Does Societal Culture Affect Public Service Motivation? Evidence of Sub-national Differences in Switzerland

Abstract

This article explores societal culture as an antecedent of public service motivation. Culture can be a major factor in developing an institution-based theory of public service motivation. In the field of organization theory, culture is considered a fundamental factor for explaining organization behavior. But our review of the literature reveals that culture has not been fully integrated into public service motivation theory or carefully investigated in this research stream. This study starts to fill this gap in the literature by using institutionalism and social-identity theory to predict how the sub-national Germanic and Latin cultures of Switzerland, which are measured through the mother tongues of public employees and the regional locations of public offices, affect their levels of public service motivation. Our analysis centers on two large data sets of federal and municipal employees, and produces evidence that culture has a consistent impact on public service motivation. The results show that Swiss German public employees have a significantly higher level of public service motivation on the whole, while Swiss French public employees have a significantly lower level overall. Implications for theory development and future research are discussed.

Full text

1
Does Culture affect Public Service Motivation?
Evidence of sub-national Differences in Switzerland
Adrian Ritz
University of Bern
Centre of Competence for Public Management
3001 Bern
Switzerland
Gene A. Brewer
University of Georgia
School of Public and International Affairs
Dept. of Public Administration and Policy
Athens, GA 30602-1615
USA
Published in: Ritz, A. and Brewer, G. A. 2013. “Does culture affect public service motiva-
tion? Evidence of sub-national differences in Switzerland.” International Public Man-
agement Journal, 16(2): 224-251.
The authors would like to thank the Federal Office of Personnel of the Swiss Federal Administration for the
support of this paper as well as for their help concerning the data collection at the federal level. The data from
the cantonal and municipal level were collected within the framework of a research project financed by the Swiss
National Science Foundation (project no. 100012-116083).

2

3
Does Culture affect Public Service Motivation?
Evidence of sub-national Differences in Switzerland
Abstract
This article explores sub-national culture as an antecedent of Public Service Motivation
(PSM). Culture can be seen as a major factor for further development of an institution based
theory of PSM. Whereas cultural relativity in organization theory is fundamental for the ex-
planation of employee behavior, present PSM research lacks of either theoretical work or
comparative studies tackling culture as an origin. Against the background of institutionalism
and social-identity theory this study analyzes the impact of regional and linguistic context
between the Germanic and Latin culture of Switzerland. The analysis uses two comparable
datasets at the federal and municipal level and results in a consistent cultural impact on PSM.
Public employees in the Swiss French region have lower level when compared to the German-
ic region. Conclusions and implications for further research are discussed.
Keywords
Public Service Motivation, National Culture, Identity, Public Values, Social Capital

4
Introduction
PSM proceeds from the assumption that work in public administration is more likely to satisfy
the motives of persons wishing to serve the public and who place the public good before their
own (Perry and Wise 1990). This view has a longstanding tradition in the USA (Horton
2008), but can also be found in Europe, Asia and Australia (Vandenabeele and Van de Walle
2008). The term ‘public service motivation’ (PSM) originated in the 1970s in discussions
about public service ethics and the differences in reward motivation between public and pri-
vate sector employees (Buchanan 1975; Rainey 1982) and today various definitions exist
naming PSM a motivation that goes beyond self-interest and serves a larger community of
people or political entity (Perry and Wise 1990; Brewer and Selden 1998; Rainey and
Steinbauer 1999; Vandenabeele 2007: 549). Each of these well-known definitions features the
elements “public” and “service” as motivational triggers. The term “service” in combination
with “public” emphasizes an ethical component of service to a community based on public
values which makes PSM distinct from other concepts of motivation. Although PSM is often
treated as a universal construct, attitudes and behavior to serve a political entity based on pub-
lic values has its origins in individuals’ values and identities which are largely linked to ones’
social environment and cultural context (Perry 2000; Perry and Vandenabeele 2008; Shamir
1991). In relation to individuals’ identity formation a corresponding process of socialization
plays an important role for the individual degree of PSM and raises question about the univer-
sality of the construct. Many open questions still exist concerning the origins and the devel-
opment process of PSM caused by socialization (Wright 2008).
This article investigates the impact of socialization through public employees’ cultur-
al context. Whereas cultural relativity in organization theory is fundamental for the explana-
tion of employee behavior, present PSM research lacks of either theoretical work or compara-
tive studies tackling culture as an origin. The study analyzes the impact of culture within a
framework of similar economic, regulatory, and political environment within one nation and

5
links PSM to institutional theory. Therefore, two datasets from Switzerland, one from the fed-
eral level and one from the municipal level, are used. Switzerland with its two cultural areas,
the Germanic and the Latin, and its multilingual and multicultural administrations offers
unique opportunities to study how attitudes towards work in government and public service
motives are affected by institutional differences. The country is often cited as a model of Fed-
eralism, especially because it is one of the wealthiest countries in the world despite linguistic
and cultural divisions. This study uses data from German Swiss, French Swiss, and Italian
Swiss public employees to identify the impact of culture on PSM.
The structure of the paper is as follows: First, a literature review on antecedents of
PSM is conducted in order to identify present research gaps and the themes of interest ad-
dressed by this study. In a second step, theories about institutional origins of PSM are dis-
cussed. Afterwards the theoretical perspective is applied to the cultural background of the
empirical study and four hypotheses are deduced. In section four, the data and method is de-
scribed. After the presentation of the results the last part of this article discusses the results
and draws conclusions for further research.
Literature Overview
Although strong arguments about the influence of institutional and cultural antecedents on
PSM exist, only a few studies investigate this issue. A literature review about the main ante-
cedents of PSM shows according to Perry (2000) and Pandey and Stazyk (2008) that a dis-
tinction between social-demographic antecedents such as gender or age (Camilleri 2007;
DeHart-Davis et al. 2006; Naff and Crum 1999), institutional antecedents like family sociali-
zation, religious or volunteering experience and education (Camilleri 2007; Perry 1997; Perry
et al. 2008), and organizational antecedents like job characteristics or incentives (Moynihan
and Pandey 2007; Scott and Pandey 2005; Taylor 2007) can be made.
Table 1 shows that most studies investigate PSM at the level of demographic and in-
dividual antecedents (e. g. age, gender, education) or relate it to organizational antecedents

6
perceived by the employees (e. g. role states, red tape, reform orientation). The overview re-
ports corroborated positive relationships by at least two studies between the antecedents edu-
cation, gender (female), religion, hierarchical grade as well as professionalism and PSM. But,
there exists still a great lack of research concerning institutional origins of PSM. Studies
which look at different institutional variables like organizational levels of government or
characteristics of agencies are missing. Furthermore, there are only a few studies which argue
from a more societal and cultural perspective. These studies show some empirical evidence
for PSM antecedents like volunteering, family socialization, and religion or union member-
ship. Nevertheless, the samples of all studies show that they were conducted within specific
cultural, national or organizational boundaries.
Against the background of this literature overview it is necessary to relate systematic
differences in public employees’ values and preferences to their cultural and institutional envi-
ronment. So far, PSM research lacks of empirical work investigating this issue.
[Insert Table 1 about here]
Theory and Hypotheses
Cultural Origins of Public Service Motivation
From a theoretical viewpoint institutional theory can give further insights into the cultural
origins of public employees’ motivation. An institutionalist approach takes into account that
“state, politics, and society are – for better or worse – inevitably intertwined” (Skocpol et al.
2000: 542). Institutions go beyond instrumental utility and an explanation of PSM based on
institutional theory reflects motivation to be an endogenous variable in contrary to it exoge-
nous characteristic in rational choice theories (Vandenabeele et al. 2006). Values, norms,
structures and rules as defining part of institutions infuse social structures with values and
promote stability and persistence over time (Peters 2000; Selznick 1984). The values of insti-
tutions “[...] are embedded across societies in religion, family, and other social structures”

7
(Perry and Vandenabeele 2008: 58). Institutions in the meaning of organizations are shaped in
reaction to their external environment and they transcend the individual level of an employ-
ee’s identity influencing its values, motives and defining a range of permissible and prohibited
behavior (Perry 2000; Perry and Vandenabeele 2008; Scott 1987). Thus, values and norms at
the individual level have societal and social roots and are learned through interactions within
communities, groups and organizations (Bandura 1986; March and Olsen 1995; Ostrom
1998).
As part of this contextualist approach following a logic of appropriateness (March
and Olsen 1989) various theories of socialization and social identification explain how identi-
ty and preferences of administrators are defined by the societal and political environment (for
an overview see Perry and Vandenabeele 2008; Vandenabeele 2007). Identity as the image a
person has of him or herself, as an individual and as part of a group (social identity), is always
done in the context of a position within an institution or of a particular culture (Stryker 1980;
Waldron 2000). Public values as values mainly associated with the culture of public institu-
tions and the public sector’s contribution to society are rooted, ultimately, in society as well as
in individuals’ and groups’ identities and preferences (Beck Jorgensen and Bozeman 2007;
Melchior and Melchior 2001). Identity changes individual preferences and is fundamental to
one’s behavior (Akerlof and Kranton 2000). Against the backdrop of social identity theory, for
example, language is seen as major institutional factor influencing identity formation although
less permeable group boundaries (e. g. skin color) increase the likelihood that a group mem-
ber will internalize group identity (Huddy 2001). Aspachs et al. (2008) show in their study
about language instruction in Spain, that different language education affects individual pref-
erences as well as a person’s identity and that “[t]he knowledge of local languages can foster
attachment to local institutions […]” (Aspachs-Bracons et al. 2008: 435).
Due to only a few studies which investigate the universal character of PSM there is a
need for more research about the effects of regional and cultural differences on the content

8
and levels of PSM (Vandenabeele and Van de Walle 2008). It is a matter of common
knowledge that nations and regions differ, above all in values, traditions and language
(Hofstede 1980), but also in their institutional characteristics like, for example, structures of
the politico-administrative system. Bangcheng et al. (2008) point out how important contex-
tualization is for PSM research in non developed and traditionally different nations like Chi-
na. His study could not confirm the “compassion” dimension in China and the authors specu-
late that benevolence for others is restrained due to traditional, upward oriented Chinese pub-
lic administration (Bangcheng et al. 2008). Hondeghem and Vandenabeele (2005) and
Vandenabeele et al. (2006) emphasize in their comparative work different foci of dimensions
of PSM between the countries compared which are in line with Norris’ (2003) findings about
different work values among government workers. For example, the dimension of compassion
relies in European countries to a welfare state tradition and government’s duty taking care of
the underprivileged whereas in the U.S. this responsibility is less assigned to the government.
The theoretical work discussed above lead us to a general proposition stating that so-
cietal culture is a highly important antecedent of PSM.
Cultural Differences between Germanic and Latin Regions
Switzerland belongs to two main cultural areas (Hofstede 2001); the Germanic area
with its German-speaking population and the Latin area with the French- and Italian-speaking
parts. Figure 1shows on the one hand the geographical border divided by the Alps between the
main language regions Latin and German. On the other hand there is a sharp divide between
the French and German language groups going from North to South which is not related to
geographical or institutional borders. Romansch as fourth national language is spoken in the
alpine regions in the East of Switzerland, both south and north of the Alps.
[Insert Figure 1 about here]

9
In Switzerland language is a signifier of deeper dissimilarities although there is
mono-nationality in spite of multilingualism. One of the decisive reason for the former was
that language equality ”never had to be fought for” (McRae 1983: 41) and when nationalism
increased on the European continent the country had already become multilingual. In compar-
ison to other European multilingual regions like for example Belgium or East-central Europe,
in Switzerland no significant language stratification by class existed, no central pressure was
imposed upon one language group and the high degree of federalism within the political sys-
tem could avoid cultural conflicts (Dardanelli forthcoming; Hofstede 1980).
But languages form a distinct barrier to migration between the Swiss regions (Linder
and Steffen 2006) and identities between the language regions differ (Watts 1988). For exam-
ple, “[l]arge parts of French-speaking Switzerland have been dominated by the German Swiss
oppressors from Berne during 250 years creating a desire for the French Swiss to distinguish
themselves from the ruling German elites and their cultural heritage” (Brügger et al. 2009: 7).
In addition, the Latin Swiss are often seen as living in a southern climate been very forthcom-
ing in contrast to the founding myth of the mountain peasant hard working German Swiss.
Whereas the Latin Swiss lean towards their neighbors France and Italy and their lingual affili-
ation with their neighbor communities is strong the German-speaking Swiss emphasize neu-
trality and independence and have always seen their variant of German language as means of
identification and separation from the German-speaking communities (Brügger et al. 2009;
Andres 1990).
Identity as a defining part of employees personality has a clear influence on work atti-
tudes and perceived importance of job characteristics (Levin and Stokes 1989). People of the
Latin regions report in the International Social Survey Programme of 1997 and 2005 signifi-
cant higher disagreement when asked about the enjoyment of a paid job even if one does not

10
need the money. Or citizens of Latin areas prefer more strongly longer vacations, less working
hours and a reduction of retirement age according to voting results (Brügger et al. 2009).
Looking at the Latin countries around Switzerland, empirical evidence for cultural
impact on work attitudes exists, too. For example, the French have been criticized for being
more concerned with taking vacations than making deals and the 35-working hour week in-
troduced in 2000 is an issue of great discussion (Gubian et al. 2004; Scherrer 1987). And
comparative research show cultural influences on work attitudes like Latin European manag-
ers prefer clear-cut job descriptions, centralized decision-making, and clearer hierarchical
structures (Laurent 1983) and French managers value the achievement of objectives as sec-
ondary to the acquisition of power whereas American managers’ goal attainment has primacy
over the aspiration to attain power (Inzerilli and Laurent 1983). “In line with Voltaire's (Can-
dide) advice, 'II faut cultiver notre jardin,' French managers tend to respond to problems by
focusing on their own 'turf,' referring problems up the hierarchy, and letting the government
from the environment (Crozier 1964).” (Schneider and Meyer 1991: 310).
According to the above mentioned regional peculiarities in Switzerland and empirical
evidence about cultural differences related to work attitudes in Latin and Germanic areas the
first two hypotheses read as follows:
H1: To be a member of the Germanic language group is positively correlated with
the level of PSM and its subdimensions.
H2: To be a member of the Latin language group is negatively correlated with
the level of PSM and its subdimensions.
Another argument about cultural differences between Swiss regions gives Freitag
(2004; 2006) with his research on social capital. According to Fukuyama (1999) the character-
istics of Latin areas like family ties, kinship and friendship moderate the engagement in for-
mal voluntarily associations. Freitag’s (2004) results confirm this for Switzerland where so-
cial capital in terms of associational attachment is higher in German-speaking areas and social

11
capital in terms of family ties is higher in the Latin areas (see also Joye and Laurent 1997). A
greater vivacity of the associational life in the Swiss German-speaking part of Switzerland
can be observed due to different development of political institutions. In Latin cantons the
model of democracy is inspired by representative government whereas in the German cantons
a more direct-democratic model is influenced by the French revolution (Kölz 1992; Linder
2005). One consequence, for example, is that local autonomy is less developed in the French-
speaking regions (Linder 2005). “It is […] plausible to assume that participatory political in-
stitutions provide an incentive to become not only more active in the political sphere, but to
become generally more civic minded and, therefore, more active in the associations of civil
society” (Kriesi and Baglioni 2003: 27). Frey’s (1997) work supports this showing empirical
evidence for the crowding-in of civic virtue with an increase of direct-democratic participa-
tion in Swiss cantons. In relation to individuals’ public service motives Wernli (1998) shows
that citizens of German-speaking Switzerland compared to the French Swiss are more inter-
ested in politics, have adopted a stronger political consciousness and show more intense polit-
ical participation. Thus, we formulate the following two hypotheses in accordance to our data
which has measures of social capital for the Swiss French and Swiss German language
groups:
H3: To be a member of the Swiss French language group is negatively correlated with
political participation as part of individuals’ social capital.
H4: To be a member of the Swiss French language group is negatively correlated with
associational attachment as part of individuals’ social capital.
Data
The data of this study were collected with two surveys. First, 26’544 employees took part in
the federal human resources survey. A random sample of all 7 ministries was taken and
14’482 responses were collected (response rate 54.6 %). The online and paper-pencil survey

12
was conducted in the period of February/March 2009. Second, the data at the municipal level
were collected within a research project of the Swiss National Science Foundation covering
279 of 1736 municipalities in German and French part of Switzerland. 9’852 employees took
part in the online and paper-pencil survey what resulted in a response rate of 38.1% and 3’754
responses. This survey was conducted in the period of October/November 2008.
Although the dataset cannot claim to be representative for both levels of government, the two
datasets can be compared. Both dataset contain mostly the same measures for each level. Ex-
ceptions are the following: Italian as language group was only part of the federal survey, Edu-
cation as a control variable was only available at the municipal level, and the variable Head-
quarter only exists at the federal level.
The distribution among the control variables shows major differences between the
federal sample and the local sample (see Table 2). The federal sample shows comparably
more men, a slightly younger workforce, fewer employees with supervisory position, higher
salary, higher tenure, and less part-time workers. These distinctive sample characteristics be-
tween the federal and the other two levels have to be taken in account when comparing PSM
levels.
[Insert Table 2 about here]
Although the datasets might not reflect the distribution in reality the proportion of
languages in the whole dataset can be compared to Swiss reality. Whereas the population of
language regions is about 72% for German, 24% for French, and 4% for Italian, the main lan-
guages of federal employees is 71% for German, 20% for French, and 7% for Italian (Lüdi
and Werlen 2005). Accordingly, our sample contains 70% of German, 20% of French, and 7%
of Italian speakers. The location of offices at the federal level is 89% within the German, 8%
within the French, and 3% within the Italian-speaking region. At the municipal level our sam-

13
ple contains 79% of municipalities located within the German-speaking part of the country
and 21% within the French region. The surveys were conducted in German, French, and Ital-
ian (only at federal level). The items were translated from professional translators for each
language. The translations of PSM items were discussed and adapted within a research group
with native speaker for German and French. An Italian native speaker was consulted for the
Italian translations. All questionnaire items were pretested with samples across the three lan-
guage groups.
Method and Measures
According to the aggregated metric dependent variables various models were tested with or-
dinary least squares regression analysis (OLS) for the federal data and hierarchical regression
analysis (OLS) for the municipal data. A hierarchical regression analysis for the federal data
including simultaneously all three language region could not be conducted due to the three
dummy variables which exhausted itself the whole dataset. The blocks of independent varia-
bles were entered into the regression equations either reflecting causal priority in the sense of
temporal ordering or relevance of research (Cohen et al. 2003). The models are statistically
significant at the .001 level except for two models. There are no problems of multicollinearity
– the highest variance inflation factor (VIF) is below 3.
The control measures gender, age, salary, tenure, supervisory position, and full-/part-
time employment were used across both samples. To control best possible in each dataset
some control variables differ between the samples. Education and religious activity were only
available at the municipal level and a separate variable headquarter was generated for the fed-
eral level in order to control for a potential minority effect because most offices of the federal
administration are located within the German-speaking area around the capital of Bern. This
variable helps to control for such an effect, although the proportional representation among
the linguistic minorities are given in the federal administration (Linder and Steffen 2006).

14
However, it should be kept in mind that proportional representation does not necessarily guar-
antee proportional influence in practice.
Public Service Motivation is measured taking Perry’s (1996) multidimensional scale
as baseline. A factor analysis could not produce the same four dimensions of PSM in both
samples. Therefore, the dimensions are measured as follows accordingly to the original com-
position of Perry (Perry 1996): three items for each dimension of attraction to public policy
making (APM), compassion (COM) and self-sacrifice (SS) and two items for commitment to
public interest (CPI) (see Appendix). For both sample the reliability coefficients for the com-
prehensive PSM scale were above the suggested lower boundary of 0.60 (Hair Jr. et al. 2009)
(PSMFed = 0.83; PSMMun = 0.71). For the federal sample all measures were above the thresh-
old (APM = 0.87, CPI = 0.61, COM = 0.70, SS = 0.83) whereas for the municipal sample one
of the reliability coefficients for each subscale was below the general threshold (APM = 0.63,
CPI = 0.62, COM = 0.68, SS = 0.54). Although this scale therefore lacks some internal con-
sistency (e. g. due to the limited number of items), the scale remains in the analysis in order to
have the same dimensional structure as at the federal level.
For measuring cultural influence two variables, namely mother tongue and language
region, are chosen: At the federal level, first, language group was measured with dummy vari-
ables using self-reported mother tongue for the sample at the federal level. Second, the locali-
ty of federal offices was assigned to their language region. At the municipal level, all munici-
palities were assigned to their language region. According to the geographic location of all
municipalities 79 percent of respondents work in a German municipality and 21 percent in a
French local administration. Thus, the measures represent the distribution of main language
and of the population of language regions quite well. Whereas at the federal level mother
tongue and location of office could be operationalized for Swiss French and Swiss Italian, the
municipal survey covered only the Swiss German and the Swiss French regions.

15
Three aspects of social capital were measured (Putnam 1998; Onyx and Bullen 2000).
First, an index for quality of social relations using eight questions for social and institutional
trust was applied in the analysis (Hughes et al. 2000). The measure shows a reliability coeffi-
cient of 0.88. Second, volunteering was measured with two questions regarding associational
membership and volunteering or donating money to a charity organization which were aggre-
gated to one index. Third, political participation is based on five questions asking respondents
about their political activities and participation during the last year.
Results
Mean Comparison
The univariate analysis compares the means of the variables and dimensions of PSM. Public
service motivation as variable of special interest is therefore analyzed at the item level, at the
dimensional level and as a whole construct. The results in Table 3 show two main effects.
First, the employees in the French region score lowest across all items in the federal sample
and are also lower when compared to the Swiss German municipalities across 8 of 11 items.
At the dimensional level both samples show lowest scores for the French region except for the
dimension of compassion at the municipal level. Second, the comparison with a comparable
small sample of offices in the Italian region at the federal administration shows a surprising
effect. Employees within this region score higher for all items and dimensions, respectively,
except for attraction to public policy making. Furthermore, the univariate analysis shows the
comparable low scores for attraction to public policy making and self-sacrifice among the
French samples. Above all, the first item of self-sacrifice (“Much what I do is for a cause big-
ger than myself”) is very low rated among the respondents of the French regions of both sam-
ples. This will be discussed in the last section of the article.
In general, the mean comparisons indicate preliminary evidence for the hypothesized
lower level of PSM of the Latin language and higher level of PSM of the Germanic language
group when only the French and Germanic regions of both sample are taken into account.

16
[Insert Table 3 about here]
Bivariate Analysis
In order to be able to recognize the importance of the independent variable culture influencing
PSM the standardized coefficients of regression analysis are examined. Therefore three re-
gressions for each language region at the federal level and three models of the hierarchical
regression analysis for the municipal dataset were calculated. This procedure was repeated for
each subscale of PSM. Tables 4 and 5 present the multiple regression results obtained by re-
gressing the five dependent variables (PSM and four subscales) on the set of independent var-
iables. The independent variables vary between the two datasets as explained in the Measures
Section. The adjusted multiple coefficient of determination is between 1.2 and 11.3 percent at
the federal level, which is comparably low. A major reason for that might be the selective and
small choice of independent antecedents of PSM. For the present study, the focus of the anal-
ysis was limited to the comparable indicators of controls and cultural variables within the two
datasets. For example, at the federal level factors regarding individuals’ family, religious or
societal background and variables in relation to the organizational environment which have
elsewhere shown to have high explanatory power (e. g. job satisfaction, pay satisfaction, lead-
ership, p-o-fit) were not included. For the municipal sample the adjusted multiple coefficient
of determination for model 3 is between 11.6 and 22.3 percent which is indicative for the ex-
planatory power of the model. Although it shows potential for improvement, it stays within
the results of other studies (e. g. Perry 1997).
The expected relationships between certain control variables and PSM can be ob-
served among the samples and mostly support the findings of other studies about PSM ante-
cedents (see Table 1). In detail, men have higher levels of PSM at the municipal level and
have higher levels of APM, CPI at municipal level, lower COM, and higher SS at municipal
level than women. Higher age is positively correlated with PSM and its subscales except for

17
APM at the municipal level and a non-significant relationship for SS in the municipal sample.
Education, only available at the municipal level, is positively correlated with PSM, especially
APM, and COM. Leadership position is positively correlated with PSM and its subdimensions
for federal employees except for non-significant values for COM. In the municipal sample
leadership position is significant negatively correlated with COM. Higher salary shows a
positive relationship with PSM and APM in both samples, with CPI at the federal level and a
significant negative correlation with COM in the municipal sample. Lower tenure has a posi-
tive relationship with PSM at both levels, with APM, CPI, COM and SS at the federal level.
Furthermore, part-time work is positively correlated with PSM, APM, and COM in both sam-
ples, but shows a significant negative relationship with CPI for federal employees. In the fed-
eral sample headquarter was included in the analysis and shows to have a significant positive
effect for PSM and all subscales except for COM. This gives evidence that working closer to
the capital or closer to the center of politics positively influences public employees’ PSM.
Of all control variables the highest correlation in both samples shows salary with
APM except for male in the municipal sample. This can also be seen in relation to APM as a
measure of rational and self-interested motives which are more strongly correlated to salary
than any other dimensions of this study.
In a further step of the hierarchical regression analysis the influence of culture was
assessed and the first two hypotheses tested. Due to the fact that the Italian region was only
included in the federal sample, the influences of the French and Italian region on PSM are
analyzed separately. For the federal sample mother tongue as self-reported by the respondents
was additionally included in the analysis to validate the regional effect.
[Insert Table 4 about here]
[Insert Table 5 about here]

18
The influence of the dummy variable German region with the reference groups
French region and Italian region only shows a significant positive effect on PSM and APM at
the federal level. The other subdimensions are not significantly influenced. For mother tongue
German a statistically positive influence on PSM, APM, and SS is found and a negative on
CPI. In the municipal sample, the effect is quite similar, although only compared to the Swiss
French as reference group; except for COM Swiss Germans correlate positively. Summariz-
ing, the Germanic influence on PSM is mostly positive when compared to the Latin region.
When looking at the influence of French region this effect get even reinforced. First, employ-
ees of the federal sample score significantly lower for PSM on all subscales and the same af-
fect exists at the municipal level except for COM. Second, the same direction of effects but
quite a bit stronger beta values results for French mother tongue, except for CPI. This effect
surprises in its consistency for both samples and both type of cultural variables (region, moth-
er tongue). The influence of the small amount of offices located in the Italian region shows
consistent non-significant effects on all dimensions. Nevertheless, a positive impact of mother
tongue Italian on PSM as a whole and on CPI and COM when compared with mother tongue
German can be observed. When taking into account the extent of effect, then a major differ-
ence between CPI and SS and all other dimensions is observed. Mother tongue French for the
federal sample and French region show a comparable strong and negative influence on CPI
and SS. Especially the effect on SS across both samples stands out.
How can these results about the impact of culture on PSM and its subscales, meas-
ured by either region or mother tongue, be summarized? First, Germanic as major culture of
Swiss public administrations has positive effect on PSM and most of its subdimensions what
supports hypothesis 1. Second, the findings about the influence of the Latin culture as stated
in hypothesis 2 are mixed. A clear negative impact of Swiss French on PSM can be observed
but this does not hold for Italian. And third, the dimension of SS can be seen as a major dif-
ference in how PSM is affected by either Swiss German or Swiss French culture.

19
In comparison to culture measured either by region or mother tongue, social capital
variables as a part of the societal socialization of public employees account for most of the
increase in variance explanation for the municipal sample in the analysis. But also the in-
crease of the variables region and mother tongue in both samples was statistically significant.
All four social capital variables show a positive effect on PSM with political participation
having the largest extent. At the level of subdimension religious activity is non-significantly
correlated with APM and political participation does not have a significant influence on COM
and SS. In order to test the third and fourth hypotheses about differences in associational at-
tachment and political participation a further regression analysis for social capital was con-
ducted (see Table 6). Although the additional explanatory power of the variables of social
capital measured is small, significant effects can be found for volunteering and political par-
ticipation. Public employees of municipalities in the French region have lower levels of vol-
unteering and of political participation. The latter result corroborates hypothesis three and
four.
[Insert Table 6 about here]
Discussion, Limitations and Conclusion
The ambition of this paper was to look at empirical evidence for variables addressing institu-
tional factors which explain the influence of culture on levels of PSM. Cultural theory and
institutionalism enlarge the theoretical framework of PSM as issue of motivation research.
Besides employees’ goals, expectations, capabilities and organizational factors like strategies,
structures or human resources measures, the cultural and societal environment influences in-
dividuals’ self-concepts and identities. To investigate these issues this study is based on two
datasets at the federal and local levels of Switzerland. Cultural impact was measured by loca-
tion of public organizations and mother tongue of public employees.

20
Before discussing the results one major limitation of this study has to be addressed.
Socialization effects are investigated based on cross-sectional data and regions and language
as proxy variables for culture is used. Longitudinal data with independent measures of indi-
viduals’ values for culture would provide a more accurate analysis. But such data are not yet
available as far as the authors know. The present study uses at least real independent variables
from different sources what can be described as an improvement within PSM research.
Against the background of the studies of Bangchen (2008) and Vandenabeele et al.
(2006) which showed the importance of cultural impact on PSM, cultural and societal differ-
ences in Switzerland are of high importance when explaining PSM. French as part of the Lat-
in language group in Switzerland has a negative impact on PSM. This supports other empiri-
cal findings on the effect of language and culture on different work values and work outcomes
for Swiss French (Brügger et al. 2009; Ichino and Maggi 2000). A cultural explanation for this
finding, for example, could be the different way French culture sees government and bureau-
cracy. According to Crozier (1964) problems in French business firms are often referred up
the hierarchy to the top of a company and to the government. This indicates other values
about the exercise of authority which differ from values in other cultures (Hofstede 1983).
“These cultural factors contribute to the stronghold of bureaucracy, as well as the strong hand
of the government […]”(Schneider and Meyer 1991: 310) in France and can be at least used
as an argument for the presented results which have to be tested more specifically for Swiss
French as well as French culture.
The results of this study is in line with research showing that differences in political
institutions like lower direct-democratic access or associational affiliation can lead to lower
interest in politics, political participation and civic orientation (Frey 1997; Kriesi and Baglioni
2003; Wernli 1998). From four tested variables of social capital two show an effect which
clearly supports these arguments: Volunteering, measured by associational affiliation, and

21
political participation, measured by the intensity of exercising votes and participating at polit-
ical meetings, were lower among respondents of the Swiss French region.
Astonishingly and in contrary to our hypothesis, the levels of PSM of Italian-
speaking employees measured at the federal level are significantly higher than the levels of
their Swiss German (and also Swiss French) colleagues. We can provide two possible expla-
nations. First, religion might be a stronger predictor of PSM in the Italian region of Switzer-
land due to a comparatively high proportion of Catholics (around 75 percent) in this part of
the country. Following Max Weber (2002), it can be argued that catholic values, for example
obedience and loyalty, have an impact on PSM and this is more spread in catholic regions and
among people with catholic background, respectively. But, if religion would be the main ex-
planation we would have expected to see a significant effect of Italian on self-sacrifice which
does not exist. On the contrary, France as one of the most secular state might have an influ-
ence on Swiss French as Freitag (2006: 134) points it out: “[…] with regard to the Swiss re-
gions, the influences of the respective linguistic cultures are reinforced by the strong impact
of the neighboring countries’ media and literature on the respective sectors of the Swiss popu-
lation.” An explanation for the differences in levels of PSM between the Swiss French and the
Swiss German can be given by the Protestant work ethic which is closely related to needs for
achievement or self-sacrifice and which is a stronger anchor within the German-speaking cul-
ture (Furnham 1989). Empirical evidence at least for general religious activity as an anteced-
ents of PSM is found in the municipal dataset and supports the findings of other research
(Perry 1997; Perry et al. 2008). In our study this is exemplified by the relative low levels of
self-sacrifice among the Swiss French when compared to the Swiss German. Future research
should investigate PSM in relationship to diverse religious background.
Second, besides religion, another explanation comes from the concrete work situation
of Italian-speaking employees in the federal administration. Employees who leave the Italian
region and move to the German part of Switzerland, where most federal offices are located,

22
might have a motivation for work in the public service which is above average. These are not
people who found a job by accident and took it from one-day to another. Literally they rather
moved “over the Alps” to the German-speaking region for their year of studies and were in-
terested in a federal post or they made a profound decision to leave the Italian part of Switzer-
land for a federal job in the capital. This argument is supported that in contrary to the Swiss
French people from the Italian part cannot expect that many colleagues in the federal admin-
istration speak Italian. Although the proportional representation of the different ethno-
linguistic groups is respected to a high degree in the federal administration, it is easier for
Italian-speaking people if they speak French or German. Therefore, we summarize that Ital-
ian-speaking employees show in average higher levels of PSM than the other two groups and
self-selection might be the main reason for that.
In conclusion, our study raises one important question about PSM research in relation
to organizational research. The results of cultural influence on PSM shed light on a major
origin of PSM – societal and public values – and points to the cultural relativity of PSM theo-
ry. This might also make PSM distinct from work motivation and general motivation theory.
Culture matters because public values are culture as well as institution bound and they foster
attitudinal and behavioral outcomes.

23
References
Akerlof, G. A. and R. E. Kranton. 2000. “Economics and Identity.” Quarterly Journal of
Economics, 115(3): 715-753.
Andres, F. 1990. “Language Relations in Multilingual Switzerland.” Multilingua, 9(1): 11-45.
Aspachs-Bracons, O., I. Clots-Figueras, J. Costa-Font, and P. Masella. 2008. “Compulsory
Language Educational Policies and Identity Formation.” Journal of the European
Economic Association, 6(2/3): 434-444.
Bandura, A. 1986. Social Foundations of Thought and Action: A Social Cognitive Theory.
Englewood Cliffs, NJ: Prentice-Hall.
Bangcheng, L., T. Ningyu, and Z. Xiaomei. 2008. “Public Service Motivation and Job
Satisfaction in China: An Investigation of Generalisability and Instrumentality.”
International Journal of Manpower, 29(8): 684-699.
Beck Jorgensen, T. and B. Bozeman. 2007. “Public Values: An Inventory.” Administration
Society, 39(3): 354-381.
Brewer, G. and S. Selden. 1998. “Whistle Blowers in the Federal Civil Service: New
Evidence of the Public Service Ethic.” Journal of Public Administration Research and
Theory, 8(3): 413-440.
Bright, L. 2005. “Public Employees With High Levels of Public Service Motivation.” Review
of Public Personnel Administration, 25(2): 138 - 154.
Brügger, B., R. Lalive, and J. Zweimüller. 2009. Does Culture Affect Unemployment?
Evidence from the Röstigraben: Institute for the Study of Labor IZA Bonn.
Buchanan, B. 1975. “Red-Tape and The Service Ethic.” Administration and Society, 6(4):
423-444.
Camilleri, E. 2006. “Towards Developing an Organisational Commitment - Public Service
Motivation Model for the Maltese Public Service Employees.” Public Policy and
Administration, 21(1): 63-83.
Camilleri, E. 2007. “Antecedents Affecting Public Service Motivation.” Personnel Review,
36(3): 356-377.
Castaing, S. 2006. “The Effects of Psychological Contract Fulfilment and Public Service
Motivation on Organizational Commitment in the French Civil Service.” Public Policy and
Administration, 21(1): 84-98.
Cohen, J., P. Cohen, S. G. West, and L. S. Aiken. 2003. Applied Multiple
Regression/Correlation Analysis for the Behavioral Sciences, 3 edn. Mahwah, New Jersey:
Lawrence Erlbaum Associates.
Crozier, M. 1964. The Bureaucratic Phenomenon. Chicago, IL: University Press.
Dardanelli, P. forthcoming. “Multi-lingual but Mono-national: Exploring and Explaining
Switzerland's Exceptionalism.” in M. Caminal and F. Requejo, eds, Democratic
Federalism and Multi-national Federations. Barcelona: Institut d’Estudis Autonòmics.
DeHart-Davis, L., J. Marlowe, and S. K. Pandey. 2006. “Gender Dimensions of Public
Service Motivation.” Public Administration Review, 66(6): 873-887.
Freitag, M. 2004. “Schweizer Welten des Sozialkapitals: Empirische Untersuchungen zum
sozialen Leben in Regionen und Kantonen.” Swiss Political Science Review 10(2): 87-118.
Freitag, M. 2006. “Bowling the State Back in: Political Institutions and the Creation of Social
Capital.” European Journal of Political Research, 45(1): 123-152.
Frey, B. S. 1997. “A Constitution for Knaves Crowds out Civic Virtues.” Economic Journal,
107(443): 1043-1053.
Fukuyama, F. 1999. The Great Disruption: Human Nature and the Reconstitution of Social
Order. New York: Free Press.
Furnham, A. 1989. The Protestant Work Ethic. London: Routledge.
Gubian, A., S. Jugnot, F. Lerais, and V. Passeron. 2004. “Les Effets de la RTT sur l’Emploi:
Des Simulations ex ante aux Evaluations ex post.” Économie et Statistique(376-377).

24
Hair Jr., J. F., W. C. Black, B. J. Babin, and R. E. Anderson. 2009. Multivariate Data
Analysis, 7 edn. Upper Saddle River, NJ: Pearson Prentice Hall.
Hofstede, G. 1980. Culture’s Consequences. International Differences in Work - Related
Values. Newbury Park/London/New Dehli: Sage.
Hofstede, G. 1983. “National Cultures in Four Dimensions.” International Studies of
Management & Organization, 13(1/2): 46-74.
Hofstede, G. 2001. Culture’s Consequences: Comparing Values, Behaviors, Institutions and
Organizations across Nations. London: Sage.
Hondeghem, A. and W. Vandenabeele. 2005. “Valeurs et Motivation dans l'Administration
Publique: Perspective Comparative.” Revue Française d'Administration Publique, 115:
463-480.
Horton, S. 2008. “History and Persistence of an Idea and an Ideal.” Pp. 17-32 in J. L. Perry
and A. Hondeghem, eds, Motivation in Public Management: The Call of Public Service.
New York et al.: Oxford University Press.
Huddy, L. 2001. “From Social to Political Identity: A Critical Examination of Social Identity
Theory.” Political Psychology, 22(1): 127-156.
Hughes, P., J. Bellamy, and A. Black. 2000. “Building Social Trust through Education.” Pp.
225-249 in I. Winter, ed., Social Capital and Public Policy in Australia. Melbourne:
Australian Institute of Family Studies.
Ichino, A. and G. Maggi. 2000. “Work Environment and Individual Background: Explaining
Regional Shirking Differentials in a Large Italian Firm*.” Quarterly Journal of Economics,
115(3): 1057-1090.
Inzerilli, G. and A. Laurent. 1983. “Managerial Views of Organization Structure in France
and the USA.” International Studies of Management & Organization, 13(1/2): 97-118.
Joye, D. and A. Laurent. 1997. “Associative and Political Participation in Switzerland and
France.” in J. W. Van Deth, ed., Private Groups and Public Life. London, New York:
Routledge.
Kölz, A. 1992. Neuere schweizerische Verfassungsgeschichte. Ihre Grundlinien vom Ende der
Alten Eidgenossenschaft bis 1848. Bern: Stämpfli.
Kriesi, H. and S. Baglioni. 2003. “Putting Local Associations into their Context: Preliminary
Results from a Swiss Study of Local Associations.” Swiss Political Science Review, 9(3):
1-34.
Laurent, A. 1983. “The Cultural Diversity of Western Conceptions of Management.”
International Studies of Management & Organization, 13(1/2): 75-96.
Levin, I. and J. P. Stokes. 1989. “Dispositional Approach to Job Satisfaction: Role of
Negative Affectivity.” Journal of Applied Psychology, 74(5): 752-758.
Linder, W. 2005. Schweizerische Demokratie: Institutionen - Prozesse - Perspektiven, 2nd
edn. Bern/Stuttgart/Wien: Haupt.
Linder, W. and I. Steffen. 2006. Ethnic Structure, Inequality and Governance in the Public
Sector in Switzerland.
Lüdi, G. and I. Werlen. 2005. Sprachenlandschaft in der Schweiz: Eidgenössische
Volkszählung 2000. Neuchatel: Bundesamt für Statistik.
March, J. G. and J. P. Olsen. 1989. Rediscovering Institutions: The Organizational Basis of
Politics. New York, London.
March, J. G. and J. P. Olsen. 1995. Democratic Governance. New York: Free Press.
McRae, K. 1983. Conflict and Compromise in Multilingual Societies - Switzerland. Waterloo,
ON: Wilfried Laurier University Press.
Melchior, M. B. and A. Melchior. 2001. “A Case for Particularism in Public Administration.”
Administration Society, 33(3): 251-275.
Moynihan, D. P. and S. K. Pandey. 2007. “The Role of Organizations in Fostering Public
Service Motivation.” Public Administartion Review, 67(1): 40 - 53.

25
Naff, K. C. and J. Crum. 1999. “Working for America: Does Public Service Motivation Make
a Difference?” Review of Public Personnel Administration, 19(4): 5 - 16.
Norris, P. 2003. “Is There Still A Public Service Ethos? Work Values, Experience, and Job
Satisfaction Among Government Workers.” Pp. 72-89 in J. D. Donahue and J. S. J. Nye,
eds, For the People - Can we Fix Public Service. Washington DC: Brookings Institution
Press.
Onyx, J. and P. Bullen. 2000. “Sources of social capital ” Pp. 105-134 in I. Winter, ed., Social
Capital and Public Policy in Australia. Melbourne: Australian Institute of Family Studies.
Ostrom, E. 1998. “A Behavioral Approach to the Rational Choice Theory of Collective
Action.” American Political Science Review, 92: 1-22.
Pandey, S. K. and E. C. Stazyk. 2008. “Antecedents and Correlates of Public Service
Motivation.” Pp. 101-117 in J. L. Perry and A. Hondeghem, eds, Motivation in Public
Management: The Call of Public Service. New York et al.: Oxford University Press.
Park, S. M. and H. G. Rainey. 2007. “Antecedents, Mediators, and Consequences of
Affective, Normative, and Continuance Commitment: Empirical Tests of Commitment
Effects in Federal Agencies.” Review of Public Personnel Administration, 27(3): 197-226.
Perry, J. L. and L. R. Wise. 1990. “The Motivational Bases of Public Service.” Public
Administration Review, 50(3): 367-373.
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.
Perry, J. L. 1997. “Antecedents of Public Service Motivation.” Journal of Public
Administration Research and Theory, 7(2): 181-197.
Perry, J. L. 2000. “Bringing Society In: Toward a Theory of Public-Service Motivation.”
Journal of Public Administration Research and Theory, 10(2): 471 - 488.
Perry, J. L., J. L. Brudney, D. Coursey, and L. Littlepage. 2008. “What Drives Morally
Committed Citizens? A Study of the Antecedents of Public Service Motivation.” Public
Administration Review, 68(3): 445-458.
Perry, J. L. and W. Vandenabeele. 2008. “Behavioral Dynamics: Institutions, Identities, and
Self-Regulation.” Pp. 56-79 in J. L. Perry and A. Hondeghem, eds, Motivation in Public
Management: The Call of Public Service. New York et al.: Oxford University Press.
Peters, B. G. 2000. Institutional Theory in Political Science: The New Institutionalism.
London: Continuum.
Putnam, R. D. 1998. “Foreword.” Housing Policy Debate, 9(1): V-viii.
Rainey, H. G. 1982. “Reward Preferences among Public and Private Managers: In Search of
the Service Ethic.” American Review of Public Administration, 16(4): 288-302.
Rainey, H. G. and P. Steinbauer. 1999. “Galloping Elephants: Developing Elements of a
Theory of Effective Government Organizations.” Journal of Public Administration
Research and Theory, 9(1): 1-32.
Scherrer, V. 1987. La France Paresseuse. Paris: Seuill.
Schneider, S. C. and A. D. Meyer. 1991. “Interpreting and Responding to Strategic Issues:
The Impact of National Culture.” Strategic Management Journal, 12(4): 307-320.
Scott, P. G. and S. K. Pandey. 2005. “Red Tape and Public Service Motivation: Findings from
a National Survey of Managers in State Health and Human Services Agencies ” Review of
Public Personnel Administration, 25(2): 155 - 180.
Scott, W. R. 1987. “The Adolescence of Institutional Theory.” Administrative Science
Quarterly, 32(4): 493-511.
Selznick, P. 1984. Leadership in Administration: A Sociological Interpretation. Berkeley, Los
Angeles, London.
Shamir, B. 1991. “Meaning, Self and Motivation in Organizations.” Organization Studies,
12(3): 405 - 424.

26
Skocpol, T., M. Ganz, and Z. Munson. 2000. “A Nation of Organizers: The Institutional
Origins of Civic Voluntarism in the United States.” The American Political Science
Review, 94(3): 527-546.
Stryker, S. 1980. Symbolic Interactionism: A Structural Version. Menlo Park, CA:
Benjamin/Cummings.
Taylor, J. 2007. “The Impact of Public Service Motives on Work Outcomes in Australia: A
Comparative Multi-Dimensional Analysis.” Public Administration, 85: 931-959.
Vandenabeele, W., S. Scheepers, and A. Hondeghem. 2006. “Public Service Motivation in an
International Comparative Perspective: The UK and Germany.” Public Policy and
Administration, 21(1): 13-31.
Vandenabeele, W. 2007. “Towards a Public Administration Theory of Public Service
Motivation: An Institutional Approach.” Public Management Review, 9(4): 545-556.
Vandenabeele, W. and S. Van de Walle. 2008. “International Differences in Public Service
Motivation: Comparing Regions Across the World.” in J. L. Perry and A. Hondeghem, eds,
Motivation in Public Management: The Call of Public Service. New York et. al.: Oxford
University Press.
Waldron, J. 2000. “Cultural Identity and Civic Responsibility.” Pp. 155-174 in W. Kymlicka
and W. Norman, eds, Citizenship in diverse societies. Oxford: Oxford University Press.
Watts, R. J. 1988. “Language, Dialect and National Identity in Switzerland.” Multilingua -
Journal of Cross-Cultural and Interlanguage Communication, 7(3): 313-334.
Weber, M. 2002. The Protestant Ethic and "The Spirit of Capitalism". New York: Penguin
Books.
Wernli, B. 1998. Contraintes Institutionnelles, Influences Contextuelles et Participation aux
Elections Fédérales en Suisse. University of Geneva: Ph. D. Dissertation.
Wright, B. E. 2008. “Methodological Challenges Associated with Public Service Motivation
Research.” Pp. 80-98 in J. L. Perry and A. Hondeghem, eds, Motivation in Public
Management. The Call of Public Service. New York et al.: Oxford University Press.

27
Appendix:
Questionnaire Items:
Public Service Motivation: (Scale: 1[lowest] to 5 [highest]; Midpoint: 3.0)
- APM_1: I am very interested in politics
- APM_2: I like to discuss political subjects with others
- APM_3: The give and take of public policy making appeals to me
- CPI_1: I consider working in public administration a part of my commitment to society
- CPI_2: It is important for me to unselfishly contribute to the common good
- COM_1: I'm often reminded by daily events how dependent we are on one another
- COM_2: I am highly moved by the plight of the underprivileged
- COM_3: Most social programs are too vital to do without
- SS_1: Much of what I do is for a cause bigger than myself
- SS_2: It is important that people give back to society more than they get from it
- SS_3: I am one of those rare people who would risk personal loss to help someone else
Control Variables:
- Gender: 0=male; 1=female
- Age: 1=below 20 year, 2=21-29, 3=30-39, 4=40-49, 5=50-59, 6=above 60
- Leadership position: 0=no, 1=yes
- Salary: 1=salary classes 1-11 (lowest), 2=12-17, 3=18-23, 4=24-29, 5=30-38 (highest)
- Tenure: 1=until 1year, 2=1-5, 3=6-10, 4=more than 10 years
- Full-/Part-Time work: 1=less than 50% employment, 2=50-90%, 3=90-100%
Language and Regional Location Variables:
- French region: Federal Level: 0=German and Italian region; 1=French region; Municipal Level:
0=German region; 1=French region
- Italian region: Federal Level: 0=German and French region; 1=Italian region
- Mother tongue French: Federal Level: 0=German, Italian, and others; 1=French
- Mother tongue Italian: Federal Level: 0=German, French, and others; 1=Italian
Social Context Variables:
- Religious activity: Have you been regularly involved in a religious or spiritual activity during the
last four weeks (e. g. praying, meditation, church attendance)? (0=no, 1=yes)
- Social Capital:
Trust:
- According to me the following groups of people are totally trustworthy: Work colleagues /
People in my neighbourhood / Municipal authorities / Swiss people / Foreigners / Swiss-
Germans / Swiss-French / Swiss-Italians (Scale: 1[lowest] to 5 [highest]; Midpoint: 3.0)

28
Volunteering : (Scale: 0=none, 1=either membership or volunteering/donation, 2=both)
- Of how many associations (e. g. cultural, political, sports) are you a member ? (Numbers :
1 to 8 or 9 and more)
- Have you been volunteering or did you give money to a charity organization during the last
twelve months? (0=no, 1=yes)
Political Participation: (Scale 1[lowest] to 5 [highest]: aggregation of five dummy-variables)
- Have you signed a petition, an initiative or a referendum within the last 12 months? (0=no,
1=yes)
- Have you attended a political assembly within the last 12 months? (0=no, 1=yes)
- Have you attended a demonstration within the last 12 months? (0=no, 1=yes)
- Have you ever been elected for a political mandate (e. g. municipal council, commission)?
(0=no, 1=yes)
- Have you participated in all votes and elections within the last 12 months? (0=no, 1=yes)

29
Figures and Tables:
Study
Organizations
and Sample
Size
Measures of
PSM (No. of
Items)
Correlations between Antecedents and PSM
Bright
2005
U. S. County
Govt. Oregon
(349)
PSM (24)
Education (+); Female (+); Grade (+)
Camilleri
2006 and
Camilleri
2007
Maltese Govt.
(2431)
PSM (24); APM
(3); CPI (5);
COM (8); SS (8)
Salary (+) ; Tenure (+); Age (+), Male (+); Grade (+);
Grade Tenure (+PSM, CPI, COM, SS); Education (+);
Family Life Cycle (+PSM, CPI, COM, SS); Alternative Job
Opp. (- PSM, APM, CPI, SS); Role States (- PSM, APM,
CPI); Perception of Org. (+); Empl.-Leader Rel. (+); Job
Attributes (+); Cust. Focus (+); Entrepreneurship (+APM,
CPI, SS); Leadership (+);
Castaing
2006
French Local
Govt. (754)
CPI (4)
Age (+); Tenure (+); Fulfillment of Expectations (-); OC
(+); Grade (+)
DeHart-
Davis et al.
2006
50 U.S. States
(247)
APM (3); COM
(3); CPI (3)
Education (+ APM, +CPI); Female (+ COM); Professional-
ism (+)
Moynihan
and
Pandey
2007
U. S. State Govt.
(274)
PSM (7); APM
(3); CPI (4)
Education (+); Female (+APM); Professional Identification
(+); Red Tape (-PSM, APM); Reform Orientation (+PSM,
CPI); Hierarchical Authority (+); Tenure (-PSM, APM);
Naff and
Crum 1999
U.S. Federal
Govt. (9710)
PSM (6)
Education (+); Female (+); Minority Employment (+)
Park and
Rainey
2007
U.S. Federal
Govt. (6900)
PSOM (6)
OC (+); Transformational Leadership (+); Goal Clarity (+);
Empowerment (+); Objective Perf. Appraisal Syst. (+);
Proc. Equity Perc. (+); Job Experience (+); Job Satis. (+);
Perc. Perf. (+); Perc. Work Quality (+)
Perry 1997
U. S. Mixed Pop-
ulation (376)
PSM (24); APM
(3); CPI (5);
COM (8); SS (8)
Education (+PSM, CPI); Age (+CPI); Income (-CPI); Male
(+CPI, SS); Church Involvement (-PSM, COM, SS);
Closeness to God (+PSM, CPI, SS); Parental modeling
(+PSM, CPI, SS); Professional Identification (-APM; +CPI,
SS); Liberalism (+APM); Conservatism (-SS)
Perry et al.
2008
U. S. Voluntary
Award Winners
(525)
COM (5); CPI
(3); SS (4)
Family Socialization (+); Religious Activity (+); Formal
Volunteering (+)
Scott and
Pandey
2005
U. S. State Govt.
(274)
APM; CPI; COM
(11 total) [PSM
as Antecedent in
Model]
Global Red Tape (-APM); Personnel Red Tape (-APM,
CPI); Procurement Red Tape (-APM); Reinvention Efforts
(+APM, CPI)
Taylor
2007
Australian State
Govt. (2274)
PSM (2 adapt.)
Education (+); Female (+); Income (-) Intrinsic Rewards
(+); Job Sat. (+); OC (+); Job Locus (+); Union Member-
ship (+)
Explanatory example: The study of De-Hart-Davis et al. was conducted within the 50 states of the U. S. and is based on a
sample of 247 respondents. PSM is measured by three dimensions, each with three items. The results show that education
has a positive relationship with APM and COM etc.
Table 1: Antecedents of PSM in empirical studies

30
Figure 1: Language Regions at cantonal and local levels of Switzerland (Source: Brügger et
al. 2009 and Swiss Federal Statistical Office)

31
N % N % N % N % N %
Gender
Female 3631 29.3 216 20.3 62 13.0 1332 45.8 312 41.2
Male 8473 68.4 843 79.2 409 85.9 1539 52.9 427 56.3
Missing 279 2.3 6 .6 5 1.1 37 1.3 19 2.5
Total 12383 100.0 1065 100.0 476 100.0 2908 100.0 758 100.0
Age
<20 years 6.0 1.1 0.0 102 3.5 13 1.7
20-29 Years 1090 8.8 74 6.9 20 4.2 383 13.2 65 8.6
30-39 Years 3232 26.1 299 28.1 152 31.9 526 18.1 191 25.2
40-49 Years 4241 34.2 367 34.5 203 42.6 811 27.9 235 31.0
50-59 Years 3114 25.1 295 27.7 90 18.9 735 25.3 170 22.4
> 59 Years 366 3.0 20 1.9 5 1.1 230 7.9 43 5.7
Missing 334 2.7 9 .8 6 1.3 121 4.2 41 5.4
Total 12383 100.0 1065 100.0 476 100.0 2908 100.0 758 100.0
No 8138 65.7 663 62.3 362 76.1 1614 55.5 476 62.8
Yes 4015 32.4 395 37.1 108 22.7 1276 43.9 278 36.7
Missing 230 1.9 7 .7 6 1.3 18 .6 4.5
Total 12383 100.0 1065 100.0 476 100.0 2908 100.0 758 100.0
Salary
Class 1-11 490 4.0 68 6.4 54 11.3 577 19.8 176 23.2
Class 12-17 2778 22.4 590 55.4 274 57.6 783 26.9 264 34.8
Class 18-23 4304 34.8 284 26.7 102 21.4 787 27.1 198 26.1
Class 24-29 4135 33.4 104 9.8 29 6.1 421 14.5 68 9.0
Class 30-38 339 2.7 2 .2 2.4 252 8.7 32 4.2
Missing 337 2.7 17 1.6 15 3.2 88 3.0 20 2.6
Total 12383 100.0 1065 100.0 476 100.0 2908 100.0 758 100.0
Tenure
< 1 year 687 5.5 43 4.0 18 3.8 52 1.8 17 2.2
1 - 5 years 2462 19.9 149 14.0 43 9.0 1079 37.1 266 35.1
6 - 10 years 2949 23.8 176 16.5 92 19.3 685 23.6 149 19.7
> 10 years 6097 49.2 694 65.2 319 67.0 980 33.7 294 38.8
Missing 188 1.5 3 .3 4.8 112 3.9 32 4.2
Total 12383 100.0 1065 100.0 476 100.0 2908 100.0 758 100.0
Full-/Part-time
less than 50% 190 1.5 6 .6 2.4 238 8.2 55 7.3
50-90% 2661 21.5 147 13.8 27 5.7 677 23.3 161 21.2
more than 90% 9354 75.5 907 85.2 443 93.1 1972 67.8 538 71.0
Missing 178 1.4 5 .5 4.8 21 .7 4.5
Total 12383 100.0 1065 100.0 476 100.0 2908 100.0 758 100.0
758
476
German
French
Leadership
Position
Municipalities (N=3666)
Federal Administration (N=13924)
2908
12383
1065
Italian
German
French
Table 2: Distribution of control variables among the geographical region of offices

32
Table 3: Mean comparison between language groups

33
Gender (1=male) .006 -.004 -.002 .080 *** .092 *** .057 ** .077 *** .072 *** .073 *** .162 *** .165 *** .140 *** .031 ** .024 * .022
Age .150 *** .156 *** .166 *** .095 *** .111 *** .086 *** .091 *** .094 *** .108 *** -.024 -.020 -.063 ** .119 *** .123 *** .126 ***
Education .066 ** .084 *** .062 ** .101 *** .106 *** .078 ***
Leadership Position (1=yes) .069 *** .078 *** .084 *** .044 .045 -.014 .071 *** .074 *** .082 *** .074 ** .074 ** .008 .074 *** .080 *** .075 ***
Salary Class .106 *** .109 *** .099 *** .109 ** .053 .013 .196 *** .202 *** .194 *** .174 *** .160 *** .120 *** .048 *** .047 *** .046 ***
Tenure -.059 *** -.061 *** -.064 *** .018 .030 .019 -.050 *** -.050 *** -.056 *** -.037 -.034 -.037 -.017 -.017 -.014
Full-/Part-Time -.029 ** -.036 ** -.026 * -.138 *** -.115 *** -.046 * -.042 *** -.045 *** -.036 ** -.117 *** -.110 *** -.056 * .028 ** .023 * .020
Headquarter .037 ** .034 ** .038 ** .041 *** .039 *** .038 *** .026 * .023 * .021
Region German (1=yes) .024 * .032 ** .007
Mother Tongue German (1=yes) .097 *** .056 *** -.058 ***
Region French (1=yes) -.053 *** -.211 *** -.192 *** -.040 *** -.057 ** -.036 * -.026 *
Mother Tongue French (1=yes) -.149 *** -.072 *** .015
Region Italian (1=yes) .000 -.014 .002
Mother Tongue Italian (1=yes) .081 *** .013 .131 ***
Religious Activity .068 *** .020
Trust .164 *** .072 ***
Volunteering .168 *** .084 ***
Political Participation .191 *** .328 ***
R2
adj. R2
∆R2
F
∆F
N
For hierarchical regressions of municipal sample all changes in F are significant at p<0.001 level except for model 2 for APM and COM (p<0.01).
Beta
Beta
Beta
Beta
Beta
Beta
Beta
Beta
Beta
Beta
Beta
Beta
Beta
Beta
Beta
.055
.056
9671
.099
.036
.035
.223
.103
.100
12373
.113
154.863
11486
.225
.221
.116
.099
.227
German
Model 3
Model 3
.097
.036
.060
PSM Federal
PSM Municipal
APM Federal
APM Municipal
German
French
Italian
Model 1
Model 1
Model 3
48.193
81.724
German
French
Italian
.066
.109
.063
.106
French
.096
.036
Model 2
Model 2
.061
58.783
.066
.043
.003
.124
144.430
51.426
26.026***
124.521***
96.272***
2585
2585
2585
11468
10629
42.590***
9.360**
108.207***
2709
8982
12314
11393
9587
2709
2709
.069
.094
.101
.068
.093
.100
94.511
122.042
141.494
.095
Italian
.113
CPI Federal
Table 4: Results of Regression Analyses for PSM and its Subdimensions

34
Gender (1=male) .045 * .061 ** .034 -.118 *** -.129 *** -.126 *** -.054 * -.057 * -.082 *** .017 .009 .021 .071 ** .090 *** .067 **
Age .078 ** .100 *** .084 *** .181 *** .186 *** .179 *** .204 *** .200 *** .195 *** .070 *** .072 *** .089 *** -.009 .019 .011
Education -.040 -.012 -.026 .106 *** .101 *** .093 *** -.026 .006 -.003
Leadership Position (1=yes) .065 ** .065 ** .031 .006 .011 .022 * -.038 -.038 -.061 ** .037 *** .050 *** .061 *** .009 .012 -.013
Salary Class .092 ** .015 -.012 .011 .015 -.002 -.061 * -.047 -.068 * -.022 * -.020 -.025 * .058 -.033 -.051
Tenure .039 .058 * .049 * -.057 *** -.059 *** -.056 *** .027 .023 .014 -.028 ** -.031 ** -.039 ** .032 .053 * .043
Full-/Part-Time -.059 * -.027 .022 -.073 *** -.076 *** -.068 *** -.106 *** -.112 *** -.068 ** -.003 -.009 .000 -.078 ** -.038 .002
Headquarter .004 *** .012 .029 ** .026 * .037 **
Region German (1=yes) .017 .001
Mother Tongue German (1=yes) .005 .239 ***
Region French (1=yes) -.301 *** -.291 *** -.046 *** .055 ** .062 ** -.045 *** -.354 *** -.345 ***
Mother Tongue French (1=yes) -.049 *** -.296 ***
Region Italian (1=yes) .013 .024
Mother Tongue Italian (1=yes) .103 *** .015
Religious Activity .051 ** .073 *** .053 **
Trust .135 *** .130 *** .093 ***
Volunteering .123 *** .099 *** .137 ***
Political Participation .091 *** .036 .028
R2
adj. R2
∆R2
F
∆F
N
For hierarchical regressions of municipal sample all changes in F are significant at p<0.001 level except for model 2 for APM and COM (p<0.01).
Beta
Beta
Beta
Beta
Beta
Beta
Beta
Beta
Beta
Beta
Beta
Beta
Beta
Beta
Beta
2715
.053
2647
.168
2705
.116
.120
Model 3
Model 3
2715
.183
.186
Model 3
German
.012
2647
.046
.134
.043
.131
.046
.088
18.544***
274.011***
43.988***
CPI Municipal
Model 1
Model 3
Model 2
2715
COM Federal
German
French
Italian
.048
.087
.046
69.928
103.068
52.616
12322
11405
9648
COM Municipal
Model 1
Model 3
.078
.081
.076
.078
Model 2
.078
.003
.039
2705
32.697***
8.737**
29.912***
2705
French
Italian
.067
.054
.011
95.767
62.040
12.538
11822
10924
9284
SS Municipal
Model 1
Model 3
Model 2
SS Federal
.009
.164
.049
.088
.047
.068
.055
.011
.132
.130
2647
.121
.036
4.340***
367.821***
28.436
.011
Table 5: Results of Regression Analyses for PSM and its Subdimensions (cont.)

35
Beta p Beta p Beta p Beta p Beta p Beta p Beta p Beta p
Gender (1=male) -.061 .004 -.063 .003 .082 .000 .081 .000 .082 .000 .085 .000 .032 .107 .036 .075
Age .023 .319 .022 .348 -.022 .358 -.023 .342 .075 .002 .078 .001 .107 .000 .110 .000
Education -.016 .401 -.018 .344 .013 .511 .012 .551 .036 .068 .040 .042 .063 .001 .067 .000
Leadership Position (1=yes) -.036 .117 -.036 .116 .050 .033 .050 .033 .075 .001 .075 .001 .167 .000 .167 .000
Salary Class -.004 .897 .002 .941 .078 .009 .081 .007 .058 .048 .047 .117 .110 .000 .098 .000
Tenure -.038 .085 -.040 .075 .024 .294 .023 .311 .043 .055 .046 .042 -.008 .713 -.005 .822
Full-/Part-Time .111 .000 .108 .000 -.124 .000 -.125 .000 -.141 .000 -.136 .000 -.123 .000 -.117 .000
Region French (1=yes) .024 .178 .012 .497 -.048 .009 -.051 .003
R2
adj. R2
∆R2
∆F
N
All changes in F are significant at p<0.01 level except for model 2 for volunteering and political participation (p<0.01) and for religious activity and trust (n.s.).
.000
.054
.056
.052
.054
.054
.002
.101
.104
.099
.102
3179
8.884
51.069
3179
.101
.003
Political Participation
Model 1
Model 2
23.892
2935
Volunteering
Model 1
Model 2
6.914
3001
10.020
0.461
3001
5.182
1.811
3130
3130
.011
.001
.011
.012
.009
.010
Model 2
Model 2
Model 1
Religious Activity
Trust
Model 1
.023
.021
.023
.020
.023
2935
Table 6: Results of Hierarchical Regression Analyses for Social Capital