How Do Directors of Nonprofit Organizations Perceive the Donor Evaluation Process?

Prior studies indicate that donors, especially large donors, use financial indicators as the main criterion to evaluate efficiency of a nonprofit organization (Weisbrod and Dominguez 1986; Khumawala and Gordon 1997; Okten and Weisbrod 2000; Parsons 2007). For example, donors tend to contribute to an organization reporting a higher program ratio, a commonly used financial metric computed as the organization's program spending divided by total spending (Weisbrod and Dominguez 1986; Khumawala and Gordon 1997). However, following recent scandals involving expense misreporting to make nonprofit organizations financially desirable, advocacy groups (e.g., Imagine Canada, GuideStar, BBB Wise Giving Alliance, and Charity Navigator) have called for donors—and the donors themselves have expressed enthusiasm for the idea—to shift their evaluation focus from financial metrics only (i.e., financially focused evaluation process) to the balance of an organization's financial and nonfinancial metrics (i.e., balanced evaluation process) (Association of Fundraising Professionals 2008; Perry 2013).¹

However, a shift to include nonfinancial metrics in the donor evaluation process does not guarantee that nonprofit organizations obtain donations, especially when nonfinancial performance is not good. In other words, the two types of donor evaluation processes (financially focused versus balanced) not only differ in what information they consider, but can also change the likelihood that nonprofit organizations obtain a favorable donation outcome. Prior studies find that both the informational feature of an evaluation process and the favorability of an outcome resulting from the evaluation process, affect fairness perceptions of the process (Korsgaard and Roberson 1995; Lind, Kanfer, and Earley 1990).² The perceived fairness of an evaluation process has been demonstrated as impacting people's behaviors (e.g., Brockner and Wiesenfeld 1996, 2005; Libby 2001; Cohen, Webb, Sharp, and Pant 2007). For example, perceived procedural fairness mitigates employees' poor performance (Libby 1999, 2001) and their lack of commitment to their jobs (Burney, Henle, and Widener 2009; Lau and Moser 2008; Lau and Sholihin 2005). Perceived procedural fairness is also shown to reduce self-interested behavior such as division managers' price resistance in transfer price negotiations (Kachelmeier and Towry 2002), managers' intentions to commit earnings management (Cohen et al. 2007), taxpayers' noncompliance with tax regulations (Porcano 1984), and employees' decisions to steal from the organization (Greenberg 1990, 1993). Turillo, Folger, Lavelle, Umphress, and Gee (2002) even found that third-party observers, who have no self-interest in a situation, tend to punish the party treating others unfairly and to help the unfairly treated party at their own cost.

In a fundraising context, it is important to know whether the board of directors of a nonprofit organization perceives the donor evaluation process to be fair for several reasons. First, directors often act as fundraisers along with management (Gill 2005; O'Regan and Oster 2005). They care about the donation outcome and would like donors to evaluate their nonprofit organization fairly. Accordingly, their reaction to a negative donation outcome is likely to depend on their perceived fairness of the donor evaluation process. Second, and more important, directors have a responsibility to monitor management, and the perceived fairness of the donor evaluation process could influence their decision to implement due diligence that exposes management misconduct (M. Yetman and R. Yetman 2012; Burney et al. 2009; Lau and Moser 2008; Donovan 2007; Lau and Sholihin 2005; Glaeser 2003; Fama and Jensen 1983). Despite the importance of directors' fairness perceptions of the donor evaluation process, no studies, of which I am aware, examine it. Hence, I have designed two experiments to address the following research questions: (1) whether directors of nonprofit organizations perceive greater fairness under a balanced donor evaluation process than under a financially focused donor evaluation process; and (2) which factor, the informational feature of donor evaluation process or the favorability of donation outcome generated from the process, contributes to directors' procedural fairness judgments.

The first experiment captures a typical fundraising context in which nonprofit organizations are likely to debate whether potential donors should adopt a balanced or financially focused evaluation process. The experiment holds constant the relatively poor financial performance of nonprofit organizations. Given that financial performance is poor, nonprofit organizations likely hope donors adopt a balanced evaluation process, so that consideration of nonfinancial performance could provide a greater chance to obtain donations. Sixty-five nonprofit directors participated in the first experiment. They were asked to assume the role of a director on the board of a hypothetical family center, whose donors adopt either a financially focused or balanced evaluation process to decide on donations. The favorability of donation outcome is operationalized as the perceived likelihood that the family center can obtain donations, manipulating the center's nonfinancial performance to be good or poor while holding constant poor financial performance. Results from this experiment indicate that directors perceive the balanced evaluation process to be procedurally fairer than the financially focused evaluation process. Furthermore, the procedural fairness judgments are not impacted by the favorability of donation outcome, but are triggered only by the informational feature of the donor evaluation process.

To increase the external validity, a second experiment highlights a context in which a balanced donor evaluation process is likely to be less desirable. More specifically, the experimental context is one where the organization has good financial, but poor nonfinancial, performance such that adopting a balanced evaluation process potentially damages the opportunity to obtain donations, as compared to a financially focused evaluation process. I find that, although the balanced donor evaluation process makes the organization appear worse because of the relatively poor nonfinancial performance, directors still perceive the balanced donor evaluation process to be procedurally fairer than the financially focused evaluation process.

Results from this study have the following theoretical and practical implications. First, this study finds that the perceived favorability of evaluation outcomes does not influence perceived procedural fairness judgments of an evaluation process. This result contrasts with previous studies for for-profit organizations (e.g., Lind and Tyler 1988; Lind et al. 1990; Burney et al. 2009; Lau and Moser 2008; Lau and Sholihin 2005), which implies that the nonprofit setting is unique. In the nonprofit setting, directors respect and accept donors' donation decisions regardless of whether they donate. The directors' low expectation for donors' plausibly explains the findings' inconsistency with the previous literature. Hence, this study adds to a stream of studies indicating that social context is a factor that must be considered when examining procedural fairness judgments (i.e., van den Bos 2002; Schminke, Cropanzano, and Rupp 2002; Kray and Lind 2002).

Second, if directors of nonprofit organizations internalize fundraising as their first priority, then directors can think of themselves as third-party observers overseeing management and as evaluatees. Although the psychology literature separately investigates how third-party observers and evaluatees develop procedural fairness perceptions of an evaluation process (Skarlicki, Ellard, and Kelln 1998; Turillo et al. 2002), no study examines whether and how people who assume both roles develop procedural fairness perceptions. This study provides evidence regarding how directors assuming both roles form procedural fairness perceptions of the donor evaluation process.

Finally, both practitioners and academic researchers advocate a balanced evaluation process (Association of Fundraising Professionals 2008; Herman and Renz 2008; Silverman and Beatty 2006, 2007; Silvergleid 2003; Campbell 2002; Kaplan 2001). This study provides empirical evidence on how a balanced evaluation process affects the fairness perceptions of nonprofit directors. Such perceptions can render directors less likely to sacrifice their governance role to their fundraising role in the cases in which the balanced donor evaluation process generates a negative donation outcome (Burney et al. 2009; Lau and Moser 2008; Lau and Sholihin 2005; Gill 2005; Libby 1999, 2001; Brockner and Wiesenfeld 1996).

The second section further develops the institutional background for this study, followed with hypotheses. The third and fourth sections describe the first experiment and its results. The fifth section presents the second experiment and its results. The sixth section concludes.

Nonprofit organizations in developed countries (e.g., the United States, Canada, the United Kingdom, and Australia) are an important economic force, contributing a significant portion of GDP (e.g., 6.9 percent in the United States and 8.5 percent in Canada) and providing employment for a large segment of the workforce—as examples, nearly 8.6 million fulltime-equivalent workers in the United States and over 2 million in Canada (Hall, Bar, Easwaramoorthy, Sokolowski, and Salamon 2005; Sokolowski and Salamon 1999). Contributions from donors make up 17 percent of all revenues for nonprofit organizations in Canada (Statistics Canada 2005) and the higher figure of 44 percent in the United States (Yan, Denison, and Butler 2009). Given the significance of donations to nonprofit organizations, directors care about whether their organizations can obtain sufficient donations and what performance evaluation metrics that donors use to decide whether and how much to donate (Keating, Parsons, and Roberts 2008; Krishnan, M. Yetman, and R. Yetman 2006; Callen, Klein, and Tinkelman 2003).

This study considers two archetypes of donors' evaluation processes. One donor type focuses solely on financial metrics, especially the ratio of charitable program expenses to total expenses (i.e., the program ratio). This donor type wants to ensure that as much of the total funds donated as possible are spent on charitable programs, rather than on administrative or fundraising activities (Charity Intelligence Canada 2008; Roberts 2005; Voluntary Sector Initiative 2003). They view the program ratio as a “sufficient statistic” for evaluating the achievements of charitable programs and tend to contribute to organizations reporting a higher program ratio (Feltham and Xie 1994). The evaluation process adopted by these donors is labeled herein as the “financially focused evaluation process.”

The second donor type considers the program ratio (i.e., the indicator of financial performance) as well as nonfinancial indicators of the organization's “service efforts and achievements” (Silverman and Beatty 2007; Campbell 2002; Kaplan 2001). This donor type believes that, while the portion of donations devoted to charitable programs compared to the administrative and fundraising activities is important, it is also important to know the impact of their donations in advancing charitable causes. The evaluation process used by these donors is labeled as the “balanced evaluation process.”

Prior studies find that people form fairness judgments about an evaluation process through which a decision or an outcome is determined; this type of fairness judgment is defined as procedural fairness (Leventhal 1980; Skarlicki et al. 1998; Turillo et al. 2002). Leventhal (1980) identifies six rules that people use to judge the symbolic and informational characteristics of a procedurally fair evaluation process. People need not apply all six rules to assess the procedural fairness of an evaluation process in a particular setting, as normally the most salient rules or “norms” are applied (Ambrose and Arnaud 2005; Leventhal 1980). In the context of this study, two of the rules, namely “accuracy” and “representativeness,” are the most relevant to the assessments of the fairness of the donor evaluation process (Dipboye and de Pontbriand 1981; Landy, Barnes-Farrell, and Cleveland 1980; Landy, Barnes, and Murphy 1978).³ Leventhal (1980, 25, 30) defines these two rules as:

• • 

  Accuracy: the procedures should be based on as much valid information and informed opinion as possible, with a minimum of error.

• • 

  Representativeness: the procedures must reflect the basic concerns, values, and outlook of the individuals and subgroups impacted by the evaluation outcome.

The definitions of accuracy and representativeness are both related to the completeness of information used in the evaluation (i.e., how much information is being used in the donor evaluation process), but from a different perspective. Accuracy emphasizes whether the donor evaluation process includes all relevant information about organizational performance, while representativeness is more precisely about whether the donor evaluation process considers the information from all basic aspects that directors care about when evaluating organizational performance.

The financially focused donor evaluation process only considers one measure, the program ratio. Analytical and empirical studies (e.g., Feltham and Xie 1994; Banker, Potter, and Srinivasan 2000) demonstrate that a single financial measure (e.g., earnings) is not usually considered a “sufficient statistic” for the performance evaluation. A single financial measure, for example, net income, only reflects the organization's short-run performance, ignores the organization's long-term strategy, fails to capture the organization's multidimensional performance, and is often subject to manipulation (Kaplan and Norton 2001a, 2001b). Therefore, it is necessary to add nonfinancial measures to financial measures to overcome these shortcomings (Feltham and Xie 1994; Banker et al. 2000). In a nonprofit setting, directors often volunteer to serve on the board out of a desire to make a difference through a charitable cause (O'Regan and Oster 2005). They also consider the social responsibilities of an organization and take into account the nonfinancial performance of an organization, as much as the financial indicators. Hence, nonprofit directors are likely to be concerned about the ability of a single financial performance measure (i.e., the program ratio) to reflect the organization's performance accurately (i.e., ignores valid information) and in a representative manner (i.e., fails to capture the outcomes of the organization's services).

Nonfinancial metrics, such as service effort and achievements, complement the program ratio, demonstrating the outcomes of charitable spending and linking financial performance to the organization's future goals (Kaplan 2001). Adding appropriate service performance metrics to the program ratio can more accurately and representatively gauge the organization's actual performance. Therefore, from the standpoint of directors, a balanced set of financial and nonfinancial metrics can be considered procedurally fairer than a financial metric alone (Leventhal 1980). Hence, I predict:

• H1: 

  (Main effect of the donor evaluation process): Directors perceive the donor evaluation process to be more procedurally fair when donors use a balanced approach than when donors focus solely on the program ratio.

In addition to the informational characteristics of an evaluation process, the capacity of the process to provide a favorable evaluation outcome may also influence procedural fairness perceptions (Lind and Tyler 1988; Lind et al. 1990). If a procedure generates an unfavorable outcome, then the procedure may be judged as unfair, even if the evaluation process meets Leventhal's (1980) normative rules. In the context of this study, the favorability of the evaluation outcome (i.e., the likelihood of an organization obtaining donations) is affected by performance on measures important to the evaluation method—the financial measures for a financially focused evaluation process and both the financial and nonfinancial measures for a balanced evaluation process. Both processes consider financial performance. The difference between the two evaluation processes is that the balanced evaluation process considers nonfinancial performance, while the financially focused evaluation process does not. Hence, it is nonfinancial performance (i.e., service performance) that differentiates the likelihood of an organization obtaining donations.

When service performance is relatively poor, provision of service performance information potentially offsets a favorable evaluation for an organization with a high program ratio, or makes the evaluation worse for an organization with a relatively low program ratio. Hence, the balanced set of evaluation metrics, although more accurate and representative and therefore procedurally fairer, does not guarantee an increased likelihood of the organization obtaining donations. Thus, this balanced approach might not be considered fairer than the financially focused evaluation process.

Relatively good service performance, however, can mitigate the negative impression given by a poor program ratio, or strengthen a positive donor evaluation brought about by a high program ratio. Hence, if the outcome favorability contributes to procedural fairness judgments of the donor evaluation process, then the procedural fairness perceived for the balanced donor evaluation process should be even higher (lower) when the nonfinancial performance is good (poor). I formalize the interactive effect in a second hypothesis.

• H2: 

  (The moderating effect of nonfinancial performance): The extent to which directors perceive a balanced donor evaluation process to be more procedurally fair than a financially focused process is stronger when nonfinancial performance is good than when nonfinancial performance is poor.

In the first experiment, I hold constant the organization's financial performance at a relatively poor level. The objective is to simulate a fundraising context in which the nonprofit organization is likely sensitive to whether donors should adopt a balanced evaluation process that considers both financial and nonfinancial performance.

The experimental design is 2 × 2. The first factor is the donor evaluation process; either balanced or financially focused. The second factor is the nonprofit organization's nonfinancial performance, which either improves or declines. The donor evaluation process is manipulated by informing participants as to whether a potential large donor adopts a single financial evaluation metric (i.e., the program ratio) or a more balanced set of performance evaluation metrics including both the program ratio and nonfinancial indicators of “service efforts and achievements.” The organization's nonfinancial performance is manipulated by varying good or poor service performance relative to its peer organizations' service performance in the current year and its own service performance in the prior year.

I enlisted contact persons, who are members of or associated with nonprofit organizations, to send an invitation email to individuals who are current or former board members of nonprofit organizations. A total of 65 directors participated in this experiment.⁴ Table 1 presents their demographic information. Randomization appears to be successful as there are no significant demographic differences across the experimental cells (all p's ≥ 0.104).⁵

I asked participants' opinions about the fundraising and monitoring responsibilities of nonprofit directors. Their responses, on a nine-point scale with −4 being “strongly disagree” and +4 being “strongly agree,” indicate that the participating directors believe that monitoring management is their important responsibility (mean agreement score is 3.11). Participants agree that fundraising is also an important responsibility (mean agreement score is 2.72). Both means are significantly larger than the midpoint of the scale 0 (p-values < 0.001). In addition, I asked participants whether nonfinancial service performance is more important than financial performance to determine the organization's chance of obtaining donations. Among the 65 participants, 50.77 percent indicated that nonfinancial performance is more important, 12.31 percent were not certain, and 36.92 percent disagreed. The mean score is 0.46, which is marginally significantly larger than 0 (p = 0.091). Hence, in general, participants believe that donors should consider the nonfinancial performance of nonprofit organizations when deciding on donations.

A pretest was run by using 46 university business students. Pretest results indicated that the experimental instrument was understandable and appropriate. After validating the experimental instrument with the pretest, contact persons were asked to send an invitation to potential participants, with a follow-up email approximately one week later. To respond to the survey, participants clicked on a web link provided in the invitation email and were automatically taken to the website with the experimental instrument. The participants first read general information about the study, including the precautions taken to ensure anonymity (i.e., ethics consent procedure) and then clicked a consent button to participate in the study. They were then randomly assigned by the survey software to one of the four experimental conditions. The software took participants through the online case in which they, acting as a director in a nonprofit organization, were asked to assess the procedural fairness of the evaluation process by a potential large donor. The participants then answered post-experimental and demographic questions.

Participants were instructed to assume the role of a director on the board of a family center.⁶ The family center needs to develop alternative funding sources to replace government funding lost due to budget cuts. Participants were told the national average program ratio is 74 percent. The program ratio of the family center is 66 percent this year, and “donors who consider the program ratio in making their donation decisions would likely evaluate a program ratio below the national average as a negative indicator when deciding about a donation to the ABC Center.” As such, the financial performance of the family center is held at the poor (below average) level.

Because procedural fairness judgments are formed based on a comparison of the alternative evaluation processes (Kahneman, Knetsch, and Thaler 1986), the participating directors were reminded explicitly that the center's potential donors adopt either a financially focused evaluation process or a balanced evaluation process. Participants then judged the procedural fairness of the donor's evaluation process to decide about a donation to the family center.

Manipulated variables are the donor evaluation process (EP) and the organization's nonfinancial performance (NFP). In the condition under which donors adopt a financially focused evaluation process, participating directors are informed that the target donor in the experimental case “focuses solely on the program ratio when making donation decisions.” Conversely, participants in the balanced evaluation process condition are informed that the target donor “focuses on both the program ratio and any voluntarily disclosed level-of-service efforts and achievements when making donation decisions.”

When the organization's service performance is relatively good, the participants are informed: “Overall, the Center's programs experienced great success this year. The number of parents and children who received services from the Center was 8 percent higher than last year and 5 percent higher than the average number of clients serviced by similarly sized family centers.” When the organization's service performance is relatively poor, however, the participants are told that: “Overall, the Center's programs did not experience as much success this year as was hoped. The number of parents and children who received services from the Center was 8 percent lower than last year and 5 percent lower than the average number of clients serviced by similarly sized family centers.”

The dependent variable is the participants' perceived procedural fairness of the donor evaluation process (PFair). Procedural fairness is measured on a nine-point scale, with −4 indicating “very unfair” and +4 indicating “very fair” in response to the statement, “Do you consider this potential large donor's evaluation criteria for making a donation to be fair?”

To check whether participants attend to the manipulations, participating directors are asked to answer on a nine-point scale (with a −4 indicating “Strongly disagree” and +4 indicating “Strongly agree”) whether they agree with each of two statements. Directors assigned to the group that is evaluated by the balanced donors agree significantly less with the statement, “The potential large donor makes donation decisions based solely on the program ratio” than the directors assigned to the group evaluated by the financially focused donors (mean = −2.14 versus 1.51; t = 6.36; p < 0.001). The directors in the condition for which the Center's nonfinancial performance improves agree significantly more with the statement, “Overall, the Center's service performance is excellent this year” than those in the condition for which the service performance declines (mean = 1.21 versus −1.09; t = 5.76; p < 0.001). Therefore, the manipulations of the donor evaluation process and the organization's nonfinancial performance are successful.

Table 2, Panel A reports descriptive statistics for the participants' procedural fairness assessment of the donor evaluation process (PFair). H1 predicts a main effect of the donor evaluation process (EP) on the directors' procedural fairness judgments (PFair), while H2 examines how nonfinancial performance moderates the effect of the donor evaluation process on procedural fairness perceptions (NFP ∗ EP). ANOVA results presented in Panel B of Table 2 indicate a significant main effect of the donors' evaluation process (EP) on the perceived procedural fairness (F_(1,61) = 10.31; p = 0.002). The planned contrast in Panel C of Table 2 further indicates that the balanced donor evaluation process is perceived to be significantly fairer than the financially focused evaluation process (PFair mean = 1.95 versus 0.33; p = 0.002). Hence, H1 is supported.

The interaction of the donor evaluation process and the organization's nonfinancial performance (NFP ∗ EP) in Table 2, Panel B, however, is not statistically significant (F_(1,61) = 2.01; p = 0.162). As ANOVA is not a powerful approach to detect the ordinal interaction predicted by H2, I further use pairwise comparisons specifically designed to test an ordinal interaction (Strube and Bobko 1989; Bobko 1986). I do not find a significant interaction between NFP and EP for this approach. Hence, H2 is not supported. That is, regardless of whether nonfinancial performance is good or poor, the balanced donor evaluation process is perceived to be fairer than the financially focused evaluation process.

The moderating effect of nonfinancial performance predicted in H2 is caused by the expected donation outcome; more specifically, the director-perceived likelihood that donors will donate. Hence, I directly measure the perceived-donation likelihood (i.e., LKDDonate) between the conditions with poor versus good nonfinancial performance. I asked participants to judge the likelihood of the donor making donations to the organization (LKDDonate) on a nine-point scale with 1 indicating “very unlikely” and 9 indicating “very likely” in response to the statement, “How likely is it that the Center will obtain a substantial donation from this potential large donor given the Center's performance?” As expected, the perceived-donation likelihood in the group with good nonfinancial performance is significantly higher than that in the group with poor nonfinancial performance (mean = 4.93 versus 3.69; F_(1,63) = 6.15; p = 0.016). Then I use the perceived-donation likelihood (LKDDonate) to replace nonfinancial performance (NFP) as the independent variable and again execute ANCOVA to test H1 and H2. The results from the ANCOVA are similar to the ANOVA results reported in the previous section. The main effect of EP on the perceived procedural fairness is significant (F_(1,61) = 7.26; p = 0.009), and the interaction between EP and LKDDonate is not significant (F_(1,61) = 2.39; p = 0.128).

The first experiment is executed in a context in which financial performance of nonprofit organizations is relatively poor. In this context, directors want donors to adopt a balanced evaluation process so that consideration of their nonfinancial performance could provide a greater chance to obtain donations. To increase the external validity, I conducted a second experiment in an alternative context where a balanced donor evaluation process is likely to be less desirable than a financially focused process. More specifically, the second experiment is the case where financial performance of nonprofit organizations is good but nonfinancial performance is poor. Hence, a balanced process that considers poor nonfinancial performance makes the organization appear worse.

The second experiment has a 2 (donor evaluation process: balanced versus financially focused) × 1 design. I manipulate the donor evaluation process in the same way as the first experiment. Experimental procedures and the experimental case are identical to the first experiment, except that financial performance is held constant at a good level and nonfinancial performance is held constant at a poor level. In total, 28 directors participated in the second experiment.⁷ Table 3 indicates that directors maintain their belief that a balanced donor evaluation process is more procedurally fair, even when the nonfinancial performance information makes the organization less likely to obtain donations (the means of PFair for balanced versus financially focused = 2.538 versus 0.333; F_(1,26) = 7.55; p = 0.011). Hence, the results from the second experiment further validate the findings obtained from the first experiment. Specifically, the informational feature of the donor evaluation process, not the favorability of donation outcome resulting from the process, appears to drive procedural fairness judgments of the donor evaluation process.

Directors of nonprofit organizations have dual roles in a fundraising context. They monitor the organization's management but also help with fundraising. Struggling with these conflicting roles, directors' procedural fairness perceptions of the donor evaluation process could be a crucial factor in determining how directors balance their activities. This study examines whether the two types of evaluation processes adopted by donors (financially focused versus balanced) affect directors' procedural fairness perceptions.

The results of this study indicate that directors consider a balanced donor evaluation process procedurally fairer than a financially focused evaluation process. Furthermore, only the informational feature of the donor evaluation process contributes to procedural fairness perceptions, while the perceived favorability of donation outcome does not contribute to the procedural fairness perceptions of directors. The latter finding contrasts with results in prior studies in the for-profit setting. In for-profit organizations, people focus on economic equivalent exchanges and expect evaluators (e.g., employers or supervisors) to reciprocate their work with a favorable evaluation outcome. Since evaluatees care about the favorability of outcomes, favorable evaluation outcomes can increase their perceived fairness of an evaluation process (e.g., Lind and Tyler 1988; Lind et al. 1990; Burney et al. 2009; Lau and Moser 2008; Lau and Sholihin 2005). However, unlike the evaluatees in for-profit organizations, nonprofit directors do not necessarily have a should-give expectation for donors, because most donors contribute out of altruism not from rational economic behavior. Hence, directors always perceive the balanced process to be more procedurally fair than a financially focused evaluation process, regardless of the donation outcome.

This study adds to a stream of studies to demonstrate that social context is an important factor that should be considered when examining perceptions of procedural fairness (e.g., van den Bos 2002; Schminke et al. 2002; Kray and Lind 2002). As discussed above, the nonprofit context likely plays an important role in explaining why directors form procedural fairness judgments in a way that differs from the judgments of individuals in the for-profit setting.

This study also has practical implications. A balanced evaluation process is widely promoted and is gradually being adopted by donors (Association of Fundraising Professionals 2008; Herman and Renz 2008; Silverman and Beatty 2006, 2007; Silvergleid 2003; Campbell 2002; Kaplan 2001). This study provides empirical evidence to examine the effects of a balanced evaluation process on nonprofit organizations. Specifically, the perceived fairness of a balanced donor evaluation process could motivate directors to focus more on their governance role and less on their fundraising role. Accordingly, a balanced evaluation process potentially enables donors and other interested parties to trust director monitoring.

This study has the following limitations. First, I construct a scenario in the experimental case in which the nonfinancial indicators for the nonprofit organization's service performance can easily be measured and compared across peer organizations. However, it is difficult in practice to design measures of service effectiveness that reflect the needs of all stakeholders while facilitating comparability across nonprofit organizations (Herman and Renz 2008; Campbell 2002; Kaplan 2001). Even so, this experiment captures the essential elements of the quality-of-service performance by simplifying the measurements of the service performance. Second, although this study investigates factors contributing to directors' perceptions of the procedural fairness of donor evaluations, it does not examine whether or how such fairness perceptions motivate directors to act on behalf of donors' interest by disciplining management's opportunistic behavior. I leave these questions to future study.