A STUDY OF PREDICTORS OF ALUMNI PHILANTHROPY IN PRIVATE UNIVERSITIES

College alumni are in a powerful position to influence all sources of voluntary support to higher education. An increase in alumni support can result in gifts and grants from foundations, corporations, and private benefactors which exceed the amount of alumni contributions manyfold. Surprisingly little is known, however, about the characteristics and attitudes of alumni donors.^ The purpose of this study was to test the ability of fourteen demographic and attitudinal variables to discriminate between alumni donors and non-donors, as well as limited donors and substantial donors, to the annual campaign of a private research university in the Northeast.^ A questionnaire survey was mailed to 1000 randomly-selected alumni who had graduated from the institution between 1960 and 1969. Applications of one-way multivariate analysis of variance for 744 respondents, blocking on year of graduation by academic performance and number of extracurricular activities, indicated that the sample was relatively homogeneous and could be analyzed as a group.^ To test the initial hypothesis of no difference between the donor and non-donor groups on each of the independent variables, analysis of variance procedures were applied. To assess the combined ability of these variables to predict group membership as donor or non-donor, a single function, two-group discriminant analysis procedure was applied; this procedure was then repeated for a disaggregated sample to assess the prediction of group membership as limited donor or substantial donor.^ Donors and non-donors differed significantly on eight variables: year of graduation, undergraduate school, receipt of institutional scholarship or grant, enrollment at another college subsequent to graduation, distance of permanent residence from campus, current occupation, satisfaction with undergraduate preparation for first job, and emotional attachment to alma mater. For the donor versus non-donor discriminant analysis, the discriminant function correctly predicted group membership for 64.11 percent of the cases, 24.8 percent better than chance. Emotional attachment, current occupation, undergraduate school, and receipt of institutional scholarship or grant were the best predictors of group status. For the limited donor versus substantial donor discriminant analysis, the discriminant function predicted group membership for 75.00 percent of the cases, 55.2 percent better than chance. Emotional attachment and current occupation were the best predictors of limited or substantial donor status.^ Suggestions for further study included replication of the study in varying types of institutions, such as public research universities, regional state universities, and liberal arts colleges. Also suggested were field validation of the optimal classification equations and applications of research designs with additional variables.