Intercollegiate athletics and institutional fundraising: a meta-analysis.

Martinez, J. Michael ; Stinson, Jeffrey L. ; Kang, Minsoo 等

Introduction

Since the late 1970s, numerous academic studies have purported to examine the effects of intercollegiate athletics on fundraising. Still, nearly 30 years later, disparate research designs and conflicting results have left researchers lacking the ability to confidently comment on how athletic programs influence donors to higher education. The goals of this meta-analysis are to clarify the common knowledge developed through previous research and to provide direction for continued research of how sports programs affect fundraising at colleges and universities.

Many of the weaknesses of this research stream lie in the various research designs employed. While most studies use fundraising, in one way or another, as the dependent variable, the calculation of fundraising varies dramatically. Some studies use aggregated institutional giving as the dependent variable (e.g., Gaski & Etzel, 1984; McCormick & Tinsley, 1990; Humphreys & Mondello, 2007; Stinson & Howard, 2007), while other studies examine individual donor behavior (e.g., Stinson & Howard 2004). Some studies include only alumni donors (e.g., Baade & Sundberg, 1996; Grimes & Chressanthis, 1994), while other studies include both alumni and non-alumni donors (e.g., Stinson & Howard, 2004; Humphreys & Mondello, 2007). Further, several studies have focused only on fundraising for intercollegiate athletics programs (e.g., Mahony, Gladden, & Funk, 2003; Gladden, Mahony, & Apostolopoulou, 2005; McEvoy, 2005), while other studies have included academic fundraising or general institutional support (e.g., Cunningham & ConchiFicano, 2002; Grimes & Chressanthis, 1994). Some studies have focused entirely on private schools (e.g., Shulman & Bowen, 2001; Turner, Meserve, & Bowen, 2001), other studies have examined only public schools (e.g., Grimes & Chressanthis, 1994; Stinson & Howard 2004), while yet other studies have included both public and private schools in the sample (e.g., Rhoads & Gerking, 2000; Humphreys & Mondello, 2007). Finally, many studies have essentially been case studies of a single institution (e.g., McCormick & Tinsley, 1990; Grimes & Chressanthis, 1994; Stinson & Howard 2004), while other studies have examined panel data for multiple institutions (e.g., Baade & Sundberg, 1996; Tucker, 2004; Stinson & Howard, 2007; Humphreys & Mondello, 2007). The end result of the inconsistency in research design is a set of confusing, often conflicting results.

The conclusions of these previous studies can essentially be classified into three categories: studies concluding there is little or no relationship between athletics and fundraising, studies concluding that intercollegiate athletics programs have a positive influence on fundraising (either for athletics and/or institutional support), and studies concluding that athletics programs have a negative effect on institutional fundraising (i.e., crowding-out effects). Several early studies concluded that intercollegiate athletics programs have little or no influence on fundraising. Notably, Gaski and Etzel (1984) constructed over 100 regressions on giving at multiple institutions and concluded that athletics programs had little influence on giving. More recently, Shulman and Bowen (2001) asserted in their study of academically elite, private schools that there was not a strong relationship between intercollegiate athletics and fundraising. In a literature review, Frank (2004) also concluded that the lack of consistent results across the research stream was suggestive of the lack of a strong relationship between athletics and giving.

Equally as many studies have concluded that there is a positive relationship between intercollegiate athletics and institutional fundraising. Sigelman and Carter (1979) were among the first authors to identify a positive effect of intercollegiate athletics on alumni giving. McEvoy (2005) also linked athletic performance positively to athletic fundraising. McCormick and Tinsley (1990) not only concluded that athletics had a positive relationship with athletic fundraising, but also with academic support. These authors noted that at Clemson University, a ten percent increase in athletic support was associated with a five percent increase in academic support. Similarly, Stinson and Howard (2008) concluded that athletic success enhanced both athletic and academic giving for NCAA Division I-AA and I-AAA institutions. Daughtry and Stotlar (2000) identified a positive effect of a NCAA Division II football national championship on institutional fundraising. Other authors have found that when included with other institutional variables and factors, intercollegiate athletics has a significant positive influence on fundraising, though other institutional factors, such as measures of academic quality may be more important determinants of giving (i.e., Rhoads & Gerking 2000, Cunningham & Conchi-Ficano 2002).

The third category of findings centers on potential crowding-out effects, where athletics programs may have a negative effect on institutional support. Crowding out effects in this context would typically be caused by donors increasing their giving to athletic programs and simultaneously decreasing their support of academic programs. Sperber (1990, 2000) asserted that crowding-out effects are responsible for a lack of academic support at some institutions, though he did not offer any empirical support. A case study at the University of Oregon (Stinson & Howard, 2004) concluded that crowding-out effects were possibly occurring amongst donors making annual gifts of over $1,000. A 2007 multi-institutional study extended these results, concluding that at most schools athletics fundraising was growing more quickly than academic fundraising and that crowding-out effects were most likely to occur at schools falling outside the top tier of academically ranked schools (Stinson & Howard, 2007).

As this brief review of research indicates, it is quite difficult to confidently draw any generalizable conclusions as to the effect intercollegiate athletics programs have on institutional giving. Given the number of studies conducted, and the over 30 years of research available, a meta-analysis is appropriate. There are four primary advantages to pursuing a meta-analytical approach in summarizing this research stream (Wilson, 2001): meta-analysis is a structured research technique that allows formal review of previous empirical work; meta-analysis is more sophisticated than traditional qualitative review studies; effect sizes across the studies may identify relationships that would otherwise be missed; and meta-analysis allows for the coordination and management of the various findings from each of the included studies. Successful meta-analysis will be useful in identifying the common effects across the research stream and for providing strong direction for future research.

The purpose of this study is to perform a meta-analytic review of the available scholarly research on the relationship between intercollegiate athletic success and institutional giving.

Methodology

Data Sources

The initial search for studies relevant to this meta-analytic review was conducted utilizing a selection of available Internet-based research databases including: ABI/INFORM Complete, Academic Search Premier, Dissertations and Theses Full-text from Proquest, EBSCO Electronic Journal Service, ERIC (Educational Resource Information Center), General OneFile, JSTOR, Management & Organization Studies: A SAGE Full-Text Collection, OmniFile Full-text Mega Edition, Social Sciences Full-text, and Web of Science. Among the key search terms included were combinations of athletic success, philanthropy, fundraising, institutional giving, and development. Articles published from 1975 through 2008 were included in this selection process. According to the literature, 1976 was the first year a peer-reviewed publication featured an article quantitatively investigating the relationship between intercollegiate athletic success and philanthropic giving (Budig, 1976). Additionally, manual ancestor searches and electronic descendent searches were conducted to compile additional studies, including relevant unpublished doctoral dissertations. This initial search resulted in more than 75 publications in the initial pool of studies. While the initial pool included more than 75 possible studies, a narrowing process was applied to determine a more targeted list of eligible studies for inclusion in the meta-analysis (i.e., quantitative exploration, comparable variables, etc.). The narrowing process focused on compiling a list of studies that met the criteria to conduct what DeCoster (2004) suggests is a "reasonable target for synthesis" (p. 6). For example, initial searches may include a study that highlights the relationship between athletic success and the benefits to a university. However, after further evaluation of that study, the relationship may make a mention of increases in institutional giving as a possible benefit, although no actual quantitative examination took place for that specific relationship. Following this process, a total of 26 publications remained from which the final dataset was taken. Although more than 65 percent of the initial pool was delimited, this is not uncommon in meta-analytic reviews (DeCoster, 2004). Further classification of the remaining 26 studies was completed as well, and that process is outlined in the following study selection process.

Study Selections

The collection of 26 studies was narrowed further based on whether or not the following criteria were included in the study: 1) measures of athletic success (win percentage, postseason appearances, and rankings); 2) a related outcome measure of philanthropic giving; 3) and designations of school type (public/private), sports of interest (football, basketball, etc.), level of athletic competition (NCAA divisions), pertinent giving targets (university, athletics, or academics), and giving bases (all donors or alumni donors only). Studies without appropriate statistical measures (i.e., regression weights) reported were excluded, as were studies that lacked the applicable data of interest. For example, a study which focused on the selected variables of interest may have been removed for only reporting the significant p-values of the relationship and not for including corresponding regression weights or correlations. Thus, the final sample consisted of 14 studies published between 1979 and 2008. A brief summary of included studies appears in Table 1.

Data Extraction

In concordance with the meta-analysis procedures outlined by DeCoster (2004), studies were then coded with pertinent study identifiers and select moderator variables extracted from the overall pool. Moderating variables included: 1) school type (public, private, or other); 2) giving targets (university, athletics, or academics); 3) giving base (all donors or alumni donors only); 4) NCAA classification (Division I, Division IAA, or other); and 5) sports of interest (football, basketball, or other)

Data Analysis

As previously mentioned, meta-analytic techniques for research synthesis are made stronger because of the added statistical evaluation used to aggregate the results of the included studies. Research indicates that the strength of any meta-analysis is the concept of effect size (DeCoster, 2004; Lipsey & Wilson, 2001). Effect size is the measure by which we can compare the varying degrees of relationships or "effects" between variables as reported in different studies. In the case of the current inquiry, the relationship explored in each study is essentially based on the effect of athletic success on fundraising. Although the included studies each operationalize athletic success in different ways (i.e., winning percentage, bowl appearances, etc.), evaluation of the overall relationship between athletic success and fundraising can be achieved using the meta-analytic techniques, which standardize the reported effects into one succinct metric of evaluation, effect size (Lipsey & Wilson, 2001).

Effect Size

For each of the selected studies, effect sizes were determined utilizing the product-moment correlation effect size (ESr) as outlined by Lipsey and Wilson (2001). Included studies reported relationships in three different ways: correlations (e.g., Sigelman & Carter 1979), t-ratios (e.g., Grimes & Chressanthis, 1994), and F-ratios (e.g., Daughtrey & Stotlar, 2000). Therefore, studies reporting values as either t-ratios or F-ratios were converted to correlations using the following formulas suggested by Lipsey and Wilson (2001, p. 201).

[MATHEMATICAL EXPRESSION NOT REPRODUCIBLE IN ASCII]

Furthermore, Lipsey and Wilson (2001) suggest that each of the calculated effect sizes be standardized using Fisher's [Z.sub.r]-transform (p. 63) in order to be able to generate proper standard error for each effect size, which is a key component for determining inverse variance weight. While the effect size statistic is the building block in any meta-analysis, the concept of inverse variance weight (w) also is important. The inverse variance weight allows researchers to more accurately compare effect sizes coming from varying sample sizes. Thus, an effect size with a sample size of 10 can easily be compared to an effect size calculated from a sample size of 100 by comparing the inverse weight variance. Accordingly, each effect size for the study was determined using the following equation:

[MATHEMATICAL EXPRESSION NOT REPRODUCIBLE IN ASCII]

Once each of the necessary effect sizes and inverse variance weights were calculated, a test of homogeneity (Q) was conducted through SPSS for overall effect size. The test of homogeneity is an important step in the meta-analysis process as it allows researchers to "determine if a grouping of effect sizes from different studies shows more variation than would be expected from sampling error alone" (Lipsey & Wilson, 2001, p. 8). Once the effect sizes are determined to be homogeneous or not, meta-analytic researchers must then determine whether a fixed-effects model or a random-effects model would be evaluated. According to Lipsey and Wilson, when the distribution of effect sizes is homogeneous (i.e., a non-significant Q value), the effect sizes differ from the population only because of subject-level sampling error (p. 119), prompting the use of the fixed-effects model. However, when the distribution is found to be heterogeneous (i.e., a significant Q-value), then a random effects model is evaluated, taking into account not only the differing effect size due to sampling error but also because of some other variables, operationalized as moderators. In the latter case, the moderator analyses were to be followed. Using the methods described by Lipsey and Wilson (2001), the effect of each moderator variable was examined. This final statistical procedure was calculated through SPSS version 17.0.

Results

Studies Selected

The initial search for applicable studies resulted in 26 empirical studies that focused on the relationship of athletic success and institutional giving in NCAA institutions. Among the 26 articles, 24 were from peer-reviewed journals, while two were unpublished dissertations. Following the inclusion and exclusion process, a final count of 14 studies were found, from which 75 different effect sizes could be calculated (Table 2).

Tests of Homogeneity

Homogeneity analysis (Q = 932.3, df = 74) was significant (p < .001). As a result, the random effects model was selected. The relationship between athletic success and giving resulted in a weighted mean effect size of 0.12 and a 95-percent confidence interval of 0.08 and 0.16. A minimum effect size (ESr) was -0.027 and a maximum effect size was 1.472, with a weighted standard deviation of 0.169. Athletic success, across the included studies, has a statistically significant, positive effect on giving.

Because of the heterogeneity of effect size distribution, follow-up moderator analysis was conducted.

Moderator Analysis

Moderator analysis for philanthropic giving (see Table 3) was tabulated on five moderators: institution type, giving target, giving base, NCAA classification, and sport of interest. Within meta-analytic studies, the aforementioned Q-statistic is necessary for comparison. While the overall Q-statistic allows researchers to establish the homogeneity of the data, evaluation of the [Q.sub.between] statistic allows researchers to determine how much of the variance can be explained by the moderating variables (DeCoster, 2004). The results indicate that only institution type did not significantly influence the strength of the relationships between athletic success and giving. For giving target, [Q.sub.between] showed a value of 20.91 (df = 2) and was significant (p < .001). While the effect sizes for all three groups are small (<.1), the results indicate that athletic success has a stronger influence on total institutional giving than either athletic or academic giving independently.

The [Q.sub.between] for giving base was 15.10 (df = 1) and was significant (p < .001). The effect size for alumni giving was twice the effect size for all donors (.08 vs. .04). Alumni donors appear to be more strongly influenced by athletic success than the general population.

[Q.sub.between] for NCAA classification had a value of 45.05 (df = 2) and was significant (p < .001). As shown in Table 3, the effect size for giving at Division I institutions was substantially higher than I-AA and other levels of NCAA membership. The effect of athletic success on giving is significantly higher at NCAA I-A schools.

Sport of interest had a [Q.sub.between] value of 26.57 (df= 2) and was significant (p < .001). Consistent with previous research documenting the power of football, football success had an effect size more than twice as large as the effect size for basketball (.09 vs. .04). The strongest effect on giving, at least at schools that compete in football, appears to be the success of the football program.

Institution type was the only non-significant (p=.072) moderator in the study with a [Q.sub.between] value of 5.26 (df = 2). The results of this meta-analysis suggest that donors at both public and private schools are influenced similarly by the success of athletic programs at those institutions.

Discussion

The primary objective of this study was to provide some conclusion to the seemingly disparate research results reported over the last 30 years regarding the influence intercollegiate athletics programs have on institutional fundraising. Using the differing ranges of correlation effect size (small [ES.sub.r] [less than or equal] .10; medium [ES.sub.r] = .25; large [ES.sub.r] [greather than or equal] .40) as cited by Lipsey and Wilson (2001), the overall results suggest that athletic success ([ES.sub.r] = .121) has a small but significant relationship with institutional giving (p < .001). In the aggregate, the 30 years of studies included in this meta-analysis indicate that intercollegiate athletics are an important influence on donors. However, the small overall effect sizes are indicative of the varying degrees of influence athletics has had in the reported studies. Both institutional heterogeneity and unmeasured influences on giving limit the ability to generalize results across institutions. Thus, further analysis was undertaken to better understand the influence of several potentially important moderating variables. (See Table 4 for a Summary of Results and Implications.)

The analysis of moderating variables provides clarification of the circumstances in which intercollegiate athletics are most likely to have the strongest influence on institutional giving. Specifically, this study identified four important moderators: giving target, alumni status of donor, NCAA classification, and focal sport of interest.

This study found that the effects of athletic success on giving were strongest when total institutional giving was considered, as opposed to athletic or academic giving specifically. Intuitively this finding makes sense. Studies examining aggregated institutional giving will pick up the positive effects of giving to both athletic and academic programs. This result, and the overall effect size of .121, is consistent with the findings of broader institutional studies that have concluded athletics has a small but positive relationship with institutional fundraising efforts (Rhoads & Gerking, 2000; Cunningham & Conchi-Ficano, 2002). Specific analysis on the differential impacts of athletic success on athletic versus academic giving reveals that athletic giving is only slightly more impacted by athletic success than academic giving. This finding, perhaps more than anything, underscores the potential value of athletics programs to assist in the growth of academic and institutional fundraising. While the effect sizes are small, it is clear that athletic success has a significant influence on all types of institutional fundraising. Similar to McCormick and Tinsley's (1990) study at Clemson University, it appears that successful athletic teams potentially benefit more than just athletic programs by spurring growth in fundraising across the campus.

Another significant moderator identified in the meta-analysis was the alumni status of the donor. Studies focusing solely on alumni giving resulted in an effect size nearly double that of studies based on giving from all institutional donors. This finding, while counter to many assertions made in the popular press (e.g., Sperber, 2000), is an important finding that may bear directly on whether athletics help or hurt institutional fundraising. Athletic performance, most notably football success, has been linked to increased numbers of alumni donors making gifts (Daughtrey & Stotlar, 2000; Stinson & Howard, 2008). Increased numbers of donors is indicative of the positive influence athletics can have on institutional fundraising. However, other studies have suggested that these alumni donors may alter their institutional giving patterns, both to athletic and academic programs based on athletic performance. In the extreme, alumni donors may cannibalize their academic giving to support athletics in times of success, potentially harming the institutional fundraising effort (Stinson & Howard, 2004; Stinson & Howard, 2007). Further work should concentrate specifically on the effect of athletics on pro-institutional alumni behaviors, including fundraising.

Not surprisingly, NCAA classification was also identified as an important moderator in the meta-analysis. Institutions competing at the NCAA I-A level of competition were more strongly influenced by the success of their respective athletic programs that institutions competing at lower levels. Several plausible explanations exist for this finding, though precise understanding awaits future research. NCAA I-A schools receive significantly more media attention, particularly for athletics, than other institutions. The high profile, focused more heavily on athletics, may prompt larger groups of donors (alums and non-alums) to identify specifically with the athletic teams at the institution. Non-alumni in particular may make gifts to the institution that would not be made in the absence of athletic success. This could potentially explain the larger effects at NCAA I-A schools. Also, the fact that the vast majority of studies examining athletic success and giving have been conducted at the NCAA I-A level provides a more thorough examination of giving at that level. Still, it is important to note that the effect sizes are significant and positive at all levels of NCAA competition. Stinson and Howard (2008) suggested that, in fact, athletic success at lower levels of competition (IAAA) was more important to increases in academic support than athletic success at the I-A level. Future research should continue to extend beyond high profile Division I-A schools.

Finally, this study identified a moderating role for the primary sport of interest. Football success displayed a much stronger influence on giving (by nearly 60 percent) than basketball or other sports. Previous studies have certainly documented the strong influence of various measures of football performance on giving (i.e., winning percentage, bowl appearances), so this finding is not necessarily surprising (Baade & Sundberg, 1996; Goff, 2000; Humphreys & Mondello, 2007; McCormick & Tinsley, 1990; Rhoads & Gerking, 2000; Stinson & Howard, 2007). Nevertheless, this result is important in that it continues to support the dominant role of football in athletic and institutional fundraising efforts. While not directly tested here, it seems possible that the ability of many football stadia to attract and seat more fans than basketball arenas heavily favors football in its influence. This is particularly true in cases where a university is able to "require" a donation for many or all season tickets in the respective arenas. Even though basketball teams typically play many more games in a season, they are more constrained than football in the number of season tickets that can be sold. This in turn constrains the required giving that can be generated through basketball programs. Thus, efforts to increase seating capacity and donor clubs that are associated with football programs appear to be paying off (Howard & Crompton, 2004). This finding is also consistent with research concluding that when football is not present (NCAA IAAA institutions), basketball as the focal sport is the primary influence on giving (Stinson & Howard, 2008). The conclusion of this meta-analysis is that the performance of the focal revenue-generating sport on a given campus is a significant influence on institutional fundraising. Researchers, and universities, should continue their efforts to understand how to use these athletic programs to best benefit the entire institutional fundraising effort.

Institution type was not statistically significant as a moderator across the studies included in this meta-analysis. In fact, the effect sizes for public and private schools were virtually identical, suggesting a similar influence of athletic success on giving at both types of institutions. This finding is somewhat surprising in relation to previous research. Many recent studies that have concluded that there is no relationship between intercollegiate athletics and fundraising have included data primarily drawn from academically elite private institutions (i.e., Shulman & Bowen 2001, Turner, Meserve, & Bowen, 2001). On the other hand, many of the recent studies concluding there are significant effects of athletics on giving include strong or sole representation of public institutions (i.e., Grimes & Chressanthis, 1994; McCormick & Tinsley, 1990; Rhoads & Gerking, 2000; Stinson & Howard, 2004). Possible explanations for this pattern of results may be indicative of other institutional factors as more powerful moderators than public/private status. For example, the level of investment in athletics programs may determine the extent of effects. Or, as Stinson and Howard (2008) reported, academic rankings may be a primary determinant of the influence of athletics on giving. Further research should continue to focus on these potential moderating factors.

Despite a thorough review and inclusion of the available research on the relationship of athletic success and institutional giving, this study is not without certain limitations. Foremost, there is a relatively small amount of quantitative-based research in the field of intercollegiate athletic relationship and giving, including a smaller percentage of studies that examine this relationship in terms of sports other than football and basketball and across differing NCAA classification levels. This fact may very well have led to the insignificant effect sizes for the institution type moderator. Still, the significant meta-analytic findings identified here are critical to understanding the influence of athletics programs on fundraising. The significant effects noted here represent a strong set of findings based on over 30 years of research in this area. As a result, we can be quite confident that intercollegiate athletic programs do have a significant influence on institutional fundraising, though the strength of that relationship is moderated by four important variables: the target of the giving, the alumni status of the donor, the institution's level of NCAA competition, and the primary sport of interest. Thus, these findings begin to set the conditions for when athletic performance is most likely to influence giving. Future research should continue to seek understanding of the underlying mechanisms causing these relationships, as well as extending itself to a consideration of other variables that may be important in the athletic performance-fundraising relationship. Further, research should be broadened to include other important variables subject to the influence of athletic programs, including institutional image, college choice and/or matriculation (Chressanthis & Grimes, 1993; Mixon, Trevino, & Minto, 2004), and academic quality (Goidel & Hamilton, 2006; McCormick & Tinsley, 1987; Mixon & Trevino, 2005).

References

* Baade, R. A., & Sundberg, J. O. (1996). Fourth down and goal to go? Assessing the link between athletics and alumni giving. Social Science Quarterly, 77(4), 789-803.

* Budig, J. E. (1976). The relationships among intercollegiate athletics, enrollment, and voluntary support for public higher education. Doctoral dissertation. Retrieved from Dissertations & Theses: Full Text. (Publication No. AAT 7630349).

Chressanthis, G. A., & Grimes, P. W. (1993). Intercollegiate sports success and first-year student enrollment demand. Sociology of Sport Journal, 10(3), 286-300.

* Coughlin, C. C., & Erekson, O. H. (1984). An examination of contributions to support intercollegiate athletics. Southern Economic Journal, 51(1), 180-195.

Cunningham, B. M., & Cochi-Ficano, C. K. (2002). The determinants of donative revenue flows from alumni of higher education. Journal of Human Resources, 37, 540-569.

* Daughtrey, C., & Stotlar, D. (2000). Donations: Are they affected by a football championship? Sport Marketing Quarterly, 9(4), 185-193.

DeCoster, J. (2004). Meta-analysis notes. Retrieved July 15, 2009, from http://www.stat-help.com/notes.html

Frank, R. H. (2004). Challenging the myth: A review of links among college athletic success, student quality, and donations. New York: Knight Foundation Commission on Intercollegiate Athletics.

Gaski, J. F., & Etzel, M. J. (1984). Collegiate athletic success and alumni generosity: Dispelling the myth. Social Behavior and Personality, 12(1), 29-38.

Gladden, J. M., Mahoney, D. F., & Apostolopoulou, A. (2005). Toward a better understanding of college athletic donors: What are the primary motives? Sport Marketing Quarterly, 14, 18-30.

Goff, B. (2000). Effect of university athletics on the university: A review and extension of empirical assessment. Journal of Sport Management, 14, 84104.

Goidel, R. K., & Hamilton, J. M. (2006). Strengthening higher education through gridiron success? Public perceptions of the impact of national football championships on academic quality. Social Science Quarterly, 87(4), 851-862.

* Grimes, P. W., & Chressanthis, G. A. (1994). Alumni contributions to academics: The role of intercollegiate sports and NCAA sanctions. American Journal of Economics & Sociology, 53, 27-41.

Howard, D. R., & Crompton, J. L. (2004). Financing sport (2nd ed.). Morgantown, WV: Fitness Information Technology.

Humphreys, B. R., & Mondello, M. (2007). Intercollegiate athletic success and donations at NCAA division I institutions. Journal of Sport Management, 21 , 265-280.

Lipsey, M. W., & Wilson, D. B. (2001). Practical meta-analysis. Thousand Oaks, CA: Sage.

* Litan, R. E., Orszag, J. M., & Orszag, P. R. (2005). The empirical effects of Intercollegiate athletics. Indianapolis, IN: National Collegiate Athletic Association.

Mael, F., & Ashforth, B. E. (1992). Alumni and their alma mater: A partial test of the reformulated model of organizational identification. Journal of Organizational Behavior, 13, 103-123.

Mahony, D. F., Gladden, J. M., & Funk, D. C. (2003). Examining athletic donors at NCAA Division I institutions. International Sports Journal, 7, 9-27.

McCormick, R. E., & Tinsley, M. (1987). Athletics versus academics? Evidence from SAT scores. The Journal of Political Economy, 95(5), 1103-1116.

McCormick, R. E., & Tinsley, M. (1990). Athletics and academics: A model of university contributions. In B. L. Goff & R. D. Tollison (Eds.), Sportometrics (pp. 193-206). College Station, TX: Texas A&M University Press.

McEvoy, C. (2005). Predicting fundraising revenues in NCAA Division I-A intercollegiate athletics. The Sport Journal, 8(1).

Meer, J., & Rosen, H. S. (in press). The impact of athletic performance on alumni giving: An analysis of micro data. Economics of Education Review.

Mixon, F. G., & Trevino, L. J. (2005). From kickoff to commencement: The positive role of intercollegiate athletics in higher education. Economics of Education Review, 24(1), 97-102.

Mixon, F. G., Trevino, L. J., & Minto, T. C. (2004). Touchdowns and test scores: Exploring the relationship between athletics and academics. Applied Economics Letters, 11(7), 421-424.

* Rhoads, T. A., & Gerking, S. (2000). Educational contributions, academic quality and athletic success. Contemporary Economic Policy, 18(2), 248-258.

Shulman, J. L., & Bowen, W. G. (2001), The fame of life, Princeton, NJ: Princeton University Press.

* Sigelman, L., & Bookheimer, S. (1983). Is it whether you win or lose? Monetary contributions to big-time college athletic programs. Social Science Quarterly, 64(2), 347-359.

* Sigelman, L., & Carter, R. (1979). Win one for the giver? Alumni giving and big-time college sports. Social Science Quarterly, 60(2), 284-294. Sperber, M. (1990). College sports inc. New York: Henry Holt.

Sperber, M. (2000). Beer and circus: The impact of big-time college sports on undergraduate education. New York: Henry Holt.

Stinson, J. L., & Howard, D. R. (2004). Scoreboards vs. mortarboards: Major donor behavior and intercollegiate athletics. Sport Marketing Quarterly, 13, 73-81.

* Stinson, J. L., & Howard, D. R. (2007). Athletic success and private giving to athletic and academic programs at NCAA institutions. Journal of Sport Management, 21(2), 237-266.

* Stinson, J. L., & Howard, D. R. (2008). Winning does matter: Patterns in private giving to athletic and academic programs at NCAA Division IAA and I-AAA institutions. Sport Management Review 11(1), 1-20.

* Tucker, I. B. (2004). A reexamination of the effect of big-time football and basketball success on graduation rates and alumni giving rates. Economics of Education Review, 23(6), 655-661.

* Turner, S. E., Meserve, L. A., & Bowen, W. G. (2001). Winning and giving: Football results and alumni giving at selective private colleges and universities. Social Science Quarterly, 82(4), 812-826.

* References marked with an asterisk indicate studies included in meta-analysis.

J. Michael Martinez, PhD, is an assistant professor in the Department of Kinesiology and Health Promotion at Troy University. His research interests include charitable giving, intercollegiate athletics, internal branding, and media consumption.

Jeffrey L. Stinson, PhD, is an assistant professor of marketing at Central Washington University. His research interests include charitable giving, intercollegiate athletics, and education institution branding.

Minsoo Kang, PhD, is an associate professor in the Department of Health & Human Performance at Middle Tennessee State University. As a measurement specialist, he has contributed to the manuscript by analysis and interpretation of meta- analysis data, and drafting and revising the manuscript.

Colby B. Jubenville, PhD, is an associate professor of sport management at Middle Tennessee State University. His research interests include branding, sport sponsorship, sport marketing, systems thinking, and sport policy.

Table 1.
Summary of Included Studies

Study Independent Dependent Findings
 Variables Variables

Baade & Records for Alumni giving Bowl appearances,
Sundberg men's football NCAA BB tournament
(1996) and basketball appearances
 teams, associated with
 post-season significant
 appearances increases in alumni
 giving at doctorate
 institutions.

Budig (1976) Football and Alumni giving No relationship
 men's at 79 public between team records
 basketball institutions and giving.
 records

Coughlin & Football and Contributions Football success
Erekson men's to the (record and bowl
(1984) basketball athletic appearance) and
 records, department basketball winning
 post-season percentage
 appearances significantly
 related to athletic
 department
 contributions.

Daughtrey & Football Alumni Positive impact of
Stotlar championships donation championship season
(2000) in NCAA I-AA, amount and on athletic
 II and III number of department donations
 alumni donors and number of
 athletic donors at
 DII and III
 institutions. Also,
 significant
 increases in
 institutional
 contributions at
 DIII schools, and
 increase in number
 of total donors at
 I-AA institutions.

Grimes & Winning Alumni Overall winning
Chressanthis percentage, contributions percentage of all
(1994) television to the three sports
 appearances, Mississippi positively
 and post-season State associated with
 appearances by University alumni
 football, men's Foundation contributions.
 basketball, and
 baseball teams

Litan, Athletic Alumni giving No relationship
Orzag, & operating between football
Orzag (2005) expenditures, success and alumni
 football giving.
 winning
 percentages

Rhoads & Post-season Total and Alumni contributions
Gerking success of alumni giving increase with bowl
(2000) football and from 87 NCAA win; decrease if
 men's Division I team placed on
 basketball institutions probation.
 teams, athletic
 probation,
 athletic
 tradition

Sigelman & Football Institutional Winning football
Bookheimer records giving teams correlate with
(1983) increased athletic,
 but not
 institutional,
 donations.

Sigelman & Football and Change in Neither record nor
Carter men's alumni giving change in record of
(1979) basketball football or men's
 records, change basketball teams
 in football and significant
 men's predictors of
 basketball changes in alumni
 records giving.

Stinson & Football Academic and Total giving to
Howard records, athletic giving top-ranked academic
(2007) post-season to NCAA I-A institutions less
 appearance, institutions susceptible to
 post-season influence of
 success, athletic success.
 tradition Increasing
 percentage of
 donations at all
 schools allocated
 toward athletics.

Stinson & Men's Academic and Successful
Howard basketball athletic giving basketball teams
(2008) records, to NCAA positively influence
 post-season I-AA/AAA the number of donors
 appearance, institutions making gifts and the
 post-season average size of
 success, those gifts, both
 tradition for academic and
 athletic programs.

Tucker Football and Alumni giving Football success
(2004) men's positively
 basketball influences alumni
 records giving; basketball
 success is not a
 statistically
 significant
 influence on alumni
 giving.

Turner, Football Individual Football won-loss
Meserve, & won-loss donor giving records not
Bowen records data from significantly
(2001) College and related to either
 Beyond dataset athletic or general
 giving
 rates/amounts.

Table 2.
Summary of Effect Sizes

Study Year School Type Ath./Univ. Alumni

Budig 1976 Combination University Alumni
Budig 1976 Combination University Alumni
Budig 1976 Combination University Alumni
Budig 1976 Combination University Total
Budig 1976 Combination University Total
Sigelman & Carter 1979 Combination University Alumni
Sigelman & Carter 1979 Combination University Alumni
Sigelman & Carter 1979 Combination University Alumni
Brooker & Klastorin 1981 Public University Total
Sigelman & Bookheimer 1983 Combination Athletics Alumni
Sigelman & Bookheimer 1983 Combination Athletics Alumni
Coughlin & Erekson 1984 Combination Athletics Total
Coughlin & Erekson 1984 Combination Athletics Total
Grimes & Chressanthis 1994 Public University Alumni
Grimes & Chressanthis 1994 Public University Alumni
Grimes & Chressanthis 1994 Public University Alumni
Grimes & Chressanthis 1994 Public University Alumni
Grimes & Chressanthis 1994 Public University Alumni
Grimes & Chressanthis 1994 Public University Alumni
Grimes & Chressanthis 1994 Public University Alumni
Grimes & Chressanthis 1994 Public University Alumni
Baade & Sundberg 1996 Private University Alumni
Baade & Sundberg 1996 Private University Alumni
Baade & Sundberg 1996 Public University Alumni
Baade & Sundberg 1996 Public University Alumni
Baade & Sundberg 1996 Liberal Arts University Alumni
Baade & Sundberg 1996 Liberal Arts University Alumni
Daughtrey & Stotlar 2000 Combination University Alumni
Daughtrey & Stotlar 2000 Combination Athletics Alumni
Rhoads & Gerking 2000 Combination University Total
Rhoads & Gerking 2000 Combination University Total
Rhoads & Gerking 2000 Combination University Alumni
Rhoads & Gerking 2000 Combination University Alumni
Turner, Meserve, & Bowen 2001 Private University Alumni
Turner, Meserve, & Bowen 2001 Private Athletics Alumni
Litan, Orzag, & Orzag 2003 Combination University Alumni
Tucker 2004 Combination University Alumni
Tucker 2004 Combination University Alumni
Tucker 2004 Combination University Alumni
Tucker 2004 Combination University Alumni
Tucker 2004 Combination University Alumni
Tucker 2004 Combination University Alumni
Stinson & Howard 2007 Combination University Total
Stinson & Howard 2007 Combination University Alumni
Stinson & Howard 2007 Combination Athletics Total
Stinson & Howard 2007 Combination Academic Total
Stinson & Howard 2007 Combination University Total
Stinson & Howard 2007 Combination University Alumni
Stinson & Howard 2007 Combination Athletics Total
Stinson & Howard 2008 Combination University Total
Stinson & Howard 2008 Combination University Alumni
Stinson & Howard 2008 Combination Athletics Total
Stinson & Howard 2008 Combination Athletics Alumni
Stinson & Howard 2008 Combination Academic Total
Stinson & Howard 2008 Combination Academic Alumni
Stinson & Howard 2008 Combination University Total
Stinson & Howard 2008 Combination University Total
Stinson & Howard 2008 Combination Athletics Total
Stinson & Howard 2008 Combination Athletics Total
Stinson & Howard 2008 Combination Academic Total
Stinson & Howard 2008 Combination Academic Total
Stinson & Howard 2008 Combination University Total
Stinson & Howard 2008 Combination University Total
Stinson & Howard 2008 Combination University Alumni
Stinson & Howard 2008 Combination University Alumni
Stinson & Howard 2008 Combination University Alumni
Stinson & Howard 2008 Combination Athletics Total
Stinson & Howard 2008 Combination Athletics Total
Stinson & Howard 2008 Combination Athletics Alumni
Stinson & Howard 2008 Combination Athletics Alumni
Stinson & Howard 2008 Combination Athletics Alumni
Stinson & Howard 2008 Combination Academic Total
Stinson & Howard 2008 Combination Academic Total
Stinson & Howard 2008 Combination Academic Alumni
Stinson & Howard 2008 Combination Academic Alumni
Stinson & Howard 2008 Combination Academic Alumni

Study NCAA Class Sport Success

Budig Combo w/DI FB Win Pct.
Budig Combo w/DI BB Win Pct.
Budig Combo w/DI BB Win Pct.
Budig Combo w/DI FB Win Pct.
Budig Combo w/DI BB Win Pct.
Sigelman & Carter Div. I FB Post
Sigelman & Carter Div. I FB Win Pct.
Sigelman & Carter Div. I BB Win Pct.
Brooker & Klastorin Div. I-A FB Post
Sigelman & Bookheimer Div. I FB Win Pct.
Sigelman & Bookheimer Div. I BB Win Pct.
Coughlin & Erekson Div. I FB Post
Coughlin & Erekson Div. I BB Win Pct.
Grimes & Chressanthis Div. I FB Post
Grimes & Chressanthis Div. I BB Post
Grimes & Chressanthis Div. I Other Post
Grimes & Chressanthis Div. I Total Post
Grimes & Chressanthis Div. I FB Win Pct.
Grimes & Chressanthis Div. I BB Win Pct.
Grimes & Chressanthis Div. I Other Win Pct.
Grimes & Chressanthis Div. I Total Win Pct.
Baade & Sundberg Combo w/DI FB Post
Baade & Sundberg Combo w/DI BB Post
Baade & Sundberg Combo w/DI FB Post
Baade & Sundberg Combo w/DI BB Post
Baade & Sundberg Combo w/DI FB Win Pct.
Baade & Sundberg Combo w/DI BB Win Pct.
Daughtrey & Stotlar Combo w/o DI FB Post
Daughtrey & Stotlar Combo w/o DI FB Post
Rhoads & Gerking Div. I FB Post
Rhoads & Gerking Div. I BB Post
Rhoads & Gerking Div. I FB Post
Rhoads & Gerking Div. I BB Post
Turner, Meserve, & Bowen Combo w/DI FB Win Pct.
Turner, Meserve, & Bowen Combo w/DI FB Win Pct.
Litan, Orzag, & Orzag Div. I-A FB Win Pct.
Tucker Div. I-A FB Post
Tucker Div. I-A BB Post
Tucker Div. I-A FB Rank
Tucker Div. I-A BB Rank
Tucker Div. I-A FB Win Pct.
Tucker Div. I-A BB Win Pct.
Stinson & Howard Div. I FB Post
Stinson & Howard Div. I FB Post
Stinson & Howard Div. I FB Post
Stinson & Howard Div. I FB Post
Stinson & Howard Div. I FB Win Pct.
Stinson & Howard Div. I FB Win Pct.
Stinson & Howard Div. I FB Win Pct.
Stinson & Howard Div. I-AA FB Post
Stinson & Howard Div. I-AA FB Post
Stinson & Howard Div. I-AA FB Post
Stinson & Howard Div. I-AA FB Post
Stinson & Howard Div. I-AA FB Post
Stinson & Howard Div. I-AA FB Post
Stinson & Howard Div. I-AA BB Post
Stinson & Howard Combined BB Post
 (I-AA, I-AAA)
Stinson & Howard Div. I-AA BB Post
Stinson & Howard Combined BB Post
 (I-AA, I-AAA)
Stinson & Howard Div. I-AA BB Post
Stinson & Howard Combined BB Post
 (I-AA, I-AAA)
Stinson & Howard Div. I-AA FB Rank
Stinson & Howard Div. I-AA BB Rank
Stinson & Howard Div. I-AA BB Rank
Stinson & Howard Div. I-AA FB Rank
Stinson & Howard Combined BB Rank
 (I-AA, I-AAA)
Stinson & Howard Div. I-AA FB Rank
Stinson & Howard Div. I-AA BB Rank
Stinson & Howard Div. I-AA BB Rank
Stinson & Howard Div. I-AA FB Rank
Stinson & Howard Combined BB Rank
 (I-AA, I-AAA)
Stinson & Howard Div. I-AA FB Rank
Stinson & Howard Div. I-AA BB Rank
Stinson & Howard Div. I-AA BB Rank
Stinson & Howard Div. I-AA FB Rank
Stinson & Howard Combined (I-AA, I-AAA) BB Rank

Study N [ES.sub.r]

Budig 58 0.696
Budig 62 0.499
Budig 56 0.588
Budig 58 0.336
Budig 58 0.200
Sigelman & Carter 100 0.098
Sigelman & Carter 100 0.131
Sigelman & Carter 100 0.001
Brooker & Klastorin 58 0.173
Sigelman & Bookheimer 57 0.335
Sigelman & Bookheimer 57 0.092
Coughlin & Erekson 56 0.290
Coughlin & Erekson 56 0.202
Grimes & Chressanthis 30 0.021
Grimes & Chressanthis 30 0.128
Grimes & Chressanthis 30 0.028
Grimes & Chressanthis 30 0.157
Grimes & Chressanthis 30 0.022
Grimes & Chressanthis 30 0.246
Grimes & Chressanthis 30 0.022
Grimes & Chressanthis 30 0.172
Baade & Sundberg 575 0.138
Baade & Sundberg 575 -0.027
Baade & Sundberg 1025 0.100
Baade & Sundberg 1025 0.096
Baade & Sundberg 1688 0.040
Baade & Sundberg 1688 -0.016
Daughtrey & Stotlar 33 0.231
Daughtrey & Stotlar 33 0.428
Rhoads & Gerking 821 0.089
Rhoads & Gerking 821 0.097
Rhoads & Gerking 821 0.110
Rhoads & Gerking 821 0.089
Turner, Meserve, & Bowen 125 0.469
Turner, Meserve, & Bowen 125 0.400
Litan, Orzag, & Orzag 371 0.900
Tucker 78 0.228
Tucker 78 0.002
Tucker 78 0.204
Tucker 78 0.006
Tucker 78 0.268
Tucker 78 0.015
Stinson & Howard 747 0.018
Stinson & Howard 795 0.035
Stinson & Howard 629 0.020
Stinson & Howard 707 0.028
Stinson & Howard 773 0.028
Stinson & Howard 856 0.002
Stinson & Howard 657 0.074
Stinson & Howard 232 0.032
Stinson & Howard 232 0.033
Stinson & Howard 631 0.064
Stinson & Howard 631 0.067
Stinson & Howard 233 0.034
Stinson & Howard 233 0.034
Stinson & Howard 428 0.068
Stinson & Howard 967 0.027
Stinson & Howard 579 0.016
Stinson & Howard 1219 0.007
Stinson & Howard 420 0.069
Stinson & Howard 965 0.028
Stinson & Howard 501 0.013
Stinson & Howard 277 0.025
Stinson & Howard 277 0.018
Stinson & Howard 501 0.046
Stinson & Howard 1032 0.008
Stinson & Howard 581 0.019
Stinson & Howard 575 0.002
Stinson & Howard 575 0.008
Stinson & Howard 581 0.004
Stinson & Howard 1286 0.014
Stinson & Howard 499 0.052
Stinson & Howard 278 0.024
Stinson & Howard 278 0.016
Stinson & Howard 499 0.045
Stinson & Howard 1033 0.010

Note: BB = Basketball, FB = Football, Win Pct. = Winning Percentage,
Post = Postseason Appearance, Rank = National Ranking

Table 3.

Moderator Analysis

Institution Type Effect Size Effect Size CI

Public 0.10 0.06-0.14
Private 0.09 0.05-0.14
Combination 0.06 0.05-0.07

Giving Target Effect Size Effect Size CI

University 0.09 0.07-0.10
Athletic 0.04 0.02-0.06
Academic 0.03 0.00-0.06

Giving Base Effect Size Effect Size CI

Total 0.04 0.02-0.06
Alumni Only 0.08 0.07-0.10

NCAA Classification Effect Size Effect Size CI

Division I-A 0.12 0.10-0.15
Division I-AA 0.05 0.03-0.06
Combination 0.04 0.01-0.06

Sport of Interest Effect Size Effect Size CI

Football 0.09 0.08-0.11
Basketball 0.04 0.02-0.05
Other 0.09 -0.09-0.29

Table 4.
Summary of Results and Implications

Key Finding

* Across 30 years of empirical study, intercollegiate athletics has
a small, but significant influence on institutional fundraising.

* Effects of athletic success are strongest when institutional
giving is considered.

* Alumni donors are more influenced by athletic success than
non-alumni donors.

* Football is the primary influence on giving at institutions where
football is played.

* Public institutions and private schools have nearly identical
effect sizes, indicating all schools may be influenced by athletic
performance.

Implications for Practice

* Institutional fundraisers should invest in understanding the role
athletics play in giving and learn to leverage athletic success to
benefit institutional fundraising efforts.

* Athletic success may be a positive influence on both athletic and
academic fundraising. Academic fundraisers should consider the
ability of athletic programs to assist in donor development.

* Contrary to some popular press, alumni should be targeted with
campaigns leveraging the role of athletics in fundraising. In
particular, successful athletics programs appear to bring more
alumni donors to the institution. Care should be taken to
successfully develop and cultivate these new donors.

* Fundraisers should pay particular attention to the football team.
Positive shaping of expectations and perceptions of success may
enhance institutional fundraising.

* Both public and private schools appear influenced, and therefore
may leverage athletic success in fundraising.