nevertheless, this nonfinding may hold clues for the test of Hypothesis 3 in conjunction with the two remaining models. Finally, it should be highlighted that APM2 does not exhibit significantly improved model fit when compared with APM1, as a robust Wald test used to analyze the statistical significance of fit differences for regressions with robust standard errors indicated (Wald statistic = 2.4, p > .10).

As for the COM dimension of PSM, only two statistically significant coefficients are found in COM1. Again, both of them concern control variables. First, an increase by one standard deviation in age (10.2 years) is associated with an increase by .331 standard deviations in COM (p < .05). Second, an increase by one standard deviation on the dichotomous variable children is associated with a decrease by .131 standard deviations in COM (p < .05), which translates to a decrease by .267 standard deviations in COM for participants with children when compared with childless participants. Note that, again, neither one of the coefficients for positive reciprocity and negative reciprocity is statistically significant, further corroborating Hypothesis 1. In COM2, both of the statistically significant coefficients from COM1 change only slightly. The coefficient for age decreases to .300 (p < .05) and the coefficient for the variable children decreases to −.135 (p < .05). However, in contrast to APM2, the interaction term between positive reciprocity and societal identification, which is positive at a value of .132, is statistically significant (p < .05). Thus, support is found for Hypothesis 2a in this model. The interaction term between negative reciprocity and societal identification, however, is negative but not statistically significant, meaning that no support is found for Hypothesis 2b. Finally, recalling the nonfindings for both interaction terms from the APM model, the statistical significance of one of the interactions in the COM model provides first partial support for Hypothesis 3. As regards model fit, note that COM2 fits the data marginally better than COM1 (Wald statistic = 5.1, p < .10).

In regard to the merged CPI/SSF dimension of PSM, only one of the predictors is statistically significant in CPI/SSF1. An increase by one standard deviation in the control variable of job satisfaction is associated with an increase by .136 standard deviations in CPI/SSF (p < .01). As with all of the previous models, the coefficients for positive reciprocity and negative reciprocity are not statistically significant, meaning that strong overall support is found for Hypothesis 1. In CPI/SSF2, the statistically significant coefficient of job satisfaction is virtually unchanged in its magnitude when compared with CPI/SSF1 (.133, p < .01). In addition, the interaction term between positive reciprocity and societal identification, which is positive and larger than in the COM model at a value of .138, is statistically significant (p < .01), supporting Hypothesis 2a. The interaction term between negative reciprocity and societal identification is negative at a value of −.125 and, unlike in the previous models, statistically significant (p < .01), meaning that support is found also for Hypothesis 2b in CPI/SSF2. Finally, given the non-significant interaction terms in APM model and the statistical significance of one of the two interactions in the COM model, the statistical significance of both interaction terms in the CPI/SSF model further corroborates Hypothesis 3. Regarding model fit, note that the CPI/SSF2 exhibits significantly better fit than CPI/SSF1 (Wald statistic = 12.0, p < .01).

To facilitate the interpretation of the statistically significant interactions from the COM and CPI/SSF models in the subsequent discussion section, I created three marginal effects plots (Figures 1 -3) based on the conventions described in Aiken and West (1991). The PSM dimension in question is found on the y-axis of each plot, whereas the x-axis shows the values of the respective dimension of reciprocity. In all plots, the slope of the gray middle line corresponds to the mean value of the moderator societal identification, whereas the light gray line corresponds to the value one standard deviation below the mean, and the black line corresponds to one standard deviation above the mean. This procedure graphically illustrates the effect of the interaction on the PSM dimensions.

Discussion

The aim of this study was to investigate whether internalized reciprocity norms may manifest themselves vis-à-vis society and thereby play a role in a person’s levels of PSM. The analysis was done based on Hypotheses 1 to 3. Results from both the correlation and regression analyses strongly support Hypothesis 1, which postulated that neither positive nor negative reciprocity norms will be related to PSM. This suggests that internalized reciprocity norms alone are, in accordance with theoretical expectations, insufficient to explain high levels of PSM. Regarding Hypothesis 2a, which postulated that the interaction between positive reciprocity norms and societal identification will be positively related to PSM, partial support was found in the regression analyses. While no interaction effect was present regarding APM, the interactions for COM and CPI/SSF corroborated Hypothesis 2a. Looking at the respective interaction plots in Figures 1 and 2, the unstandardized slopes for the relationships between both COM and CPI/SSF and positive reciprocity norms are negative at low levels of societal identification (COM: B = −.108; CPI/SSF: B = −.115), reverse their signs at the mean (COM: B = .085; CPI/SSF: B = .086), and become even steeper in the positive direction at high levels of societal identification (COM: B = .277; CPI/SSF: B = .288). This suggests that societal identification is a prerequisite for positive reciprocity norms to positively affect these PSM dimensions. Partial support was furthermore found for Hypothesis 2b, which conjectured that the interaction between negative reciprocity norms and societal identification will be negatively related to PSM. While this hypothesis did not find any support regarding APM and COM, it was corroborated by the statistically significant findings for the merged CPI/SSF dimension. As per Figure 3, the slope for the relationship between CPI/SSF and negative reciprocity norms is positive at low levels of societal identification (B = .047), reverses its sign at the mean (B = −.076), and becomes steeper in the negative direction at high levels of societal identification (B = −.198), indicating that societal identification is a necessity for negative reciprocity norms to negatively affect CPI/SSF.

The above findings also provide partial support for Hypothesis 3, which posited that the interaction effects in Hypotheses 2a and 2b will be stronger for the norm-based (CPI) and affective (COM, SSF) dimensions of PSM than for the rational (APM) dimension. Given the nonfindings for APM, the partial findings for COM, and the robust findings for the merged CPI/SSF dimension, it appears warranted to conclude that the affective and norm-based PSM dimensions with their focus on benefiting others are more relevant drivers of strong reciprocity than the rational APM dimension (Ritz, 2011).

In sum, the findings of this study provide at least partial evidence for the theoretical argument that reciprocity norms may transcend relationships between two individuals or among small collectives to become effective between the individual and society. Depending on whether positive or negative reciprocity norms are concerned, such strong reciprocity thereby either increases or decreases levels of the normative and affective components of PSM in a person. As such, these findings are coherent with Koehler and Rainey’s (2008) line of reasoning that individuals may feel obligated to engage in civic service to repay society for the benefits it offers to its members. Importantly, however, this requires the individual to strongly identify with their society (Rousseau, 1995). This last point is the most relevant contribution of this study. By allowing insights into the moderating role of societal identification in strong reciprocity, I provide first evidence that PSM, viewed as a form of strong reciprocity, may be based on a shared psychological contract (Rousseau, 1995) between the individual and society. Further empirical evidence for this line of reasoning is provided by Kim (2015), who recently demonstrated that a negative relationship exists between individualism and PSM.

As such, the findings corroborate several of the definitions of PSM that implicitly touch upon the theme of identification with society or political entities. For instance, this concerns the one by Vandenabeele (2007, p. 547) who defines PSM as “the belief, values and attitudes that go beyond self-interest and organizational interest, that concern the interest of a larger political entity and that motivate individuals to act accordingly whenever appropriate” and the one by Rainey and Steinbauer (1999) presented in the introduction (however, in addition to states and nations, these authors also refer to humankind, which goes beyond the notion used in this study). The key focus on identification with society furthermore underscores the notion that certain individuals, for instance, strong reciprocators exhibiting high levels of PSM, perceive government employment as a calling (Houston, 2006; Perry, 1996). Importantly, however, it should be noted that society is not the only possible source of identification related to PSM. For instance, the identification with the policies underlying specific types of public work may similarly determine public servants’ PSM and the importance placed on its various dimensions (Anderfuhren-Biget, Varone, & Giauque, 2014).

On the other hand, the finding that PSM is partially based on strong reciprocity vis-à-vis society challenges the notion of PSM being a predominantly other-concerned motivation that drives individuals to benefit society without expecting reciprocal benefits (see Perry & Vandenabeele, 2015). In fact, strong reciprocators serve society exactly because they feel that they benefit from being a member of society, meaning that the motives behind their acts of altruism may not be so selfless after all. Arguably, reciprocal exchanges with society are less direct and less tangible than direct reciprocal exchanges. However, it would be premature to conclude from this that strong reciprocity and PSM are purely selfless, other-concerned forms of motivation. What is more likely is that PSM can be both self-concerned and other-concerned at the same time (Taylor, 2007; see also De Dreu & Nauta, 2009). While the role of mixed motives in PSM has been discussed previously, and while self-concerned motives have been identified as not unimportant drivers of PSM (Brewer et al., 2000; Le Grand, 2003), there is further need to explore the components of PSM that are often interchangeably called (selfless) altruism, prosocial motives, and self-sacrifice (Perry et al., 2010) to investigate in how far these components might also be based on self-concern that is, however, more long-term oriented. This could be done, for instance, by intensifying qualitative research efforts and should result in improved conceptual clarity.

As regards the specificities of police, which the sample of this study was drawn from, strong reciprocity in the form of PSM is arguably of even higher importance than in other areas of public work. In emotionally challenging or demotivating situations with specific citizens, PSM may operate as a coping mechanism (Bakker, 2015; Tummers, Bekkers, Vink, & Musheno, 2015) that lets police officers maintain their efforts based on the shared psychological contract they have with society, thus deriving motivation from the common good that overrides situational demotivation. In addition, it should be noted that police work can involve situations of extreme stress as officers put their lives at risk while serving the public. Such situations may change officers’ PSM, for instance by reducing their COM and reinforcing their CPI, which may thwart efforts to have an affectively motivated police force (Brænder & Andersen, 2013).

Furthermore, strong reciprocators with high levels of PSM are not only a valuable addition to public organizations. Such individuals may also be driven to take on key roles in civic participation and act as catalysts for the formation of social capital in societies (Brewer, 2003), thus strengthening the bedrock of democracy (Paxton, 2002). Accordingly, the findings of this study are not limited to the public sector context but may hold insights for scholars of other fields, such as volunteering (see also Steen, 2006). This is illustrated, for instance, by the case of Syrian refugee Alex Assali who, according to media reports, fled from his war-torn country and was granted asylum in Germany in 2014. In November 2015, a picture of him distributing self-cooked meals to the homeless of Berlin surfaced on Facebook. Attached to the small table he had brought was a sign reading “Give something back to German people.” His acts of strong reciprocity made it to international media and were inspiring to many.

This research holds several relevant implications for practice. To recruit employees who are well motivated to advance the public weal, I recommend public organizations develop measures to assess both their candidates’ internalization of reciprocity norms and their identification with society in the selection process. Such efforts, combined with the frequent recommendation to directly assess candidates’ levels of PSM (Ritz et al., 2016), may help recruit individuals who seek opportunities to reciprocate vis-à-vis society, thus creating valuable symbiotic effects. Accordingly, human resource management should furthermore ensure that such individuals be selected into service-oriented roles involving tasks that are useful to society to optimize person–job fit (Christensen & Wright, 2011; Neumann, 2016). To complement these measures, public organizations may want to consider using forms of employer branding to specifically send out brand signals related to “giving back to society” (Wilden, Gudergan, & Lings, 2010). Such branding may help to attract public service motivated employees in times of competitive labor markets, bringing together public organizations and public service motivated individuals. Finally, public managers are encouraged to use management practices such as transformational leadership (Bass & Riggio, 2006) to instill and reinforce positive reciprocity norms in their followers. Such practices could also be utilized to raise employees’ awareness of the social collective they serve to enhance their societal identification.

Several strengths and limitations of this study are noteworthy. As for the strengths, the three subdimensions of PSM included in this study were analyzed separately, allowing for a more differentiated perspective on the various motives that constitute PSM (Andersen & Serritzlew, 2012; Taylor, 2007). In addition, the sample of Swiss state police officers was particularly suited to study reciprocity norms vis-à-vis society and their relationship with PSM because police jobs offer frequent opportunities to directly benefit society and to give something back (Giauque et al., 2012). As for the limitations, first, it has to be noted that the analyses relied on cross-sectional data which does not allow for definite statements on the causality of variable relationships. Second, no direct measure of societal identification was available and the proxy variable which was used as a measure instead may not have captured all facets of the concept appropriately. Third, one has to be cautious when generalizing the findings from a police sample from rural Switzerland to other populations given the specificities of police work. Finally, sample size was rather limited due to practical constraints. Future research investigating the role of reciprocity norms in PSM is therefore encouraged to, first, use experimental research designs that allow for causal hypothesis testing and, second, use larger and more diverse samples to investigate the generalizability of the present findings.

Conclusion

The present study contributes to the literature on PSM in two ways. First, acknowledging the key role that reciprocity plays in many social situations, it invigorates and expands upon theoretical links between reciprocity norms and PSM that have been articulated in the past (Koehler & Rainey, 2008). More specifically, it brings in the concept of shared psychological contracts (Rousseau, 1995) and draws on social identity theory (Tajfel, 1981) to substantiate previous reciprocity-based explanations of acts of altruism vis-à-vis society. Given that reciprocal exchanges are usually upheld for personal benefits derived from them (Nowak & Sigmund, 2005), PSM is thereby viewed at least partially as a self-concerned motivation. Thus, this study challenges the common notion that PSM is mainly other-concerned and selfless. Second, this research complements previous empirical studies on antecedents of PSM (see Ritz et al., 2016) by analyzing the relationship between both positive and negative reciprocity norms and PSM, utilizing societal identification as a moderator. Coherent with theoretical assumptions, internalized reciprocity norms alone were found to be unassociated with PSM. However, for certain PSM dimensions, introducing societal identification as a moderator yielded positive associations with positive reciprocity, and negative associations with negative reciprocity. These findings corroborate the claim derived from theory on shared psychological contracts that there needs to be a strong identification with society to invoke reciprocity norms vis-à-vis society, and thereby PSM.

Footnotes

Appendix

Table A1.

PSM and PNR List of Items and Confirmatory Factor Analysis Results.

Item	Factor loadings (SE)		PSM-CPI/SSF	Reciprocity positive	Reciprocity negative
Item	PSM-APM	PSM-COM	PSM-CPI/SSF	Reciprocity positive	Reciprocity negative
I’m very interested in politics	1
I like to discuss political subjects with others	.887 (.051)
It is important to me to always be informed about current political developments	.774 (.045)
I am often reminded by daily events how dependent we are on one another		1
It is difficult for me to contain my feelings when I see people in distress		.580 (.243)
It is important for me to contribute to the common good			1
Much of what I do is for a cause bigger than myself			.930 (.205)
I would prefer seeing public officials do what is best for the whole community even if it harmed my interests			.837 (.178)
I feel people should give back to society more than they get from it			.792 (.158)
I am one of those rare people who would risk personal loss to help someone else			.696 (.159)
Meaningful public service is very important to me			.549 (.116)
When someone does me a favor, I feel committed to repay him or her				1
I go out of my way to help somebody who has been kind to me before				.969 (.152)
If someone does a favor for me, I am ready to return it				.955 (.153)
I am kind and nice if others behave well with me, otherwise it’s tit-for-tat					1
If somebody is impolite to me, I become impolite, too					.890 (.118)
The way I treat others depends much on how they treat me					.874 (.128)
If somebody offends me, I will offend him or her back					.758 (.122)
Cronbach’s α	.91	.47	.65	.72	.75
Raykov’s ω	.92	.51	.65	.72	.86
AVE, total = .47	.79	.35	.25	.46	.49
Max. HTMT	.53	.55	.55	.49	.49

Note. Fit indices: RMSEA = .042; CFI = .956; TLI = .945; χ²(123) = 165.802; p < .01. N = 194; all items were measured on a 7-point scale; actual items in survey were in German; estimation procedure was maximum likelihood; listwise deletion of missings was used; software used was lavaan 0.5-20 in the R environment; covariance matrix available upon request. PSM = public service motivation; PNR = Personal Norm of Reciprocity; APM = attraction to policy making; COM = compassion; CPI/SSF = commitment to the public interest/self-sacrifice; AVE = average variance extracted; HTMT = heterotrait-monotrait ratio of correlations.

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.

Author Biography

Oliver Neumann is a postdoctoral researcher in the KPM Center for Public Management and the Institute of Information Systems at the University of Bern, Switzerland. His research interests include employee motivation, person–job fit, government digitalization, and open government data.

References

Aiken

L. S.

West

S. G.

(1991). Multiple regression: Testing and interpreting interactions. Newbury Park, CA: Sage.

Alexander

R. D.

(1987). The biology of moral systems. New York, NY: Aldine de Gruyter.

Anderfuhren-Biget

Varone

Giauque

(2014). Policy environment and public service motivation. Public Administration, 92(4), 807-825.

Anderfuhren-Biget

Varone

Giauque

Ritz

(2010). Motivating employees of the public sector: Does public service motivation matter? International Public Management Journal, 13(3), 213-246.

Andersen

L. B.

Serritzlew

(2012). Does public service motivation affect the behavior of professionals? International Journal of Public Administration, 35(1), 19-29.

Bakker

A. B.

(2015). A job demands–resources approach to public service motivation. Public Administration Review, 75(5), 723-732.

Bass

B. M.

Riggio

R. E.

(2006). Transformational leadership (2nd ed.). Mahwah, NJ: Lawrence Erlbaum.

Binmore

K. G.

(1994). Game theory and the social contract. Cambridge, MA: MIT Press.

Bolino

M. C.

Grant

A. M.

(2016). The bright side of being prosocial at work, and the dark side, too: A review and agenda for research on other-oriented motives, behavior, and impact in organizations. The Academy of Management Annals, 10, 599-670.

10.

Brænder

Andersen

L. B.

(2013). Does deployment to war affect public service motivation? A panel study of soldiers before and after their service in Afghanistan. Public Administration Review, 73(3), 466-477.

11.

Brewer

G. A.

(2003). Building social capital: Civic attitudes and behavior of public servants. Journal of Public Administration Research and Theory, 13(1), 5-26.

12.

Brewer

G. A.

Selden

S. C.

Facer

R. L.

II . (2000). Individual conceptions of public service motivation. Public Administration Review, 60(3), 254-264.

13.

Christensen

R. K.

Wright

B. E.

(2011). The effects of public service motivation on job choice decisions: Disentangling the contributions of person-organization fit and person-job fit. Journal of Public Administration Research and Theory, 21(4), 723-743.

14.

Cropanzano

Mitchell

M. S.

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

15.

Croux

Dhaene

Hoorelbeke

(2003). Robust standard errors for robust estimators (Discussion Papers Series 03.16). Leuven, BE, KU: CES.

16.

De Dreu

C. K. W.

Nauta

. (2009). Self-interest and other-orientation in organizational behavior: Implications for job performance, prosocial behavior, and personal initiative. Journal of Applied Psychology, 94(4), 913-926.

17.

DiStefano

Zhu

Mindrila

(2009). Understanding and using factor scores: Considerations for the applied researcher. Practical Assessment, Research & Evaluation, 14(20), 1-11.

18.

Falk

Fischbacher

(2006). A theory of reciprocity. Games and Economic Behavior, 54(2), 293-315.

19.

Fehr

Fischbacher

(2003). The nature of human altruism. Nature, 425(6960), 785-791.

20.

Fehr

Fischbacher

Gächter

(2002). Strong reciprocity, human cooperation, and the enforcement of social norms. Human Nature, 13(1), 1-25.

21.

Fehr

Henrich

(2003). Is strong reciprocity a maladaptation? On the evolutionary foundations of human altruism. In Hammerstein

(Ed.), Genetic and cultural evolution of cooperation (pp. 55-82). Cambridge, MA: MIT Press.

22.

Giauque

Ritz

Varone

Anderfuhren-Biget

(2012). Resigned but satisfied: The negative impact of public service motivation and red tape on work satisfaction. Public Administration, 90(1), 175-193.

23.

Giauque

Ritz

Varone

Anderfuhren-Biget

Waldner

(2011). Putting public service motivation into context: A balance between universalism and particularism. International Review of Administrative Sciences, 77(2), 227-253.

24.

Gintis

(2000). Strong reciprocity and human sociality. Journal of Theoretical Biology, 206(2), 169-179.

25.

Gintis

Bowles

Boyd

Fehr

(2003). Explaining altruistic behavior in humans. Evolution and Human Behavior, 24(3), 153-172.

26.

Godbout

J. T.

(1992). L’esprit du don [The spirit of giving]. Paris, France: La Découverte.

27.

Grant

A. M.

(2008). Does intrinsic motivation fuel the prosocial fire? Motivational synergy in predicting persistence, performance, and productivity. Journal of Applied Psychology, 93(1), 48-58.

28.

Haley

K. J.

Fessler

D. M. T.

(2005). Nobody’s watching? Subtle cues affecting generosity in an anonymous economic game. Evolution and Human Behavior, 26(3), 245-256.

29.

Henseler

Ringle

C. M.

Sarstedt

(2015). A new criterion for assessing discriminant validity in variance-based structural equation modeling. Journal of the Academy of Marketing Science, 43(1), 115-135.

30.

Holström

(1979). Moral hazard and observability. The Bell Journal of Economics, 10(1), 74-91.

31.

Houston

D. J.

(2006). “Walking the walk” of public service motivation: Public employees and charitable gifts of time, blood, and money. Journal of Public Administration Research and Theory, 16(1), 67-86.

32.

Houston

D. J.

(2011). Implications of occupational locus and focus for public service motivation: Attitudes toward work motives across nations. Public Administration Review, 71(5), 761-771.

33.

Jackson

D. L.

Gillaspy

A. J.

Jr. Purc-Stephenson

(2009). Reporting practices in confirmatory factor analysis: An overview and some recommendations. Psychological Methods, 14(1), 6-23.

34.

Kim

(2015). National culture and public service motivation: Investigating the relationship using Hofstede’s five cultural dimensions. International Review of Administrative Sciences, 83(suppl. 1), 23-40.

35.

Kim

Vandenabeele

(2010). A strategy for building public service motivation research internationally. Public Administration Review, 70(5), 701-709.

36.

Kline

(2011). Principles and practice of structural equation modeling, methodology in the social sciences (3rd ed.). New York, NY: Guilford Press.

37.

Koehler

Rainey

H. G.

(2008). Interdisciplinary foundations of public service motivation. In Perry

J. L.

Hondeghem

(Eds.), Motivation in public management: The call of public service (pp. 33-55). Oxford, UK: Oxford University Press.

38.

Koller

Stahel

W. A.

(2011). Sharpening Wald-type inference in robust regression for small samples. Computational Statistics & Data Analysis, 55(8), 2504-2515.

39.

Le Grand

. (2003). Motivation, agency and public policy: Of knights and knaves, pawns and queens. Oxford, UK: Oxford University Press.

40.

Lei

P.-W.

(2007). Introduction to structural equation modeling: Issues and practical considerations. Educational Measurement: Issues and Practice, 26(3), 33-43.

41.

Matejkowski

McCarthy

K. S.

Draine

(2011). The personal norm of reciprocity among mental health service users: Conceptual development and measurement. Psychiatric Rehabilitation Journal, 34(3), 202-213.

42.

McCabe

K. A.

Rassenti

S. J.

Smith

V. L.

(1998). Reciprocity, trust, and payoff privacy in extensive form bargaining. Games and Economic Behavior, 24(1-2), 10-24.

43.

Mummendey

Klink

Brown

(2001). Nationalism and patriotism: National identification and out-group rejection. British Journal of Social Psychology, 40(2), 159-172.

44.

Neumann

(2016). Does misfit loom larger than fit? Experimental evidence on motivational person-job fit, public service motivation, and prospect theory. International Journal of Manpower, 37(5), 822-839.

45.

Nowak

M. A.

Sigmund

(2005). Evolution of indirect reciprocity. Nature, 437(7063), 1291-1298.

46.

Organisation for Economic Co-Operation and Development. (2016). OECD.Stat. Available from https://stats.oecd.org

47.

Paxton

(2002). Social capital and democracy: An interdependent relationship. American Sociological Review, 67(2), 254-277.

48.

Peña

E. A.

Slate

E. H.

(2006). Global validation of linear model assumptions. Journal of the American Statistical Association, 101(473), 341-354.

49.

Perry

J. L.

(1996). Measuring public service motivation: An assessment of construct reliability and validity. Journal of Public Administration Research and Theory, 6(1), 5-22.

50.

Perry

J. L.

Hondeghem

Wise

L. R.

(2010). Revisiting the motivational bases of public service: Twenty years of research and an agenda for the future. Public Administration Review, 70(5), 681-690.

51.

Perry

J. L.

Vandenabeele

(2015). Public service motivation research: Achievements, challenges, and future directions. Public Administration Review, 75(5), 692-699.

52.

Perry

J. L.

Wise

L. R.

(1990). The motivational bases of public service. Public Administration Review, 50(3), 367-373.

53.

Perugini

Gallucci

Presaghi

Ercolani

A. P.

(2003). The personal norm of reciprocity. European Journal of Personality, 17(4), 251-283.

54.

Podsakoff

P. M.

MacKenzie

S. B.

Lee

J.-Y.

Podsakoff

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

55.

Rainey

H. G.

(2003). Understanding & managing public organizations (3rd ed.). San Francisco, CA: Jossey-Bass.

56.

Rainey

H. G.

Steinbauer

(1999). Galloping elephants: Developing elements of a theory of effective government organizations. Journal of Public Administration Research and Theory, 9(1), 1-32.

57.

Ritz

(2011). Attraction to public policy-making: A qualitative inquiry into improvements in PSM measurement. Public Administration, 89(3), 1128-1147.

58.

Ritz

Brewer

G. A.

Neumann

(2016). Public service motivation: A systematic literature review and outlook. Public Administration Review, 76(3), 414-426.

59.

Rousseau

D. M.

(1995). Psychological contracts in organizations: Understanding written and unwritten agreements. Thousand Oaks, CA: Sage.

60.

Schott

Neumann

Baertschi

(2016, August). Public service motivation, prosocial motivation, prosocial behavior, and altruism: Towards disentanglement and conceptual clarity. Paper presented at the 2016 European Group for Public Administration (EGPA) conference, Utrecht University, Utrecht,The Netherlands.

61.

Simmel

(1950). The sociology of Georg Simmel (Translated by Wolff

Kurt H.

). Glencoe, IL: The Free Press.

62.

Steen

(2006). Public sector motivation: Is there something to learn from the study of volunteerism? Public Policy and Administration, 21(1), 49-62.

63.

Tajfel

(1981). Human groups and social categories: Studies in social psychology. New York, NY: Cambridge University Press.

64.

Tajfel

(2010). Social identity and intergroup relations. Cambridge, UK: Cambridge University Press.

65.

Taylor

(2007). The impact of public service motives on work outcomes in Australia: A comparative multi-dimensional analysis. Public Administration, 85(4), 931-959.

66.

Tummers

Bekkers

Vink

Musheno

(2015). Coping during public service delivery: A conceptualization and systematic review of the literature. Journal of Public Administration Research and Theory, 25(4), 1099-1126.

67.

Vandenabeele

(2007). Toward a public administration theory of public service motivation. Public Management Review, 9(4), 545-556.

68.

Whatley

M. A.

Webster

J. M.

Smith

R. H.

Rhodes

(1999). The effect of a favor on public and private compliance: How internalized is the norm of reciprocity? Basic and Applied Social Psychology, 21(3), 251-259.

69.

Wilden

Gudergan

Lings

(2010). Employer branding: Strategic implications for staff recruitment. Journal of Marketing Management, 26(1-2), 56-73.