OPTIMAL ALLOCATION OF FEDERAL STUDENT FINANCIAL AID: A SIMULATION

Abstract

This study was a simulation involving packaging financial aid applicants with reduced Federal campus-based funds in order to examine the differences between two packaging models. With {dollar}1.5 million as a funding base, students were packaged twice, once using a Modified Self Help model and the second time utilizing an unused, two-stage packaging method. The population for the study consisted of financial aid applicants at Brooklyn College whose financial records were already computerized and whose needs had been assessed by an accepted uniform methodology. In order to study population factors, a +10% and a -10% population variable (based on the current 1,983 students), was also hypothesized.;Based on the three population factors, thirty hypotheses were formulated to study two alternative methods of financial aid packaging. The hypotheses were categorized by dependency (independent/dependent student) and by need (full need/partial need) as well as by population factors.;The theoretical basis for this study encompassed distributive and economic justice and social welfare theory. The main economic theorists on which this study was based were Pareto, Bergson, Kaldor, Hicks, Arrow and Little.;Data were analyzed using Chi Square (X('2)) statistics to determine whether the differences in packaging methods were statistically significant. In addition, a contingency coefficient (C) was used to determine the strength of association between the choice of Model and the number of students assigned to various aid categories. Out of the thirty hypotheses tested, three dealing with dependent full need students and two with partial need students in the -10% population category, were supported. Two other analyses were performed, one by population grouping using X('2) to test all students packaged and another testing the interaction of the Models and the total numbers of students in the Need Categories (full and partial).;Overall, looking at the various dependency and need categories, there was a statistically significant difference in the use of the new model over the old.;Thus, through the simulations, a decision maker is able to observe and compare the results of his options before he selects one for implementation.