Midterm Assignment

40 Public Administration Review • January | February 2007

In seeking to explain the antecedents of public service motivation, James Perry focuses on the formative role of sociohistorical context. Th is study tests Perry’s theory and examines the role that organizational factors play in shaping public service motivation, based on responses from a national survey of state government health and human service managers. Th e fi ndings support the role of sociohistorical context, showing that public service motivation is strongly and positively related to level of education and membership in professional organiza- tions. Th e results also underscore the signifi cant infl uence of organizational institutions, indicating that red tape and length of organizational membership are negatively related to public service motivation, whereas hierarchical authority and reform eff orts have a positive relationship. Th erefore, public organizations have both an opportunity and a responsibility to create an environment that allows employees to feel they are contributing to the public good.

Although it is of recent vintage, the concept of public service motivation (PSM) represents a positive example of theory development in public administration. Th is theory has signifi cant practical relevance, as it deals with the relationship between motivation and the public interest. Th e con- struct of the public interest is central to traditional public administration scholarship ( Appleby 1945 ; Herring 1936). In recent years, this theoretical devel- opment has been gradually joined by empirical work as scholars have sought to operationalize what public interest means for employees, why they develop a strong sense of public service, and how that sense infl uences their behavior (e.g., Alonso and Lewis 2001 ; Brewer and Selden 1998; Brewer, Selden, and Facer 2000; Crewson 1997 , Houston 2000; Perry 1996, 1997 ).

Despite such research, there remains a need for more empirical work to validate and test this theory, which has continued to expand. Th us far, much of the research has focused on establishing the existence of PSM, usually by comparing employee motivations across sectors. Th e relevance of much of these fi ndings

remains in dispute because of the use of diff erent or indirect measures of PSM and incomplete theories of how PSM occurs and the eff ects it generates. A series of publications by James Perry removed a great deal of ambiguity about the theoretical and empirical ap- proaches that are appropriate for studying PSM. Perry carefully devised a series of scales to measure PSM in 1996 and provided empirical evidence on the causes of PSM in 1997. In 2000, he articulated the most comprehensive theory of the causes of PSM thus far, identifying sociohistorical context as a primary infl uence.

Our model adds to the limited empirical research on PSM by partially testing the theory proposed by Perry (2000) and deepening that theory by focusing greater attention on the role of organizational institutions. In extending Perry’s model to account for the eff ect of organizational institutions on PSM, we employ a number of organizational variables: organizational culture, red tape, hierarchy, reform orientation, and length of organizational membership. We also include in the model some salient sociohistorical factors iden- tifi ed by Perry (level of education and membership in professional organizations), along with a number of demographic controls (age, gender, and income). We test this model on two of Perry’s four dimensions of PSM (attraction to policy making and commitment to public interest/civic duty) exhibited by a sample of health and human service managers in the 50 state governments.

The Importance of Public Service Motivation Th e literature on PSM has been thoroughly reviewed elsewhere (see, in particular, Brewer and Selden 1998; Perry 2000; Perry, Mesch, and Paarlberg 2006). Th erefore, we simply highlight the major contribu- tions to underline the importance of the topic.

Perry and Wise provide the widely accepted defi nition of PSM: “an individual’s predisposition to respond to motives grounded primarily or uniquely in public institutions and organizations” ( Perry and Wise 1990,

Donald P . Moynihan University of Wisconsin –Madison

Sanjay K . Pandey University of Kansas

Th e Role of Organizations in Fostering Public Service Motivation

Donald P. Moynihan is an assistant

professor of public affairs in the Robert M.

La Follette School of Public Affairs,

University of Wisconsin – Madison. His

research analyzes public management

capacity and reform, especially in the area

of performance management.

E-mail: [email protected] .

Sanjay K. Pandey is an assistant

professor of public administration at the

University of Kansas. He conducts research

in public management and health policy.

Most recently, he directed Phase II of the

National Administrative Studies Project.

E-mail: [email protected] .

Essays on Work Motivation and the Workplace

The Role of Organizations in Fostering Public Service and Motivation 41

368 ). 1 As Brewer, Selden, and Facer (2000) note, PSM is important not just to motivation but also to pro- ductivity, improved management practices, account- ability, and trust in government, making it one of the major topics of investigation in public administration today. Th e relevance of PSM, though particularly high for government, is not limited to this sector, as employees in the private and nonprofi t sectors also exhibit PSM to varying degrees ( Wittmer 1991 ). Th erefore, PSM not only helps us understand the traditional diff erences between the public, nonprofi t, and private sectors, but, given the increasingly blurry boundaries between sectors, it is useful in understand- ing public-regarding behaviors in organizations that are characterized by varying levels of publicness ( Bozeman 1987 ).

Although some of the empirical work off ers evidence of no diff erence between public and private organiza- tions on extrinsic and intrinsic motivators ( Buchanan 1975; Gabris and Simo 1995 ), the bulk of the empiri- cal evidence supports the existence of a public service ethic among public employees. Careful investigation of the PSM construct supports its validity (Brewer, Selden, and Facer 2000; Coursey and Pandey, forth- coming; Perry 1996, 1997 ). Support for PSM can also be found in cross-sector comparisons. Rainey (1982) , for example, found that public and private managers showed signifi cant diff erences in their perceptions of the importance of diff erent types of rewards. In con- trast to private managers, public managers regarded public service and work that is helpful to others as important, whereas higher pay, status, and prestige were less important. Wittmer (1991) came to a similar conclusion in his examination of reward preferences among government, hybrid, and business sector managers. 2

Th us, PSM provides a theory of motivation that links the pursuit of the public interest with administrative behavior. Perry and Wise (1990) argue that individu- als with a high sense of public interest are more likely to select public service careers, an assertion that is supported by evidence of diff erent levels of PSM between the public and private sectors (Houston 2000; Rainey 1982; Wittmer 1991 ). Having joined an organization, members with high levels of PSM appear to contribute in positive ways: Th ey are more willing to engage in whistle-blowing to protect the public interest (Brewer and Selden 1998); they exhibit higher levels of organizational commitment ( Crewson 1997 ); they believe that their jobs are important, which, in turn, leads them to work harder (Wright 2003); they are more likely to be high performers and enjoy higher job satisfaction; and they are less likely to leave their jobs ( Naff and Crum 1999 ). 3

In relation to the motivation literature, Perry (2000) asserts the importance of PSM as an alternative to

rational and self-interested theories of motivation, which tend to focus on pecuniary rewards. It is also possible to illustrate how PSM can shape beliefs and behavioral outcomes. Th us far, research on the sources of PSM has pointed to institutions, with Perry focus- ing particularly on the role of sociohistorical institu- tions. Earlier work (Perry 1997) confi rms the infl uence of education, family, and religion in shaping PSM. Th e next section further investigates Perry’s process theory of PSM and our contribution to it.

Extending Public Service Motivation Theory: The Role of Organizations Perry’s (2000) process theory of PSM off ers the most signifi cant theoretical development in the topic since Perry and Wise (1990) . Th e theory argues that indi- vidual behavior is not just the product of rational, self-interested choices but is rooted in normative and aff ective motives as well. Simply studying motivation from a rational, incentive-driven perspective provides only a partial understanding of motivation. We also need to study the social processes that shape individu- als’ normative beliefs and emotional understandings of the world. Similar to Perry (1997, 2000), our analysis shares a focus on the institutional shapers of indi- vidual beliefs and behavior, but, rather than focus on sociohistorical institutions, we examine organizational institutions. Indeed, Perry explicitly calls for such a study: “Investigation of organizational infl uences should seek to assess the eff ects of organizational experiences and policies on the public service motiva- tion of members over time” (1997, 193).

Perry (2000) points to March and Olsen’s (1989) work on institutions as a theoretical basis for asserting that institutions foster a logic of appropriateness in the minds of individuals — rather than a more rational- choice logic of consequentiality — which causes them to develop PSM. March and Olsen defi ne insti- tutions as “collections of interrelated rules and rou- tines that defi ne appropriate action in terms of relations between roles and situations” (1989, 21). 4 Th is broad defi nition of institutions not only con- forms to the idea of sociohistorical institutions — March and Olsen explicitly consider the role of education and socialization — but also is consistent with the concept of organizational institutions. Th e emphasis on rules and routines suggests that both formally mandated and informal aspects of organiza- tions provide institutional infl uences. 5

Our key theoretical contribution to Perry’s work is to extend the understanding of institutions that shape PSM to include organizational institutions. We argue that work-related rules and norms are organizational institutions that shape not only the administrative behavior of public servants but also the basic attitudes that these actors hold about the value of public service. 6 Our theoretical model, therefore, does not

42 Public Administration Review • January | February 2007

seek to directly contradict Perry’s theory but rather to test whether one aspect of that theory — the work environment — deserves closer attention as a signifi - cant predictor of employee beliefs. Essentially, we argue that PSM may be formed by sociohistorical factors before employees enter the organization, but it will also be infl uenced by the organizational environ- ment in which employees fi nd themselves. 7

Our approach is consistent with the theoretical prem- ises that Perry (2000) places at the heart of PSM theory: that rational, normative, and aff ective pro- cesses motivate humans; that people are motivated by their own self-concept; that preferences or values should be endogenous to any theory of motivation; and that preferences are learned in social processes. Public organizations are not just a means to produce outputs; they are also social institutions in which individuals interact and infl uence each other in the context of a structured environment. 8 Conformity to organizational norms is likely a clear instance of the endogenous nature of individual preferences. March and Olsen (1989) observe that expectations, prefer- ences, experiences, and interpretations of the actions of others are all constructed within institutions. Actors construct beliefs and behaviors based on what is appropriate in light of their environment and the norms of behavior of those around them. Th erefore, we expect that public employees’ beliefs about public service are at least partly infl uenced by the nature of the organizations they are a part of.

Sociohistorical Context Perry (2000) argues that PSM depends on how individuals are socialized through sociohistorical institutions — primarily parental relations, religion, observational learning and modeling during the course of their lives, education, and professional training. Perry (1997) tested the eff ect of some of these factors on fi ve dependent variables: a composite measure of PSM and four subscales (attraction to policy making, commitment to public interest/civic duty, compassion, and self-sacrifi ce). In his 1997 analysis, Perry found that religion had a mixed impact on PSM, with church involvement being negatively related to PSM and closeness to God being positively related to PSM. A third religion variable, religious worldview, had no signifi cant impact in either direc- tion. He found similarly mixed results for parental relations: Th e extent to which respondents had learned altruistic or helping behavior from their parents had a strong positive eff ect on overall PSM and civic duty in particular. Th e degree to which respondents had good relations with at least one parent did not have a strong impact on PSM.

Perry (1997) found that education had a signifi cant positive relationship with his overall measure of PSM and two subscales, commitment to public interest/

civic duty and compassion. In the 1997 model, educa- tion was included as a demographic control, but in formulating his process theory of PSM, Perry recog- nized the primary role of education in shaping beliefs, fi nding that the level of education was signifi cantly and positively related to the overall PSM scale and all of the subscales, with the exception of compassion. Th is leads to our fi rst hypothesis:

H 1 : Level of education is positively related to PSM.

Th e second aspect of education outlined by Perry is professional identifi cation through membership and active involvement in a professional organization. To a greater degree than level of education, which is typi- cally completed prior to joining an organization, professional membership is a type of socialization that parallels the employee’s experience in the organization. Professional organizations seek to help their members make sense of and succeed in their organizational environments. Th erefore, we expect membership in professional organizations to have a strong infl uence on PSM.

Perry (1997) confi rms the strength of this infl uence, but the direction of the relationship remains unclear. Although he had hypothesized that professional identifi cation socializes an individual to ethical behavior — and therefore should be positively related to PSM — Perry found that it had a mixed eff ect on PSM. Overall, it had no signifi cant relationship to the composite PSM scale; was negatively and signifi cantly related to attraction to policy making; was positively and signifi cantly associated with commitment to public interest/civic duty and self-sacrifi ce; and had no signifi cant relationship with compassion. Perry argues that these are among the most interesting of his fi ndings, raising questions about a confl ict between professional values and a tolerance for politics. Given the relevance of these mixed fi ndings, we retest them here without specifying the direction of the relation- ship between professional membership and PSM:

H 2 : Membership in a professional society aff ects an employee’s PSM.

Organizational Institutions To examine organizational institutions, we survey employees’ perceptions of their work environments in the areas of culture, hierarchical authority, red tape, and reform orientation. We also test the eff ects of length of organizational membership.

As Barnard (1938) observes, there are a variety of formal and informal mechanisms through which organizations may shape the beliefs and behavior of their members. Any description of informal institu- tions inevitably brings us to the concept of culture.

The Role of Organizations in Fostering Public Service and Motivation 43

Th e infl uence of organizational norms on beliefs and behavior is widely asserted in the organizational culture literature. Organizational culture shapes beliefs and practices ( Kaufman 1960; Schein 1992 ) and other aspects of administrative behavior, including patterns of interorganizational interaction and reform ( Ban 1995 ), implementation ( Ginger 1998 ), the potential for learning ( Mahler 1997 ), and entrepre- neurship (Moon 1999).

To examine the role of culture, we employ Quinn and Kimberly’s (1984) categorization of distinct value types: group, developmental, hierarchical, and rational cultures. Group cultures are associated with a focus on people rather than the organization, fl exibility rather than control, employee cohesion, and morale. Devel- opmental cultures are associated with a focus on the organization, fl exibility, adaptability and readiness, growth, and resource acquisition. Hierarchical cultures tend to focus on people, control, manage- ment of information and communication, and organizational stability. Finally, rational cultures are associated with organizational goals, control, plan- ning, goal setting, production, and effi ciency. All organizations are likely to exhibit these types to varying degrees rather than simply fall into one type or another ( Zammuto and Krakower 1991 ).

Th erefore, we test the infl uence of the presence of each type of organizational culture on PSM. Given the diffi culty of measuring culture — and the absence of discussion of organizational culture in previous work on PSM — our proposals here are clearly explor- atory, and we do not specify the direction of the rela- tionship between culture and PSM, with one exception. Given that hierarchical cultures tend to emphasize rule-based control of employees (like tradi- tional bureaucracies) and bureaucratic personality ( Merton 1940 ), we expect hierarchical culture to be negatively associated with PSM:

H 3 : Organizational culture aff ects an employee’s PSM.

H 4 : Employees who experience a hierarchical culture have lower levels of PSM.

Th e defi nition of red tape off ered by Bozeman — “rules, regulations, and procedures that remain in force and entail a compliance burden, but do not advance the legitimate purposes the rules were intended to serve” (2000, 12) — underscores the confl ict between red tape and the emphasis on actively serving the public interest that characterizes PSM beliefs. Early work (e.g., Goodsell 1994 ; Kaufman 1977) tended to off er a spirited defense of red tape as a “procedural safeguard” in the public sector without specifying what was meant by red tape, thus limiting further research on the topic. Recent work by

Bozeman and colleagues (e.g., Bozeman 1993, 2000; Bozeman and Scott 1996; Pandey and Kingsley 2000; Pandey and Scott 2002; Pandey and Welch 2005 ; Rainey, Pandey, and Bozeman 1995) has provided clearer conceptual and measurement specifi cations for the red tape concept. Th is defi nition more closely matches the perceptions of individual practitioners: Not all formal rules are red tape, just those that frustrate employees in achieving their goals.

Examinations of the relationship between red tape and individual beliefs detect a greater tolerance for red tape among employees with lower job satisfaction, greater personal alienation, and higher insecurity, pessimism, and mistrust ( Bozeman and Rainey 1998; DeHart-Davis and Pandey 2005; Pandey and Kingsley 2000 ; Rainey, Pandey, and Bozeman 1995). Th ese characteristics are at odds with the profi le of employees with high levels of PSM, who are charac- terized by higher levels of performance, achievement, organizational commitment, and job satisfaction (Brewer and Selden 1998; Crewson 1997; Perry and Wise 1990 ). 9 Indeed, in explaining lower job involve- ment on the part of public employees compared to their private sector counterparts, Buchanan (1975) points to employee frustration with red tape as the primary cause. Th erefore, we expect that employees who perceive themselves as working in an organization dominated by red tape will have lower levels of PSM:

H 5 : Employees who experience red tape have lower levels of PSM.

If public sector stereotypes such as bureaucratic personality and entrenched red tape are indeed responsible for reducing PSM, we expect that reform eff orts directly targeted at battling these maladies will improve PSM. Recent reform eff orts falling under the banner of the reinvention movement have adopted this approach. Reinvention explicitly assumes that employees wish to provide valuable public services and, with the removal of red tape and constraints, can be more eff ective in doing so ( Gore 1993; Moynihan 2006 ). To public managers, the adoption of reforms commu- nicates that leaders are intent on pursuing change that allows workers to do their jobs eff ectively (e.g., Dilulio 1990). A reformist orientation, therefore, is likely to reduce employee cynicism about existing barriers to public service and give employees hope for greater freedom to operate in a way that is consistent with their conception of public service. Naff and Crum (1999) have already found a relationship between PSM and reinvention eff orts at the federal level, and we extend this research to the state level:

H 6 : Employees who experience employee- friendly organizational reforms that seek to cut red tape and empower employees display higher levels of PSM.

44 Public Administration Review • January | February 2007

Increased levels of hierarchy are associated with many of the eff ects of red tape, frustrating the ability to achieve goals, and therefore might be expected to have a similarly negative eff ect on employee outcomes. However, we control for the primary negative eff ects of hierarchy through the inclusion of a red tape vari- able and a measure of hierarchical culture based on an emphasis of formal rules and procedures. Controlling for the dysfunctional aspects of hierarchy prompts us to consider its functional purposes and the reason why it has persisted. Wildavsky (1990) points to a ten- dency in public administration scholarship toward a distaste for hierarchy — in part because of the inequal- ity it invokes — while taking for granted its virtues of stability, continuity, and predictability. Hierarchical levels provide a means to manage the complexity of large organizations, enabling managers to undertake actions that are consistent with task complexity and the time span required ( Jaques 1990 ). Levels of hierar- chy also provide a way to manage accountability in a structured fashion. Th ese issues are likely to be of major concern for our respondents — mid-level to senior-level managers — as they attempt to successfully contribute to the delivery of public services. By con- trast, reformers who criticize hierarchical levels tend to focus on the perspective of frontline workers and urge the removal of mid-level managers as a means of reducing hierarchy and its associated problems ( Gore 1993 ). Given the competing claims about the benefi t of hierarchy, we propose the following:

H 7 : Th e number of hierarchical levels in an organization aff ects employee levels of PSM.

Finally, we test the impact of length of organizational membership on PSM. It has been argued that over time, organizational loyalty and commitment are strengthened as a result of organizational membership ( Romzek 1990 ). Crewson (1997) found that seniority and pay grade, both correlates of length of organiza- tional membership, had a positive relationship to organizational commitment. However, there are confl icting arguments about the eff ect of length of employment on PSM. A basic assumption of most reforms is that bureaucratic forms of government sap the enthusiasm of well-meaning workers ( Gore 1993 ). Merton (1940) argues that bureaucratic organizations lead employees to adopt a bureaucratic personality that is characterized by goal displacement as rule observance replaces the original purposes of the orga- nization. For most public positions — and certainly health and human services — the basic goal of helping citizens is consistent with PSM. However, from em- ployees’ perspective, these goals may become less clear the longer they remain with the organization because they must adapt their everyday work patterns to the demands of the bureaucracy, becoming part of and directing management systems that are enmeshed in a web of rules and reporting requirements. In addition,

employees may gradually recognize the ineff ectiveness of the organization in achieving its goals, further undermining PSM. Members who joined an organiza- tion with a strong commitment to public service may fi nd themselves increasingly frustrated as time passes, as their hopes to contribute are dashed (Romzek and Hendricks 1982). In short, length of organizational membership refl ects the cumulative eff ects of being in a particular work environment over time, both good and bad. Given these confl icting arguments, we propose the following:

H 8 : Length of organizational membership aff ects employee levels of PSM.

Demographic Controls In addition to the foregoing variables, we employ a number of demographic controls: gender, age, and income. Th ese controls led to some unexpected results in Perry’s 1997 study, and consequently, there is value in seeking to replicate his fi ndings. Perry found, at least in relation to commitment to public interest/ civic duty, that men were more likely to have higher levels of PSM than women. Contrary to one hypoth- esis, Perry found that higher levels of individual in- come were negatively associated with PSM among employees, particularly in terms of commitment to public interest/civic duty. He argues that this may be the result of a growing class separation between the wealthy and other classes, leading to a reduced sense of civic involvement on the part of wealthy individu- als. Finally, he found that age had, as hypothesized, a positive impact on PSM, although it is not clear whether this is the result of a generational diff erence between respondents or the diff erent environments faced by respondents of diff erent ages, given the inclu- sion of students in the survey. Our sample is made up exclusively of public employees, and therefore we can employ length of organizational membership as a variable, in addition to the age variable, to better understand the nature of this relationship.

Data, Methodology, and Measurement Th e data for this study were collected as part of Phase II of the National Administrative Studies Project (NASP-II) during 2002 – 03. Th e response rate for the survey was approximately 53 percent (274 responses). Additional detail on the data-collection process is included in appendix 1 and in Pandey (2003) .

Th e theoretical population of interest for this study consisted of managers engaged in information management activities at state-level primary health and human service agencies. Consistent with Caudle (1990) , we employed a broad defi nition of information management, including not only those who manage information systems applications but also managers involved in research and evaluation, managers dealing with public information and communication,

The Role of Organizations in Fostering Public Service and Motivation 45

and top-level program administrators. Th e average age of the managers we surveyed was almost 50 years, almost half were women, the average length of stay in the organization was more than 15 years, and the average salary was $50,000 – $75,000 (see Table 1). Th is demographic description sounds more consistent with our expectations of a typical career manager in a health and human service agency, not the popular image of information technology staff as young men who frequently change employment. However, as with any survey of a particular group, caution should be exercised in generalizing study fi ndings.

Measurement Th is section discusses how we measured the more complex variables employed in this analysis. A full listing of the questions employed, the original source of these questions, and the Cronbach’s alphas of the scales employed can be found in appendix 2.

Th e survey included three of Perry’s four measures of PSM: attraction to policy making, commitment to public interest/civic duty, and compassion. As survey questions were being selected, a deliberate choice was made to exclude questions pertaining to the self- sacrifi ce dimension, largely because it was not in- cluded in the original conception of PSM and because of its conceptual similarity to and overlap with the compassion dimension. In constructing scales for each item, we used a selection of the Likert-scale items suggested by Perry (1996) . In this analysis, we chose not to employ the compassion variable because of concerns over the internal consistency of the scale. Th e scale employed for this variable had a Cronbach’s alpha of .40, below minimally acceptable standards. Perry (1996) reports alpha measures ranging from .69 to .74 for the PSM scales that he developed. Th e diff erences in the scale reliability report may be the

result of a diff erent group of respondents: We surveyed managers working in the public service in the areas of health and human services, whereas Perry administered his survey to a wider variety of respon- dents, including graduate and undergraduate students, university employees, public employees, and managers in diff erent public organizations. Another explanation is that we employed only a subset of the compassion survey questions (four of the eight items in Perry’s compassion scale), and the alpha is dependent on the number of items in the scale. For the attraction to public policy making scale, we employed all three items used by Perry, obtaining a Cronbach’s alpha of .72. For the civic duty scale, we used four of the fi ve items employed by Perry, resulting in a Cronbach’s alpha of .67. Our overall PSM scale, made up of the attraction to public policy making and civic duty scales, had a Cronbach’s alpha of .67.

Our measure of culture examines the perceived values and norms of the organization directly in terms of shared social meanings. Measuring culture has proved diffi cult (Schein 1996), partly because defi ning orga- nizational culture along a unidimensional scale fails to capture the complexity of competing cultures in al- most all organizations. Th is complexity gave rise to Quinn and Rohrbaugh’s (1981) categorization of distinct value types, which argues that organizational approaches to performance vary along three dimen- sions: (1) organizational focus (people versus the organization), (2) diff ering preferences about structure (control versus fl exibility), and (3) diff erent foci on important organizational processes and outcomes (means and ends). Quinn and Kimberly (1984) categorize the four cultural types emerging from this model as group, developmental, hierarchical, and rational cultures. Th ese types were converted into survey questions by Zammuto and Krakower (1991) and adapted for use in the NASP-II survey.

Attitudes toward structural attributes — such as the degree of formalization and the distribution of power and discretion in the organization — are shaped by the way employees perceive formal rules or entrenched norms that communicate organizational values and practices. Th erefore, our measures of red tape, hierar- chical authority, and reform orientation refl ect em- ployee attitudes about these more structural aspects of the organization.

For the measure of reform orientation, reforms such as contracting out or privatization may produce a negative eff ect on employee morale. Th erefore, we employed a measure of reform that excludes downsiz- ing or privatization but includes reinvention reform strategies that we expect will capture a sense of purpose and action that is consistent with our hypoth- esis by enhancing a focus on results and managerial authority (Moynihan 2006). In testing the existence

Table 1 Descriptive Statistics

Variables Mean Standard Deviation

Public service motivation 25.0554 4.1007 Attraction to policy making 9.6630 2.7420 Commitment to public interest/civic duty 15.4022 2.5467 Education 2.4142 0.6506 Professional identifi cation 0.4964 0.5009 Group culture 3.5693 1.0076 Developmental culture 2.8723 1.1460 Hierarchical culture 3.5036 1.0902 Rational culture 2.7701 1.0245 Red tape 6.4210 1.9824 Reform orientation 18.3059 6.8975 Hierarchical authority 6.0239 2.1785 Length of organizational membership 15.4388 10.6607 Age 49.8910 7.6192 Income 2.4335 0.8394 Gender (Female) 0.4664 0.4998

46 Public Administration Review • January | February 2007

of a reform orientation, we employed a preexisting measure of reinvention (Brudney, Hebert, and Wright 1999). Our measure gauges the extent to which workers observed the implementation of reinvention eff orts during the previous four years, assuming that high levels of implementation raise PSM.

Th e models tested in the study satisfi ed ordinary least squares regression assumptions; therefore, we esti- mated the eff ects of each independent variable on PSM while holding the other independent variables constant. 10 Th e descriptive statistics for the data are reported in table 1 .

Findings and Discussion Th e explanatory power of our models, expressed as R 2 , ranged between .15 and .22, levels that are consistent with the results of Perry (1997) (see table 2). Th e results of our analysis off er strong support for the importance of the sociohistorical context but also recognize the infl uence of organizational institutions on the aspects of PSM tested here — attraction to policy making and commitment to public interest/ civic duty. Although our culture measures did not prove signifi cant, we found strong support for other organizational variables and, therefore, for our general premise that the organization is an important venue for shaping PSM. Th e existence of red tape appears to reduce PSM. On the other hand, the perception that the organization is actively implementing reforms is a positive and signifi cant predictor of PSM for the overall commitment to public interest/civic duty scale, and the measure of levels of hierarchical authority is a positive predictor of PSM across all scales. For our

sample, length of organizational membership was negatively associated with PSM, signifi cantly for the overall measure of PSM and attraction to policy making and slightly above generally accepted signifi - cance levels for commitment to public interest/civic duty. None of the demographic controls proved signifi cant, with the exception that women were more attracted to policy making, counter to Perry’s (1997) fi nding.

Th e most powerful predictors of PSM are the socio- historical variables: higher levels of education and professional membership. Th e results reinforce earlier fi ndings (Perry 1997) on the positive relationship between level of education and PSM. Given the clarity and strength of these results, future research on this topic might benefi t from examining how the educational process imparts values of PSM rather than simply measuring the level of education.

Th e results for professional identifi cation were also strong and uniformly positive. Th is is partly inconsis- tent with Perry’s (1997) fi ndings that professional identifi cation had no overall positive eff ect and was a negative infl uence for attraction to policy making. Th e fi ndings are, however, consistent with Perry’s original hypothesis on the positive role that professional orga- nizations play in shaping the ethical norms of their members. Th e diff erence in results may be attributable to measurement — we used a simple dummy measure of membership in a professional organization, whereas Perry used a scale that tracks activity in and attitudes toward professional organizations. Alternatively, the diff erence may be explained by the wide variety of

Table 2 Regressions to Explain Public Service Motivation

Public Service Motivation Attraction to Policy Making Commitment to Public

Interest/Civic Duty

Independent variables B (std. error) Beta B (std. error) Beta B (std. Error) Beta

Education 1.093 (.398) .172 *** .536 (.274) .127 * .565 (.257) .144 * Professional identifi cation 1.913 (.527) .230 **** .834 (.363) .151 * 1.095 (.341) .213 **** Group culture .332 (.288) .079 .089 (.198) .032 .264 (.186) .102 Developmental culture .117 (.260) .023 −.074 (.179) −.030 .178 (.168) .079 Hierarchical culture −.009 (.273) −.002 −.050 (.188) −.019 .033 (.176) .014 Rational culture −.237 (.264) −.057 −.097 (.181) −.035 −.118 (.171) −.046 Red tape −.431 (.147) −.207 *** −.337 (.101) −.244 **** .087 (.095) −.068 Reform orientation .088 (.039) .145 ** −.034 (.027) −.083 .054 (.025) .145 ** Hierarchical authority .378 (.137) .193 *** .179 (.094) .138 * .203 (.088) .168 ** Length of organizational membership −058 (.025) −.148 * −.030 (.017) −.116 * −.026 (.016) .109 Age .029 (.040) .049 −.030 (.028) .076 −.004 (.026) −.012 Income .347 (.330) .069 .356 (.226) .107 .020 (.213) .007 Gender (Female) .543 (.519) .065 .857 (.357) .155 ** −.343 (.335) −.066

N 237 238 237 R 2 .220 .160 .151 F 7.768 3.303 3.067 Signifi cance .0000002 .0001 .0003

**** Statistically signifi cant at .001; *** statistically signifi cant at .005; ** statistically signifi cant at .01; * statistically signifi cant at .05. Note: Signifi cance levels are one-tailed tests if matching a predicted direction, two-tailed tests otherwise.

The Role of Organizations in Fostering Public Service and Motivation 47

respondents that Perry surveyed, which ranged from undergraduate students to managers and employees in diff erent types of organizations. Our more homog- enous set of respondents in the health and human services fi eld might be expected to have a less varied and more uniform relationship, with a smaller range of relevant professional organizations. Overall, our results suggest that, at least in the area of health and human services, professional identifi cation appears to have a positive eff ect for the aspects of PSM that we measure here.

Th e organizational culture variables were not found to be signifi cant predictors of PSM. Th e null results for all of the measures of organizational culture were somewhat unexpected. Given the results of in-depth case analysis on the role of culture in shaping beliefs, it is hard to dismiss this variable; Kaufman (1960) and Wilson (1989) off er classic examples of how culture shapes beliefs and behavior. Although the validity of the categories and measures that we employed is confi rmed by other studies ( Zammuto and Krakower 1991 ), we acknowledge the diffi culty of measuring cultural attributes across organizations. In addition, as noted before, our scales do not cover all aspects of PSM but instead focus on a limited set of public employees. Th us, our results on culture should be viewed as preliminary, and there is value in testing this relationship with alternative measures of culture, additional survey populations, and the full PSM scale.

Red tape was negatively and signifi cantly related to the overall measure of PSM and attraction to policy making but not to commitment to public interest/ civic duty. Consistent with the message of reform advocates in the public sector, well-meaning employ- ees may become unmotivated in intensely bureau- cratic organizations. Red tape discourages employees from believing they are serving the public good, but active reform eff orts to increase managerial authority and focus on results can reinvigorate PSM among managers. Th e reinvention movement that provides the basis for our reform scale is broadly associated with generally positive reform rhetoric based on assumptions that organizational systems rather than public employees are to blame for public failures and that public employees play a central enabling role in achieving public goals ( Gore 1993; Moynihan 2006 ). For public employees who value public service and may be frustrated by perceived goal displacement in their organizations, such reforms are likely to be viewed positively and reinforce the belief that the organization provides a venue in which employees can fulfi ll their public service motivation. Th e results from our sample suggest that this eff ect is especially pronounced among employees who are interested in policy making.

However, the fi ndings from our sample question recent reforms’ emphasis on the negative eff ect of

excessive hierarchy. We found the perception that an organization has many hierarchical levels to be associ- ated with higher levels of employee PSM. Th is fi nding runs contrary to the assumptions of reform eff orts that denigrate hierarchy as an outmoded and stifl ing organizational characteristic. Such reforms have advo- cated fl atter organizations that reduce the distance between the top and bottom of the hierarchy, thereby empowering frontline employees. Th e losers in these reform eff orts are mid-level managers who supervise the work of public organizations but are deemed an unnecessary nuisance and targeted for downsizing. However, Wolf (1997) argues that such criticisms are overstated based on a mistaken belief that modern agencies maintain premodern levels of hierarchy. Using a meta-analysis of case studies of eff ective pub- lic organizations, Wolf demonstrates that hierarchy has virtually zero eff ect on agency performance. Th is is because “agencies of the modern era are decidedly less hierarchical than were their predecessors” and “most modern agencies are formalized to a degree appropriate with their mission” ( Wolf 1997, 376 ). Rather than being the hapless victims of an industrial age organizational structure, public managers are likely to fi nd themselves in organizations in which the level of hierarchy is based on an understanding of how best to achieve organizational goals.

As Jaques (1990) argues, hierarchy continues to persist because it remains the most logical and eff ective way to organize the multiple complex tasks undertaken by large groups of individuals while maintaining a system of clear accountability, adding value to work as it moves through the organization, and building consen- sus and acceptance for organizational actions. Few are in a better position to appreciate the benefi ts of hierarchy — and also to overlook its failings — than mid-level managers. Hierarchical levels provide a structure that allows these actors to perceive them- selves as exercising a powerful and positive role in providing public services, but consistent with reason- able levels of complexity and accountability. Mid-level managers in organizations that have adopted a fl atter approach face criticism, may struggle to fully under- stand their role relative to senior managers and frontline employees, and must reconceptualize how to meaningfully add value and oversee a greater number of employees. From a more self-interested perspective, the presence of many hierarchical levels also improves the potential for promotion to a higher level of the organization for these managers. Defending hierarchy may have gone out fashion, but our results off er rare advocacy for the benefi ts that hierarchy provides, not just in terms of making organizations work eff ectively but also in terms of the beliefs of employees. It also helps explain the dogged persistence of the hierarchi- cal form in the face of continuing criticism. We recog- nize, however, that this result may not be generalizable to the PSM of other groups who are less tolerant of

48 Public Administration Review • January | February 2007

hierarchy. If our survey had included frontline employees, nonprofi t staff , or students, the results might have been diff erent: Th is is clearly an empirical ques- tion and worthy of attention in future studies of PSM.

It is somewhat discouraging to fi nd that the longer our respondents were in their current organizations, the lower their levels of PSM, at least with respect to the measure of attraction to policy making and the combined measure of PSM. Th ough it would be tempting to believe that public organizations are eff ective at celebrating and communicating the value of their services to their employees, our respondents reported declining levels of PSM. We are careful, however, to caution that our results come from one functional area in state government and must be considered preliminary. As others have noted, PSM is a dynamic construct, and it may vary considerably across function or level of government (Brewer and Selden 1998). Workers in the provision of health and social services may be particularly likely to grow dis- couraged over time if they see limited progress toward solving the problems of poverty, inadequate health services, and other social issues. Workers in other areas may not perceive such a lack of progress. Another explanation is that PSM may have declined among respondents because of such life-cycle considerations as work or retirement (rather than the individual’s experience with the organization), although it bears noting that our analysis controlled for the eff ects of age as a separate variable. Again, these are clearly salient research questions that are worthy of empirical investigation.

Two additional caveats related to the generalizability of our analysis are worth considering. First, we have analyzed only two of the four elements of the standard PSM scale. Perry (1997) found some signifi cant diff erences in the infl uence of independent variables on the four diff erent aspects of PSM, so it is unclear whether our results would apply to the compassion and self-sacrifi ce elements of PSM. On the other hand, PSM has been consistently articulated as a single concept with related subcomponents, which would argue for the generalizability of our results. Indeed, Perry’s fi rst article on PSM scale development conceives of PSM as a superordinate multidimen- sional construct (Edwards 2001), a conceptualization that is in accord with the intellectual history of the concept in public administration. Furthermore, Coursey and Pandey (forthcoming) provide empirical support for this viewpoint using a latent variable technique. A second caveat relates to the independent variables employed. Clearly, the model presented here does not test all possible variables related to PSM. Because of data limitations, we did not include other sociohistorical factors previously tested, such as reli- gion and parental relations.

Conclusion Th is study has off ered evidence that the PSM of public employees is a result of not only individual sociohistorical background but also the organizational environment in which employees fi nd themselves. Clearly, ours is a preliminary model, and there remains a need for additional analysis of the antecedents of PSM. Nonetheless, the results support the basic argument on the importance of organiza- tional institutions to PSM and thus represent an empirical and theoretical advance over prior knowledge.

Recognizing the impact of organizational institutions on PSM means recognizing an accompanying set of organizational responsibilities. One of the perceived practical benefi ts of PSM is that it both helps recruit individuals into the public sector and strengthens employee ties with the public sector, providing a basis for loyalty, motivation, and commitment that is more eff ective than monetary incentives. But how can organizations seek to foster PSM in the fi rst place?

Public organizations must become attuned to the eff ect that management systems and other organiza- tional institutions have on PSM. Th is research has provided some qualifi ed insights: Reducing red tape and undertaking reform that clarifi es goals and empowers employees can have a positive eff ect on employee PSM. Other research suggests that helping employees feel as if they are meaningfully contribut- ing to organizational goals reduces employee frustra- tion and strengthens commitment (Romzek and Hendricks 1982). Th e element common in these suggestions is the importance of encouraging public employees to feel that they are personally contributing to an organization that performs a valuable service, without unnecessary restrictions or controls on their eff orts. Th e research also suggests the importance of communicating to public employees the centrality of their role in the organization and the real benefi ts that their contribution makes to society — an organiza- tional trait that is currently unfamiliar to many public employees.

Acknowledgments We would like to thank the three anonymous PAR reviewers for making valuable suggestions. Th e data analyzed in this paper were collected under the auspices of the National Administrative Studies Project, which is supported in part by the Forum for Policy Research and Public Service at Rutgers University under a grant from the Robert Wood Johnson Foundation to the Center for State Health Policy, also at Rutgers University. Naturally, this support does not necessarily imply an endorsement of the analyses and opinions presented here.

The Role of Organizations in Fostering Public Service and Motivation 49

Notes 1. Perry and Wise propose three main types of PSM:

rational motives, in which individual utility is based on creating good public policy; norm-based motives, which are based on a general sense of public interest and social equity; and aff ective motives, which are based on emotional responses to social contexts and particular aff ection for specifi c programs. Perry (1996) later tested a number of measurement tools to track PSM, devising scales for attraction to policy making, commitment to public interest/civic duty, and compassion and matching them to the rational, norm-based, and aff ective motives, respectively. Perry added self-sacrifi ce, an additional aspect of PSM that has been identifi ed in discussions of public service.

2. Off ering a ranking of diff erent reward alternatives to respondents, Wittmer found that private sector managers were signifi cantly more concerned with pay, status, and prestige than public or hybrid managers, who placed much greater emphasis on helping others and community service. Using a national sample, Houston (2000) provides further support for public employee reward preferences. However, he also reports that public sector employees place greater emphasis on job security.

3. Alonso and Lewis (2001) off er more mixed evidence on performance. Employing two sepa- rate federal government data sets, they found that PSM is signifi cantly and positively related to performance in one but not the other, identifying the need for greater empirical testing of this claim.

4. Our approach to institutions joins with and extends Perry’s application of the work of March and Olsen (1989, 1995) . However, we are aware that any eff ort to incorporate the term institution into an analytic framework is fraught with risk and disagreement. Th e success of various strands of institutional theory — Hall and Taylor (1996) count three versions, Peters (1999) counts six — encourages a tendency toward overextension ( Hall 1996 ). March and Olsen’s defi nition does point to some common attributes among the diff erent approaches — a distinction between institutions and organizations, with institutions as rules, formal or informal, that are stable over time and have an impact on individual belief and behavior (Peters 1999).

5. For example, in a later work, March and Olsen argue that public servants should be directed not only by formal mandates but also by the need to be socialized into “an ethic of administrative duty and conformity to the law” (1995, 58). Although March and Olsen conceive of the public servant in distinctly Weberian terms, the importance of formal rules and informal socialization assumes a

key role for organizational institutions in shaping the beliefs and behavior of these actors.

6. Perry (2000) does not completely overlook the role of the organization: He specifi es the role of the work environment as a source of infl uence on PSM, noting evidence on the special context of public organizations and the need to develop incentives that are appropriate to the logic of the work setting. He does not, however, specify in detail how diff erent aspects of the public work environment shape PSM, suggesting that the work environment is of secondary importance to such sociohistorical institutions as religion, education, and parental relations.

7. To illustrate, we argue that the PSM diff erences between employees in diff erent sectors are not solely the product of self-selection based on preexisting PSM levels. It may also be the case that the way employees perceive their organiza- tional environments makes them more or less likely to maintain PSM beliefs over time. As Wright (2001) points out, sectoral diff erences may be explained not only by self-selection into suitable jobs but also by a process whereby employees adapt their values to suit the organiza- tional environment (see also Hall, Schneider and Nygren 1970; Rosenberg 1957 ). Th is is consis- tent with March and Olsen’s argument that individual preferences are not exogenous and fi xed but rather endogenous to the institutions in which the individual interacts, shaped by a logic of appropriateness consistent with these institu- tions. Th ere is some empirical evidence in the context of PSM to support this claim. Romzek and Hendricks (1982) suggest that the inability of public organizations to satisfy a desire to serve will lead to disinterest or hostility as PSM turns into frustration. Crewson (1997) found that organizational commitment is partly dependent on the sense of trust and affi liation that workers have with fellow employees.

8. An acceptance of the role of the work environ- ment as a shaper of public employee behavior is central to traditional public administration scholarship, as illustrated by three giants of the fi eld: Gulick (1937) represents an early accep- tance of the central role of formal structures and procedures in shaping employee outcomes, whereas Barnard (1938) points to the role of informal norms. Selznick (1996) notes how deviations from formal structures become embed- ded normative organizational characteristics over time and the sense of value that institutionaliza- tion can infuse in the tasks. Th is acceptance of institutions — although the term institutions is rarely employed in the manner that March and Olsen propose — continues in present scholarship. For example, Wilson’s (1989) widely cited account of bureaucracy identifi es organizational

50 Public Administration Review • January | February 2007

circumstances, culture, and constraints as shapers of administrative behavior. Th e public budgeting process and civil service systems are objects of study that refl ect the way that highly formal and procedural aspects of the public sector interact with entrenched norms of appropriate and expected behavior (Ingraham, Moynihan and Andrews forthcoming).

9. Quite likely, the relationship between red tape and public service motivation is bidirectional. Consistent with our focus on the infl uence of organizational institutions, this study argues that red tape reduces public service motivation. However, some public administration literature focuses on the other causal direction, arguing that managers who are imbued with high levels of public service motivation perceive and work to reduce red tape ( Scott and Pandey 2005 ).

10. To determine whether ordinary least squares was the appropriate estimation technique, we exam- ined our data for heteroskedasticity, multicol- linearity, and infl uential data. A histogram of the standardized residuals shows a normal distribu- tion. A scatter plot illustrates that the errors were relatively constant (homoscedastic) and indepen- dent of one another. We examined the bivariate correlations and the square root of the variance infl ation factor (VIF) to detect multicollinearity ( Fox 1991, 11 ). Th e highest correlations were between red tape and hierarchical levels, which were positively correlated at .466, and between the diff erent measures of culture: bureaucratic and developmental culture were correlated at – .423, and the developmental and group culture were correlated at .406. None of the other cor- relations between the independent variables exceeded 0.4, although length of organizational membership and age were positively correlated at .396. Th e highest VIF for each model was the red tape variable, although the square root of the VIF did not exceed 1.215 in any model.

References Aiken , Michael , and Jerald Hage . 1968 .

Organizational Interdependence and Intra- Organizational Structure . American Sociological Review 33 ( 6 ): 912 – 30 .

Alonso , Pablo , and Gregory B . Lewis . 2001 . Public Service Motivation and Job Performance: Evidence from the Federal Sector . American Review of Public Administration 31 ( 4 ): 363 – 80 .

Appleby , Paul H . 1945 . Big Democracy . New York : Alfred A. Knopf .

Ban , Carolyn . 1995 . How Do Public Managers Manage? Bureaucratic Constraints, Organizational Culture, and the Potential for Reform . San Francisco : Jossey-Bass .

Barnard , Chester. I . 1938 . Th e Functions of the Executive . Cambridge, MA : Harvard University Press .

Bozeman , Barry . 1987 . All Organizations Are Public: Bridging Public and Private Organizational Th eories . Jossey-Bass : San Francisco .

— — — . 1993 . A Th eory of Government “Red Tape.” Journal of Public Administration Research and Th eory 3 ( 3 ): 273 – 304 .

— — — . 2000 . Bureaucracy and Red Tape . Upper Saddle River, NJ : Prentice Hall .

Bozeman , Barry , and Hal Rainey . 1998 . Organizational Rules and Bureaucratic Personality . American Journal of Political Science 42 ( 1 ): 163 – 89 .

Bozeman , Barry , and Patrick Scott . 1996 . Bureaucratic Red Tape and Formalization: Untangling Conceptual Knots . American Review of Public Administration 26 ( 1 ): 1 – 17 .

Brewer , Gene A. , and Sally Coleman Selden . 1998 . Whistleblowers in the Federal Civil Service: New Evidence of the Public Service Ethic . Journal of Public Administration Research and Th eory 8 ( 3 ): 413 – 39 .

Brewer , Gene A. , Sally Coleman Selden , and Rex L . Facer II . 2000 . Individual Conceptions of Public Service Motivation . Public Administration Review 60 ( 3 ): 254 – 64 .

Brudney , Jeff rey L , F .. Ted Hebert , and Deil S . Wright . 1999 . Reinventing Government in the American States: Measuring and Explaining Administrative Reform . Public Administration Review 59 ( 1 ): 19 – 30 .

Buchanan , Bruce II . 1975 . Red Tape and the Service Ethic: Some Unexpected Diff erences Between Public and Private Managers . Administration & Society 6 ( 4 ): 423 – 44 .

Caudle , Sharon L . 1990 . Managing Information Resources in State Government . Public Administration Review 50 ( 5 ): 515 – 24 .

Coursey , David H. , and Sanjay K. Pandey . Forthcoming . Public Service Motivation Measurement: Testing a Shortened Version of Perry’s Scale . Administration & Society .

Crewson , Philip E . 1997 . Public-Service Motivation: Building Empirical Evidence of Incidence and Eff ect . Journal of Public Administration Research and Th eory 7 ( 4 ): 499 – 518 .

DeHart-Davis , Leisha , and Sanjay K . Pandey . 2005 . Red Tape and Public Employees: Does Perceived Rule Dysfunction Alienate Managers? Journal of Public Administration Research and Th eory 15 ( 1 ): 133 – 48 .

DiIulio , John J . 1990 . Managing a Barbed-Wire Bureaucracy: Th e Impossible Job of Corrections Commissioner . In Impossible Jobs in Public Management , edited by Edwin C. Hargrove and John C . Glidewell , 49 – 71 . Lawrence : University Press of Kansas .

Dillman , Don A . 2000 . Mail and Internet Surveys: Th e Tailored Design Method . 2nd ed . New York : Wiley .

Edwards , Jeff rey. R . 2001 . Multidimensional Constructs in Organizational Behavior Research:

The Role of Organizations in Fostering Public Service and Motivation 51

An Integrative Analytical Framework . Organizational Research Methods 4 ( 2 ): 144 – 92 .

Fox , John . 1991 . Regression Diagnostics . Newbury Park, CA : Sage Publications .

Gabris , Gerald T. , and Gloria Simo . 1995 . Public Sector Motivation as an Independent Variable Aff ecting Career Decisions . Public Personnel Management 24 ( 1 ): 33 – 51 .

Ginger , Clare . 1998 . Interpreting Roads in Roadless Areas: Organizational Culture, Ambiguity, and Change in Agency Responses to Policy Mandates . Administration & Society 29 ( 6 ): 723 – 57 .

Goodsell , Charles . 1994 . Th e Case for Bureaucracy: A Public Administration Polemic . 3rd ed . Chatham, NJ : Chatham House .

Gore , Al . 1993 . From Red Tape to Results: Creating a Government Th at Works Better and Costs Less . Washington, DC : Government Printing Offi ce .

Gulick , Luther . 1937 . Notes on the Th eory of Organization . In Papers on the Science of Administration , edited by Luther Gulick and Lyndall Urwick , 191 – 95 . New York : Institute of Public Administration .

Hall , Douglas T , Benjamin Schneider , and Harold T. Nygren . 1970 . Personal Factors in Organizational Identifi cation . Administrative Science Quarterly 15 ( 2 ): 176 – 90 .

Hall , Peter A. , and Rosemary C. R . Taylor . 1996 . Political Science and the Th ree New Institutionalisms . Political Studies 44 ( 5 ): 936 – 57 .

Hall , Richard H . 1996 . Organizations: Structures, Processes, and Outcomes . 6th ed . Englewood Cliff s, NJ : Prentice Hall .

Herring , E. Pendleton . 1936 . Public Administration and the Public Interest . New York : McGraw-Hill .

Houston , David, J . 2000 . Public-Service Motivation: A Multivariate Test . Journal of Public Administration Research and Th eory 10 ( 4 ): 713 – 27 .

Ingraham , Patricia W. , Donald P. Moynihan and Matthew Andrews . Formal and Informal Institutions in Public Administration. Forthcoming in B. Guy Peters and Donald Savoie (Eds.), Institutional Th eory in Political Science . Manchester. Manchester University Press .

Jaques , Elliot . 1990 . In Praise of Hierarchy . Harvard Business Review 68 ( 1 ): 127 – 33 .

Kaufman , Herbert . 1960 . Th e Forest Ranger . Baltimore, MD : Johns Hopkins University Press .

— — — . 1977 . Red Tape: Its Origins, Uses, and Abuses . Washington, DC : Brookings Institution Press .

Mahler , Julianne . 1997 . Infl uences of Organizational Culture in Learning in Public Agencies . Journal of Public Administration Research and Th eory 7 ( 4 ): 519 – 40 .

March , James G. , and Johan P . Olsen . 1989 . Rediscovering Institutions: Th e Organizational Basis of Politics . New York : Free Press .

— — — . 1995 . Democratic Governance . New York : Free Press .

Merton , Robert . 1940 . Bureaucratic Structure and Personality . Social Forces 18 ( 4 ): 560 – 68 .

Moon , M. Jae . 1999 . Th e Pursuit of Managerial Entrepreneurship: Does Organization Matter? Public Administration Review 59 ( 1 ): 31 – 43 .

Moynihan , Donald P . 2006 . Managing for Results in State Government: Evaluating a Decade of Reform. Public Administration Review 66 ( 1 ): 78 – 90 .

Naff , Katherine C. , and John Crum . 1999 . Working for America: Does Public Service Motivation Make a Diff erence? Review of Public Personnel Administration 19 ( 4 ): 5 – 16 .

Pandey , Sanjay K . 2003 . National Administrative Studies Project (NASP-II): A National Survey of Managers in State Health and Human Services Agencies . Camden, NJ : Rutgers University .

Pandey , Sanjay K. , and Gordon A . Kingsley . 2000 . Examining Red Tape in Public and Private Organizations: Alternative Explanations from a Social Psychological Model . Journal of Public Administration Research and Th eory 10 ( 4 ): 779 – 99 .

Pandey , Sanjay K. , and Patrick G . Scott . 2002 . Red Tape: A Review and Assessment of Concepts and Measures . Journal of Public Administration Research and Th eory 12 ( 4 ): 553 – 80 .

Pandey , Sanjay K. , and Eric W . Welch . 2005 . Beyond Stereotypes: A Multistage Model of Managerial Perceptions of Red Tape . Administration & Society 37 ( 5 ): 542 – 75 .

Perry , James L . 1996 . Measuring Public Service Motivation: An Assessment of Construct Reliability and Validity . Journal of Public Administration Research and Th eory 6 ( 1 ): 5 – 22 .

— — — . 1997 . Antecedents of Public Service Motivation . Journal of Public Administration Research and Th eory 7 ( 2 ): 181 – 97 .

— — — . 2000 . Bringing Society In: Toward a Th eory of Public-Service Motivation . Journal of Public Administration Research and Th eory 10 ( 2 ): 471 – 88 .

Perry , James L. , and Lois R . Wise . 1990 . Th e Motivational Bases of Public Service . Public Administration Review 50 ( 3 ): 367 – 73 .

Perry , James L. , Debra Mesch , and Laurie Paarlberg . 2006 . Motivating Employees in a New Governance Era: Th e Performance Paradigm Revisited . Public Administration Review 66 ( 4 ): 505 – 14 .

Peters , B. Guy . 1999 . Institutional Th eory in Political Science: Th e New Institutionalism . New York : Pinter .

Quinn , Robert E. , and John R . Kimberly . 1984 . Paradox, Planning, and Perseverance: Guidelines for Managerial Practice . In Managing Organizational Translations , edited by John R. Kimberly and Robert E . Quinn , 295 – 313 . Homewood, IL : Dow Jones/Irwin .

Quinn , Robert E. , and John Rohrbaugh . 1981 . A Competing Values Approach to Organizational Eff ectiveness . Public Productivity Review 5 ( 2 ): 122 – 40 .

52 Public Administration Review • January | February 2007

Rainey , Hal . 1982 . Reward Preferences among Public and Private Managers . American Review of Public Administration 16 ( 4 ): 288 – 302 .

Rainey , Hal , Sanjay K. Pandey , and Barry Bozeman . 1995 . Public and Private Manager’s Perceptions of Red Tape . Public Administration Review 55 ( 6 ): 567 – 74 .

Romzek , Barbara . 1990 . Employee Investment and Commitment: Th e Ties Th at Bind . Public Administration Review 50 ( 3 ): 374 – 82 .

Romzek , Barbara and , J. Stephen Hendricks . 1982 . Organizational Commitment and Representative Bureaucracy: Can We Have It Both Ways? American Political Science Review 76 ( 1 ): 75 – 82 .

Rosenberg , Morris . 1957 . Occupations and Values . New York : Arno Press .

Schein , Edgar . 1992 . Organizational Culture and Leadership . 2nd ed . San Francisco : Jossey Bass .

— — — . 1996 . Culture: Th e Missing Concept in Organization Studies . Administrative Science Quarterly 41 ( 2 ): 229 – 41 .

Scott , Patrick G. , and Sanjay K . Pandey . 2005 . Red Tape and Public Service Motivation . Review of Public Personnel Administration 25 ( 2 ): 155 – 78 .

Selznick , Philip . 1996 . Institutionalism “Old” and “New.” Administrative Science Quarterly 41 ( 2 ): 279 – 27 .

Wildavsky , Aaron . 1990 . Administration without Hierarchy? Bureaucracy without Authority? In Public Administration: Th e State of the Discipline , edited by Naomi B. Lynn and Aaron Wildavsky , xiii – xix . Chatham, NJ : Chatham House .

Wilson , James Q . 1989 . Bureaucracy: What Government Agencies Do and Why Th ey Do It . New York : Basic Books .

Wittmer , Dennis . 1991 . Serving the People or Serving for Pay: Reward Preferences among Government, Hybrid Sector, and Business Managers . Public Productivity and Management Review 14 ( 4 ): 369 – 83 .

Wolf , Patrick J . 1997 . Why Must We Reinvent the Federal Government? Putting Historical Developmental Claims to the Test . Journal of Public Administration Research and Th eory 7 ( 3 ): 353 – 88 .

Wright , Bradley E . 2001 . Public Sector Work Motivation: Review of Current Literature and a Revised Conceptual Model . Journal of Public Administration and Research Th eory 11 ( 4 ): 559 – 86 .

— — — . 2003 . Toward Understanding Task, Mission, and Public Service Motivation: A Conceptual and Empirical Synthesis of Goal Th eory and Public Service Motivation. Paper presented at the 7th National Public Management Research Conference, Georgetown Public Policy Institute , Washington, DC , October 9 – 11 .

Zammuto , Raymond F. , and Jack Y . Krakower . 1991 . Quantitative and Qualitative Studies of Organizational Culture . Research in Organizational Change and Development 5 : 83 – 114 .

Appendix 1: Data Collection for Phase II of the National Administrative Studies Project Th e theoretical population of interest for this study consisted of managers engaged in information management activities in state-level primary health and human service agencies. Primary health and human service agencies were identifi ed on the basis of the defi nition used by the American Public Human Services Association (APHSA), which includes agencies and housing programs related to Medicaid, Temporary Assistance for Needy Families, and child welfare. Th e sampling frame was developed with the aid of the most widely used directory of human service agency managers, namely, the APHSA directory. Th e application of the study criteria resulted in a sampling frame of 570 managers from the 50 states and Washington, D.C. Given the small size of the sampling frame, a decision was made to administer the survey to the entire sampling frame (i.e., conduct a census).

As with most survey research projects, minimizing nonresponse, both to the survey and to specifi c questionnaire items, was a primary survey administra- tion goal. Dillman’s (2000) comprehensive tailored design method was employed to maximize the response rate. Th is method uses the following elements:

· Questionnaire with well-designed content · Survey questionnaire formatted in accordance with the latest advances in cognitive research · Multiple personalized contacts, each accompanied by a carefully crafted message to encourage the respondent to complete the survey questionnaire · Use of real stamps on return envelopes · Use of features such as a pre-notice letter, a fax message, and a phone call at key points in the survey administration · Use of special delivery (combination of two-day delivery by Airborne Express and Priority Mail service of U.S. Postal Service)

Th e data-collection phase of the study began in the fall of 2002 and concluded in the winter of 2003. Approximately one week following the initial alert letter, the survey questionnaire was mailed to the respondents. About 10 days later, a combination thank you/reminder postcard was sent to all respon- dents, thanking those who had responded and en- couraging those who had not to respond as soon as they possibly could. Nearly a month after this post- card was mailed, a new cover letter and replacement survey were sent to nonrespondents. Th e fi nal step in survey administration took place about two months later, when nonrespondents were sent a new cover letter and a second replacement survey with a request to complete the survey.

The Role of Organizations in Fostering Public Service and Motivation 53

Based on the information collected during this period, the size of the sampling frame was reduced from 570 to 518. Th e principal reason for deletion from the sampling frame was that managers had left the organization before survey administration eff orts. Other reasons for deletion from the sampling frame were retirement and death. By the time the survey administration concluded in the winter of 2003, a total of 274 responses had been received. Th us, the response rate for the study was 53 percent.

Appendix 2: Measurement of Study Variables Attraction to Policy Making (adapted from Perry 1996) Summative index (Cronbach’s alpha = .72) of responses to the following questions, taken from Perry (1996) , where 1 = strongly disagree and 5 = strongly agree.

· Politics is a dirty word. (reversed) · The give - and-take of public policy making doesn’t appeal to me. (reversed) · I don‘t care much for politicians. (reversed)

Commitment to Public Interest/Civic Duty (adapted from Perry 1996 ) Summative index (Cronbach’s alpha = .67) of responses to the following questions, taken from Perry (1996) , where 1 = strongly disagree and 5 = strongly agree.

· I consider public service my civic duty. · Meaningful public service is very important to me. · I would prefer seeing public offi cials do what is best for the whole community even if it harmed my interests. · I unselfi shly contribute to my community.

Rational Culture (adapted from Zammuto and Krakower 1991 ) 1 = strongly disagree, 5 = strongly agree · My agency is very production oriented. · A major concern is with getting the job done. · People aren‘t very personally involved.

Group Culture (adapted from Zammuto and Krakower 1991 ) 1 = strongly disagree, 5 = strongly agree · My agency is a very personal place. · It is an extended family. · People seem to share a lot of themselves.

Developmental Culture (adapted from Zammuto and Krakower 1991 ) 1 = strongly disagree, 5 = strongly agree · My agency is a very dynamic and entrepreneurial place. · People are willing to stick their necks out and take risks.

Hierarchical Culture (adapted from Zammuto and Krakower 1991 ) 1 = strongly disagree, 5 = strongly agree · My agency is a very formalized and structured place. · Bureaucratic procedures generally govern what people do.

Reform Orientation Summative index (Cronbach’s alpha = .77) of responses to the following question, taken from Deil Wright’s American State Administrators Project (Brudney, Hebert, and Wright 1999): From time to time, state agencies undertake to change the way in which they do things. Please indicate the extent to which your agency as implemented each of the following over the last four years (1 = no changes considered, 5 = changes fully implemented):

· Training programs to improve client or customer service · Systems for measuring customer satisfaction · Benchmarks for measuring program outcomes · Strategic planning to produce clear mission statements · Quality improvement programs to empower employees · Simplifi cation of human resource rules · Greater discretion to carry over funds · Reduction in hierarchical levels · Decentralization of decision making · Greater discretion in procurement

Red Tape (from NASP-I; see Bozeman 2000; Pandey and Scott 2002 ) If red tape is defi ned as burdensome administrative rules and procedures that have negative effects on the organization’s performance, please assess the level of red tape in your organization. Please enter a number between 0 and 10, with 0 signifying no red tape and 10 signifying the highest level of red tape.

Hierarchical Authority (from NASP-I; see Bozeman 2000; Pandey and Scott 2002) Please assess the extent of hierarchical authority in your organization. Please enter a number between 0 and 10, with 0 signifying few layers of authority and 10 signifying the many layers of authority.

Gender Are you male (0) or female (1)?

Education Level · Some college (1) · Bachelor’s degree (2) · Graduate degree (3)

Age In what year were you born? (Converted to age in years)

Length of Organizational Membership How many years have you worked for this organization?

Income Which of the following categories best describes your income from the agency in the previous year? · Less than $50,000 (1) · Between $50,000 and $75,000 (2) · Between $75,000 and $100,000 (3) · Between $100,000 and $150,000 (4) · Over $150,000 (5)

Professional Identifi cation (based on Aiken and Hage 1968 ) Are you a member of a professional society? (e.g., ASPA, APHSA, APHA, AMA, ANA) · Yes (1) · No (0) Note: NASP-I refers to Phase I of the National Administrative Studies Project.