Abstract
Although considerable research has developed exploring effective management of nonprofit boards of directors, there is limited understanding of the motivations of nonprofit board members to serve on boards. Using a sample of nearly 700 nonprofit board members, this study examines antecedent conditions and dimensions of public service motivation (PSM) as they apply to nonprofit board members and the differences in levels of PSM between board members who have worked primarily in the nonprofit, public, or private sectors. Board members with primary employment in the public sector show the highest levels of PSM. This study illustrates that nonprofit board members who work in the private sector exhibit fewer values associated with public service motivation. Other variables that predict public service motivation among board members include gender, level of education, and formal volunteering activity, among several others.
Introduction
The financial crisis of 2008–2009 presented an enormous challenge for the nonprofit sector. As budgets shrank and the fundraising environment became increasingly difficult, demand for services provided by nonprofit organizations rose (Lin & Wang, 2016). As a result, nonprofit organizations have been asked to provide increased services with fewer resources. Given this difficult operating environment, strong board leadership and commitment have become more important. Although considerable research has explored effective management of nonprofit boards of directors, there is limited understanding of the motivations of nonprofit board members to serve on the board, and very little scholarship examines the impacts of primary employment as it relates to nonprofit board participation. This study addresses these issues by considering the relationship between primary sector of employment and board member motivation by asking the question: How does primary sector of employment over a lifetime relate to board service? To more deeply examine this relationship, the public service motivation (PSM) framework is applied to ask: How does primary sector of employment relate to dimensions of PSM among board members, including commitment to public interest, compassion, and self-sacrifice? Are there differences in the levels of PSM among board members who have been primarily employed in the nonprofit, public, or private sectors?
To provide structure and validity to our analysis, the PSM framework is employed to better understand why individuals are attracted to serve on nonprofit boards. The motivations of volunteer nonprofit board members have been largely understudied and likely differ from other types of volunteers and paid employees. Over the past two decades, PSM literature has developed into the foremost framework for explaining motivation to serve in the public sector, and more recently, it has been applied to nonprofit settings. In the nonprofit context, PSM has been used to explain volunteer motivation and it has demonstrated success in explaining the values that differentiate nonprofit employees from their public and private sector counterparts (Clerkin et al., 2009; Coursey et al., 2011; Lee, 2012; Miller-Stevens et al., 2014).
As a stream of research that can be found in the social and behavioral sciences, PSM helps explain the relationships between motivations and incentive structures where PSM is considered altruistic, prosocial behavior, or “the desire to expend effort to benefit other people” through voluntary acts without return (Asseburg & Homberg, 2020; Grant, 2008, p. 49; Perry et al., 2010; Piatak & Holt, 2020). PSM explores how exposure to specific socialization processes, institutions, events, and ideologies may shape individuals’ “general, altruistic motivation to serve the interests of a community of people, a state, a nation or humankind” (Rainey & Steinbauer, 1999, p. 20). In this context, nonprofit board members have demonstrated a propensity toward prosocial behavior, as evidenced through their volunteer activities on boards and their engagement in policy activities within the governance context of the board. This volunteer work is often accomplished in tandem with their paid careers in the public, private, and nonprofit sectors. Thus, PSM is a useful framework for examining the motivations of individuals to serve on nonprofit board of directors, as well as an appropriate lens to compare the motivational differences of board members with primary employment in the nonprofit, public, or private sectors.
The contributions of this study are threefold. First, we aim to better understand why individuals are motivated to serve on nonprofit boards of directors. We believe this research will help guide nonprofits in their efforts to recruit and retain motivated and committed board members. Second, we aim to expand the nonprofit and theoretical literature by linking theory and practice within the context of PSM and nonprofit board participation. Third, we adapt a well-developed framework, PSM, to a new environment within the nonprofit sector.
The article is organized as follows: First, a brief overview is provided of PSM as a comparative tool to study motivation in the public, private, and nonprofit sectors. This is followed by a discussion of PSM and board service. Next, the PSM framework is described. The sample and method are then explained. Finally, the analyses and discussion of results are presented, which examine how PSM applies to nonprofit board members and whether differences in levels of PSM exist between nonprofit board members with primary employment in the nonprofit, public, or private sectors. The article concludes with implications for future research.
Review of Literature and Theoretical Model
Recently, volunteer and employee motives in the nonprofit sector have been ascendant on the nonprofit research agenda. Studies on nonprofit sector motivations have addressed incentives and drive to work in the nonprofit sector (Park & Kim, 2016), filling leadership roles in nonprofit organizations (Willems et al., 2015), and propensity to volunteer (Coursey et al., 2011; Coursey et al., 2008; Perry et al., 2008). These research veins have employed numerous theoretical lenses to explain motivation, including social resource theory (Dekker & Halman, 2003), economic utility maximization theory (Ziemek, 2006), PSM (Coursey et al., 2011; Miller-Stevens et al., 2014), and intrinsic/extrinsic reward theory (Park & Word, 2012). This increased attention to nonprofit sector motivation has been inspired by the flourishing research agenda relating to PSM to explain how and why individuals are attracted to serving others through altruistic and prosocial behaviors (Perry et al., 2010), as well as how PSM may affect levels of individual and organizational performance. This literature review examines two distinct veins of nonprofit research relating to PSM as a comparative tool and its applicability to the nonprofit board service. This is followed by a proposed model of the antecedents to PSM that link the dimensions of PSM, board service, and primary employment.
PSM as a Comparative Tool
Over the past several decades, a debate has emerged on the fundamental characteristics of management in the public and private sectors (Buchanan, 1974; Perry & Porter, 1982; Perry & Rainey, 1988; Rainey, 1982, 2003; Rainey & Chun, 2005; Rainey et al., 1976; Simon, 1995, 1998). The PSM literature was born out of this debate, and primarily argues that public and nonprofit sector employees hold a unique set of values that distinguish them from their private sector counterparts. As there is generally consensus among management scholars that employees in the public and nonprofit sectors hold unique values, the PSM literature has grown in an effort to help both scholars and practitioners better understand the motives and values of public, nonprofit, and private sector employees. A 2016 systematic review of the PSM literature identified 323 PSM articles (Ritz et al., 2016), with dozens more published since.
Among the leading and most promising research veins in public administration and management literature, PSM posits that workers attracted to public service likely hold a combination of rational and prosocial motivations to serve the public. Perry and Wise’s (1990) original definition of PSM referenced the construct as “an individual’s predisposition to respond to motives grounded primarily or uniquely in public institutions and organizations” (Perry & Wise, 1990, p. 368) and was later augmented to include nonprofit and other nongovernmental institutions and organizations (Perry et al., 2010; Wise, 2000).
In their original discussion of PSM, Perry and Wise (1990) suggest that interests such as commitment to public interest, compassion, self-sacrifice, and attraction to public policymaking are all associated with PSM. They distinguish between three types of motives that influence an individual’s desire to participate in public service, including rational, normative, and affective motives. Rational motives are based on individual gain, normative motives relate to beliefs of altruism, and affective motives are associated with an individual’s emotional processes and “genuine conviction” of the importance of a program or organization (Perry & Wise, 1990, p. 369). Other scholars have expanded on this definition to incorporate motivations of meaningful public, community, and social service that extend beyond the public sector (Brewer & Selden, 1998; Rainey & Steinbauer, 1999) and the beliefs and values that motivate individuals to act outside of their own self-interests (Perry & Hondeghem, 2008; Schott et al., 2019; Vandenabeele, 2007, 2011).
Several scholars (Behn, 1995; Perry et al., 2006; Perry & Hondeghem, 2008) have identified motivation as a “big question” in public management that requires additional scholarly attention to create a more effective public sector. This renewed focus on employee motivation is in response to the push toward creating more efficient and effective public sector organizations that occurred in the early-to-mid-1990s, the so-called “New Public Management” movement (Perry & Hondeghem, 2008). Similarly, the emergence of the nonprofit sector has contributed to a blurring of the lines between sectors (Allison, 1984; Dahl & Lindblom, 1953; Haque, 2001; Weisbrod, 1997). This blurring of the lines is evident in recent research applying PSM to management issues. For example, Mann (2006) explores the potential impacts of PSM on human resource management in the public, private, and nonprofit sectors. In this study, he notes that “suggestions for research on PSM in the nonprofit sector appear to have been unheeded thus far” (p. 40). He argues that the tenets of PSM are equally as applicable to the inherent characteristics of the nonprofit sector, and thus the framework should be applied and used to explore motivation in all three sectors.
As Bozeman and Su (2015) note in their critique of the PSM literature, the nonprofit sector has been “one seemingly compelling locus for PSM” application (p. 702). In the nonprofit literature, PSM has also been applied to explore employee intrinsic and extrinsic motivation, although there are few studies to draw from. In one recent study, Word and Carpenter (2013) apply an adapted version of the PSM scale developed by Perry (1996) to explore how the “nonprofit sector’s history and culture affect and uniquely shape the motivation of [nonprofit] workers” (p. 316). Their findings show that much like public sector employees, nonprofit sector employees are intrinsically motivated. There is also higher employee motivation when employees feel as though their values are in congruence with the organization’s mission, and workers that have multiple roles in the nonprofit sector tend to be more motivated in the workplace. In other studies, PSM has been found to have a bi-directional relationship with employment in the nonprofit sector. Individuals with high levels of PSM may be attracted to work in both the public and nonprofit sectors (Ballart & Rico, 2018; Word & Park, 2015; Wright et al., 2017). Public and nonprofit sector managers hold similar intrinsic and extrinsic reward values (Park & Word, 2012).
In their systematic review of PSM research, Ritz et al. (2016) found that less than 3% of PSM research focused exclusively on the nonprofit sector, while another 2% of studies compared the public and nonprofit sectors, compared with nearly 14% of studies conducting public–private sector comparisons. In total, only 10% of studies compared PSM across sectors (public/private/nonprofit). This article builds on the aforementioned theoretical and practical research on motivation by exploring the dimensions of PSM among nonprofit board members and the comparison between board members with different career histories.
PSM and Board Service
Nonprofit leaders serving as board members often come from professional backgrounds (BoardSource, 2014) and are recruited for board service for their skills, expertise, fundraising ability, and social networks (Inglis, 1994; Miller-Stevens & Ward, 2019). Several scales and measurement instruments have been developed to examine motivations for board membership, and many of these scales incorporate the rational, normative, and affective elements of PSM (Clary et al., 1998; Inglis, 1994; Searle, 1989). In one study exploring individual board member motivations in nonprofit organizations, Inglis and Cleave (2006) identify six components of board member motivation, including enhancement of self-worth, learning through community, developing individual relationships, unique contributions to the board, helping the community, and self-healing purposes. In their summary, the authors determine that five of the components are associated with “needs of the individual volunteer,” while the sixth, helping the community, reflects “more altruistic values” (p. 97). This desire to help others indicates these individuals may be motivated by a public service ethos or other prosocial motivation.
In another study, Preston and Brown (2004) explore the relationship between board member commitment and individual performance using a three-component model of commitment that emphasizes affective and normative commitment as well as continuance. They conclude that board members who are “emotionally attached” to their organization and who have “strong affective commitment” are regarded as highly valuable because they tend to be actively involved with the board (p. 233). They find that only affective commitment predicts perceived participation, while normative commitment and continuance do not contribute to the model.
In a large study of nonprofit board members, Miller-Stevens and Ward (2019) find that affective motives are prominent during board recruitment processes, and rational and normative motives grow in importance during a board member’s tenure. Additional subsequent research reveals that attraction and attrition motives are dynamic and change once an individual joins a nonprofit board (Miller-Stevens & Ward, 2019). Individuals reported being attracted to board positions for personal gain or rational motives and community-level reasons, but these reasons become less important after serving on a board.
While a few studies apply elements of PSM to nonprofit board members, the research is limited. This study extends the application of the PSM framework (Perry, 1996, 1997) to explore PSM in a sample of nonprofit board members in aggregate. Specifically, this lens is applied to examine the differences of PSM between groups of nonprofit board members who have worked primarily in the nonprofit, public, or private sectors. This research applies PSM to a new context within the volunteer motivation literature—that of nonprofit boards of directors—while expanding the theoretical literature of the PSM framework.
An Adapted Model: Antecedents of PSM and Board Service
Within the study of PSM, a subset of research has examined the antecedent conditions and determinants of PSM, or the factors that predispose individuals to develop values and traits consistent with prosocial behavior (Ballart & Rico, 2018). In their recent systematic review of the PSM literature, Ritz et al. (2016) find that roughly 17% of all PSM research, or 88 studies, have examined the potential antecedents.
Gender, age, and education are the most frequently examined antecedents of PSM (Ritz et al., 2016). Job grade, organizational tenure, sector of employment, income, job attributes, religiousness, and parental socialization constitute the second tier of commonly explored determinants of PSM. On the aggregate, Ritz et al. (2016) find that women tend to exhibit higher levels of public service motivation, and the variables age, education, job grade/management level, job tenure, place of work in the public sector, and good employee-leader relations tend to increase levels of public service motivation, although the results are not always consistent across studies. (p. 419)
Our research draws on an adapted model of antecedents of PSM. In studying a group of award-winning volunteers, Perry et al. (2008) develop a framework for examining the antecedents of PSM. Perry and his colleagues posit that gender, level of education and income will predict volunteering behaviors. They also posit that family socialization, religious socialization, and youth volunteering will predict volunteer activity, but these dimensions may also mediate and moderate the relationship between volunteering and PSM. We build on Perry et al.’s (2008) model by adding a dimension relating to primary sector of employment throughout an individual’s career. We also omitted one dimension of the original PSM scale, attraction to public policymaking, which is discussed further in the “Method and Data” section. The adapted model is presented in Figure 1.
Adapted model of the antecedents of public service motivation.
Method and Data
Survey Method and Location
The survey instrument included questions from the PSM scale (Perry, 1996), the antecedent conditions of PSM scale (Perry, 1997), reasons for joining and staying on a nonprofit board, current and lifetime employment experiences, and other organizational and individual attribute variables. Survey data were collected electronically using a controlled chain-referral method of survey administration in partnership with the Georgia Center for Nonprofits. The Georgia Center for Nonprofits is a nonprofit association that serves a membership of nonprofit organizations and professionals by providing research and policy information, networking events, job opportunities, and professional development activities to help organizations and individuals grow and built capacity in the nonprofit sector (Georgia Center for Nonprofits, 2020).
The chain-referral method of survey administration relies on point-of-contact participants to serve as a gatekeeper for survey distribution. This method is common when surveying hard-to-reach populations (Penrod et al., 2003). Staff at the Georgia Center for Nonprofits sent a prewritten recruitment email and survey link to executive directors, or other primary points of contact, of all of their member organizations, approximately 3,000 organizations at the time of this study. These individuals were asked to forward the email containing the survey link to members of their nonprofit boards of directors. This chain-referral method provided access to a population that otherwise would not have been easily accessible.
Survey administration was conducted between January 11 and February 11, 2013. At the end of this 1-month period, 1,046 surveys were returned of which 679 were usable. The remaining 367 surveys were not included in the analysis because large portions of the survey were incomplete.
In administering the survey, we deployed Perry’s (1996) original full 24-item PSM scale to allow maximum flexibility in the analysis of the results, as there have been numerous attempts to adapt, shorten, and reduce the number of items used to measure PSM (Kim, 2009; Ritz et al., 2016). Ultimately, we elected to use Coursey et al.’s (2008) rendition of the PSM scale, primarily because they refined and tested the model using a sample of exemplary volunteers, a group that is likely to share traits and values with our sample of nonprofit board volunteers. In refining the PSM scale Coursey et al. (2008) retained three of the original four factors including commitment to public interest, compassion, and self-sacrifice, and dropped the attraction to public policy dimension. Furthermore, after conducting psychometric verification and performing validity tests on the instrument, Coursey et al. (2008) only included 12 of the original 24 survey items in their final analysis, including three items on the commitment to public interest dimension, five items on the compassion dimension, and four on the self-sacrifice dimension. They also found solid reliability and good general fit for all three dimensions in the adapted model among their sampling frame. We also drew inspiration for Perry et al.’s (2008) adapted model of the antecedents of PSM, which was derived from the same sample of exemplary volunteers.
Sample
The study gathered data on the age, race, gender, educational attainment, yearly household income, and sector of employment of the board members who completed the survey. As respondents were not asked to provide the name of the organization for which they serve, the information presented below is in aggregate across the entire database of 679 surveys, though some tables report lower samples due to instances of cases with missing data.
Descriptive statistics show that 83% of board members report they are White, non-Hispanic, 11% African American, and 3% Latino. With educational attainment, 87% have earned a college degree. Sixty-four percent have an annual household income over US$100,000, and 27% have an income within the range of US$50,000 to US$99,999. Only 9% of the sample have an income of less than US$50,000 per year. Relating to gender, 51% of respondents identified as female and 49% identified as male. Descriptive statistics also show that 52% of board members in the sample have primarily worked in the private sector, while 30.5% have primarily worked in the public sector throughout their careers. Approximately 18% of nonprofit board members in the sample have mostly worked in the nonprofit sector.
Measures
Common analyses of reliability and dimension reduction were performed to verify the psychometric fit of the PSM latent variables among this population. Confirmatory factor analyses were employed to determine fit of the PSM model among a population of nonprofit board members. Next, group differences in goodness-of-fit measures for public, private, and nonprofit sector employees were examined. Data were analyzed using the STATA 12 software package that includes structural equation modeling capabilities. In addition, individuals were grouped by their sector of primary employment (i.e., public, private, nonprofit) and group differences of levels of PSM model were examined.
Results
As noted, the purpose of this study is threefold: to expand our understanding of the motives of individuals to serve on nonprofit boards to guide practitioners in their efforts to recruit and retain quality board members; to expand the nonprofit and theoretical literature by linking theory and practice within the context of PSM and nonprofit board service; and to adapt a well-developed framework, PSM, to a new environment within the nonprofit sector. The results below are organized by research question.
How Does Primary Sector of Employment Over a Lifetime Relate to Board Service?
Survey respondents were asked questions about their current and past employment, including the sector in which they have been primarily employed over their career. Furthermore, respondents were asked how many hours per month they spent on board duties. Table 1 reports the frequencies and crosstabs for these items. Over half of all surveyed members have been primarily employed in the private sector (52%), while roughly 31% hail from the public sector and only 17% have primarily worked at nonprofit organizations. Private sector board members spent approximately 15% fewer hours per month on board duties (8.6 hr) when compared with their public (10.1 hr) and nonprofit (10.33 hr) sector counterparts.
Frequencies of Board Membership by Primary Sector of Employment and Mean Values of Hours Per Month Spent on Board Responsibilities (N = 658).
Analysis of board member roles and responsibilities by sector of employment, reported in Table 2, reveal that private sector employees were overrepresented in board chair positions (60.9% of board chairs are from the private sector, while 52.1% of all board members are from the private sector). Both public and nonprofit sector employees were underrepresented in the board chair role. Nonprofit sector employees were marginally overrepresented in other board officer positions (19.5% vs. 17.3% of overall board positions). Public sector employees were slightly overrepresented in regular board positions.
Mean Comparison of Board Membership Roles and Responsibilities and Primary Sector of Employment (N = 658).
Comparison of mean values of board membership tenure, terms of service, and primary sector of employment were also examined (see Table 3). Overall, 73% of board members have served on other boards. However, board members who have worked primarily in the nonprofit sector have served on an average of 4.77 other boards, while public and private sector workers both report serving, on average, on fewer than four boards (3.82 and 3.91, respectively). Supplemental analysis demonstrates that there is not a significant difference in the size of the organizations, measured by budget, that nonprofit, public, and private sector employees serve on. This finding potentially raises the question of whether there may be a small, but potentially dense network of nonprofit workers who serve on many boards. This finding warrants further investigation, as there is only limited information relating to the mission, scope, or structure of the organizations that these board members are representing.
Mean-Comparison of Board Membership Tenure, Terms of Service, and Primary Sector of Employment (N = 658).
Scale used: 1 = <1 year; 2 = 1 to 3 years; 3 = 4 to 6 years; 4 = 7 to 10 years; 5 = 10+ years. bScale used: 1 = >US$250,000; 2 = US$250,000 to US$499,999; 3 = US$500,000 to US$999,999; 4 = US$1M to US$4.99M; 5 = US$5M to US$9.99M; 6 = US$10M to US$24.99M; 7 = US$25M+.
How Does Primary Sector of Employment Relate to the Dimensions of PSM Among Board Members, Including Commitment to Public Interest, Compassion, and Self-Sacrifice?
Using Perry’s (1996, 1997) PSM scales, survey participants were asked to think about their life experiences that may have influenced their desires to participate in public service and serve on a board of directors. As previously mentioned, Coursey et al.’s (2008) measures of PSM were ultimately employed and reliability results for the PSM construct are presented. An analysis of variance (ANOVA) table examining differences between the variables is also presented and discussed, along with regression tables examining the antecedent variables of PSM on the various dimensions of PSM are examined.
Exploratory factor analyses, which are not reported in tables, revealed that several of the specific items used to measure the latent dimensions of PSM were highly correlated and suggested good reliability, particularly for self-sacrifice and commitment to public interest. However, several items from Coursey et al.’s (2008) measure of compassion, when rotated using varimax rotation, cross-loaded onto other factors or demonstrated weak loadings. This factor was further examined using confirmatory factor analytic techniques. Confirmatory factor analyses are the most reliable and robust measures of reliability and psychometric verification for theoretically driven scale testing (Acock, 2013; Byrne, 2010; Kline, 2010). Figure 2 reports the results from a first-order confirmatory factor analysis among the entire sample of nonprofit board members. The compassion dimension has been shown to present measurement problems (Wright, 2008), which were also present in our analysis—three of the five items did not demonstrate a strong relationship. Modification indices were performed to address the compassion dimension, resulting in three items being dropped from the analysis, for a total of nine questions on three factors. The final survey items and corresponding dimensions are presented in Table 4.
First-order confirmatory factor analysis testing the public service motivation construct among nonprofit board members (n = 676).
Items and Dimensions of PSM.
Source. Adapted from Coursey et al. (2008).
Note. Respondents were prompted to “Please select the response that most closely represents your opinion:” and were presented with the following 5-item scale: 1 = strongly disagree; 2 = disagree; 3 = neutral; 4 = agree; 5 = strongly agree. PSM = public service motivation.
Table 5 reports common goodness-of-fit statistics testing the PSM construct among nonprofit board members in the sample. Measures of χ2 (adjusted for the degrees of freedom in the model) reveal strong model fit. Root mean standard estimate of errors (RMSEA), a parsimony correction class of goodness-of-fit indicator is included. Acceptable RMSEA scores are typically between .0 and .1 (Brown, 2006; Byrne, 2010; Hu & Bentler, 1999; Kline, 2010). RMSEA values between .06 and .1 are considered good fit, while values below .06 are considered very good fit (Hu & Bentler, 1999)—the current model fit is approaching very good at 0.069. The comparative fit index (CFI) is also reported—scores range between 0 and 1, with “a cutoff value close to 0.95” commonly used to indicate good model fit (Hu & Bentler, 1999, p. 1). The current CFI score of 0.936 in this model reveals adequate model fit. The Tucker–Lewis index (TLI) behaves similar to CFI (ranging from 0.0 to 1.0, with scores closer to 1.0 demonstrating stronger fit) and is reported in Table 5.
Goodness-of-Fit Indicators Testing the Public Service Motivation Construct Among Nonprofit Board Members (n = 679).
Note. .RMSEA = root mean square error of approximation; CFI = comparative fit index; TLI = Tucker–Lewis index.
p < .001.
Group-level of goodness fit indicators are reported in Table 6. Group-level goodness-of-fit statistics are limited in many structural equation modeling software packages, including STATA, M+, and AMOS. Typical group-level goodness-of-fit measures include χ2 and the coefficients of determination (CD). A CD value near 1.0 indicates good model fit (Stata, 2013). All groups demonstrate strong model fit with CDs approaching 1.0.
Group-Level Goodness-of-Fit Indicators for First-Order CFA by Primary Sector of Employment (N = 679).
Note. CFA = confirmatory factor analysis; CD = coefficient of determination.
p < .001.
Furthermore, an ANOVA was run to analyze the differences in the dimensions of PSM by primary sector of employment, reported in Table 7. Significant differences were identified between each primary sector of employment on the various dimensions of PSM. Notably, the private sector board members reported the lowest levels across all of the dimensions of PSM, while the public sector employees reported the highest levels of PSM on all three of the dimensions, including commitment to public interest, compassion, and self-sacrifice. Although, the nonprofit sector employees came in a close second on all of these dimensions.
ANOVA Tables for Dimensions of Public Service Motivation (N = 679).
Note. ANOVA = analysis of variance.
Three regression analyses were conducted, with the three dimensions of PSM as the dependent variables. In each model, measures of the antecedents of PSM were included as independent variables, as well as two dummy variables for primary sector of employment (nonprofit and private sectors). A dummy variable for public sector employees was excluded to avoid issues of multi-collinearity in the models. All three models were significant, explaining between 8.6% and 14.3% of the variance in each model. In all three of the models, being employed in the nonprofit sector did not predict the dimensions of PSM.
Regression Analysis Predicting Three Dimensions of Public Service Motivation (N = 520).
p < .05. **p < .01. ***p < .001.
Being employed in the private sector had a significant, negative relationship with all three of the dimensions of PSM among board members. It should also be noted that related models were also run including dummy variables for employment primarily in the public and private sectors, thus excluding nonprofit sector employment, and performed similarly. Two indexes were created to measure family socialization and religious activity, using items from Perry et al. (2008). Family socialization significantly predicted PSM dimensions of commitment to public interest and compassion, while religious activity was negatively associated with self-sacrifice among board members, but was not significant on the other two dimensions.
Another index was created to assess how many hours board members engaged in formal volunteering and was included in the models. This item was meant to capture volunteering engagements outside of traditional board duties. In all three models, formal volunteering significantly and positively contributed to predicting the dependent variables, dimensions of PSM. Similarly, informal volunteering had a positive association with commitment to public interest and self-sacrifice. Gender only significantly predicted one of the three dimensions of PSM—males were more likely to self-report self-sacrifice behavior. Finally, age, income, and highest level of education attained were only significant on the model predicting compassion, with age and education demonstrating a negative relationship.
In What Ways are There Differences in the Levels of PSM in Board Members With Primary Employment in the Nonprofit, Public, and Private Sectors?
The PSM construct appears to hold among nonprofit board members. Goodness-of-fit indicators of confirmatory factor analysis suggest that PSM among this population is good, although falls short of very good fit. This finding supports previous work that has demonstrated that public service motives are likely driven by an “other-regarding” disposition (Chen & Lee, 2015; Perry & Hondeghem, 2008). Thus, measures of PSM provide explanatory power among both public and nonprofit populations (Buchanan, 1974; Gabris & Simo, 1995; Perry & Porter, 1982; Perry & Wise, 1990; Rainey, 1982). While others have studied the impact of sector on levels of PSM, particularly the differences between the public and nonprofit sector (Clerkin et al., 2009), few studies have broadly examined differences between the public, nonprofit, and private sectors.
This study reveals that nonprofit board members’ motivations appear to be largely consistent with the PSM construct. However, descriptive and inferential statistics revealed significant differences between primary sectors of employment. While public sector employees hold higher levels of PSM, private sector employees are overrepresented both on boards, as well in the board chair position.
Limitations
The study has some limitations that should be considered when interpreting the results. The data were derived from a cross-sectional self-report survey, which could result in degrees of social desirability bias or common method bias. In addition, response rates for the survey are not available because the Georgia Center for Nonprofits did not have information on how many of the organizational points of contact forwarded the survey or how many board of directors members opened the email. This supports the general understanding that response rates are difficult to track with chain-referral survey methods (Penrod et al., 2003). Alternatively, access to more robust state-level data on board membership would help address issues of representativeness of our sample. The study also would have benefited to collect more board-level data for the ability to control for board-level attributes. However, this would have been difficult to do without compromising anonymity.
Finally, there are known shortcomings with PSM and its measurement. Many PSM studies only test the relationship of PSM with desirable constructs while ignoring those that are undesirable, thus limiting our full understanding of PSM (Ritz et al., 2016). In addition, inconsistencies are present between studies using indicators of PSM with differing operational definitions of the same construct (Wright, 2008). Furthermore, considerable efforts have been made to shorten or modify the original 24-item scale, and few studies have used the original four-factor measure of PSM because internal reliability issues have been found. Such diversity in operational definitions and approach limits replicability of PSM studies and efforts to build on previous research in this area (Wright, 2008).
Discussion and Future Research
The PSM scale is a useful tool for examining the differences in motivations among employees from different sectors. Yet, applications of PSM within the nonprofit sector are scarce (Ritz et al., 2016), despite being a clear locus for PSM application (Bozeman & Su, 2015). Although previous research has found similarities in levels of PSM among public and nonprofit employees (Park & Word, 2012; Word & Park, 2015; Wright et al., 2017), limited studies have examined PSM among nonprofit board members. In this study examining PSM among nonprofit board members, differences in levels of PSM among board members hailing from different sectors of employment were found. Being employed primarily in the private sector over one’s lifetime, along with other antecedent conditions of PSM, was negatively associated with levels of PSM, even among a cadre of citizens who have demonstrated prosocial behaviors through volunteering.
Studying board member motivation has important implications, both for practitioners and researchers attempting to more fully understand the relationships between attributes, attitudes, and behaviors of board members. From a practitioner perspective, a clearer understanding of motivations of board members could be used as a guide in their efforts to recruit and retain board members. As the results in this study indicate, nonprofit board members appear to be motivated by a public service ethos, which is consistent with previous research in this area (Inglis & Cleave, 2006), but the level of PSM differs between board members when considering primary employment. Understanding the relationship between employment background and PSM could produce new strategies to define and target a specific type of board member during recruitment periods. This knowledge could also have an impact on retention of board members through improved board recruitment and onboarding practices, experiences, and expectations that take into consideration board members’ employment backgrounds and the types of work environments they are used to.
This research also extends upon Inglis and Cleave’s (2006) findings that board member motivations may be captured by measuring six latent variables, including enhancement of self-worth, learning through community, developing individual relationships, unique contributions to the board, self-healing purposes, and helping the community. Although Inglis and Cleave find that five of the six motivational factors relate primarily to rational or extrinsic factors, this research more closely examines the prosocial reasons identified in the PSM literature that align with the factor helping the community. As we previously found (Miller-Stevens & Ward, 2019), board members may be initially attracted to board service for some of the rational reasons identified by Inglis and Cleave, but board members also reported continuing their service on boards for affective and normative reason. Research by Preston and Brown (2004) has found that board members with strong affective commitment are more likely to be engaged members. Thus, a closer look at the affective and normative motivations of board members is warranted.
Furthermore, this study highlights the overrepresentation of private sector employees in board chair positions. Although individuals primarily employed in the nonprofit sector serve on boards at a far lower rate than their private and public sector counterparts, these nonprofit workers report serving on considerably more boards of directors. This suggests there may be a small, but potentially well-connected group of nonprofit employees serving on boards. Future research might examine the social networks of board members to better discern where individual board members are serving and identify cliques within nonprofit board service. In addition, taken with Word and Carpenter’s (2013) findings that public and nonprofit employee motivation is higher when there is alignment between personal values and those of the organization, nonprofit boards might consider adding more public and nonprofit sector employees to their board leadership positions.
Finally, our findings relating to the antecedent conditions suggest that primary sector of employment among board members has an effect on levels of PSM. We offer an adapted conceptual model of the antecedents of PSM to include primary sector of employment (Figure 1) and use regression analysis to measure these relationships. Taken with other research linking PSM to organizational performance (Alonso & Lewis, 2001; Bright, 2007; Paarlberg & Lavigna, 2010), boards might consider adding sector of employment to their board recruitment matrices, or they could be more open to the recruitment of public and nonprofit sector employees. However, board members primarily employed in the private sector may serve a boundary spanning function by increasing the reach and visibility of the nonprofits they serve. More research examining the motivations of boards during recruitment processes is necessary.
Although this article focuses on the application of PSM to the nonprofit context, the development of specific theories and frameworks of motivation within the nonprofit sector has been relatively lagging. Currently, nonprofit scholars have not developed a unified theory or framework of effective board governance. As the motivation component is important to performance and board effectiveness, this continued research effort could potentially provide contributions to the nonprofit governance literature and ultimately nonprofit practitioners wishing to improve board governance practices.
Future research in this area should also consider the context within which nonprofit organizations are operating and the internal pressures to run and be managed like private sector organizations. These pressures may impact the type of employees that are attracted to work in the nonprofit sector, which in turn may impact board member participation. Furthermore, person–environment fit theories should be used to explore educational attainment and degree type as they relate to employment choice, PSM, and board service.
Finally, moving beyond simplified explanations relating to altruism should be another goal of nonprofit theorists. As it stands, explanations of nonprofit employee and board motivations are too often oversimplified to motives associated with altruism. Although prosocial values are undoubtedly an important component in explaining board membership, the public sector motivation research suggests the explanation is likely more complex and should include such non-altruistic reasons as peer pressure, social norms, promise, and legal issues. Extrinsic and intrinsic motivations may offset each other in forming overall self-determination, and future motivation research should consider this more holistic view.
Conclusion
This study is of particular interest to PSM scholars because it carries on the original spirit of the PSM body of literature that was born out of the examinations of the similarities and differences between employees across the public and private sectors. Others have applied PSM to nonprofit settings (Gabris & Simo, 1995; Park & Word, 2012; Word & Carpenter, 2013) with mixed results. The population examined in this study represents employment across all sectors, with the additional filter of an attraction to serve on a nonprofit board of directors. These findings suggest that the PSM scale can be effectively applied to a new subset of the nonprofit sector, that of board members. As board members are both representatives of the nonprofit sector and are often volunteers, these findings support the application of PSM in both the nonprofit and volunteer literatures.
Footnotes
Declaration of Conflicting Interests
The author(s) declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.
Funding
The author(s) received no financial support for the research, authorship, and/or publication of this article.
