STUDENT LOANS: A MULTIVARIATE ANALYSIS OF PLANNERS, USERS, AND NON-USERS

Abstract

Student financial-aid programs have changed drastically over the past twenty-five years. Many new loan and grant programs have come into existence at the federal, state, and institution level over this period. While more financial-aid programs are available now than in the past, little has been published regarding the financial-aid plans of high-school seniors and the likelihood of realization of these plans for students of different socioeconomic status (SES) or ability. Nor has there been much published regarding the dropout rates of students using various forms of financial aid. The purpose of this paper was to investigate these two topics with a special emphasis on comparing loan users with users of other forms of aid. The financial-aid groups used most often in these analyses were loans only, loans in combination with other aid, other aid only, and familial aid only. Differences in SES, ability, type of school, and type of loan for students planning various types of aid were analyzed statistically using multivariate analysis of variance (MANOVA), analysis of variance (ANOVA), or Chi-square tests. Similar analyses were performed for students using various forms of financial aid in their first postsecondary schooling. Comparisons of the planners and users are made for the different types of aid and other variables. Differences in dropout rates for students in the several financial-aid categories were analyzed using ANOVA. The variables SES, ability, grade-point average, and parental income were used in these ANOVA's as blocking factors to obtain separate estimates of dropout rates for categories of these variables as well as to test for differences in these rates. The final set of analyses in this paper tested for differences in SES, ability, and grade-point average of students in the several financial-aid groups who persist, stop out, or drop out. These tests were accomplished using MANOVA. The dropout rate analyses and the persist, stopout, and dropout analyses were accomplished for academic years beginning in 1972, 1973, and 1974 and thus allow longitudinal estimates of the dropout rates and other variables tested. Data for these analyses were extracted from the National Longitudinal Study of the High School Class of 1972. Major findings follow. The planning and using analysis in this paper showed that students planning to use loans had significantly lower SES than those not planning loans. Further, when first-year loan users are compared larger percentages of low SES students were using loans than students in the two higher categories of SES. Vocational school students not only plan to use loans at higher rates than students planning other schools, but have a much higher percentage use of loans only than do students at other schools. Low SES students at these schools seem to bear a disproportionate loan burden when compared to students at other schools. Vocational students also use less desirable loan programs at higher rates than other students. When dropout rates of financial-aid groups are compared for this three-year study, it is evident the first year of schooling is critical. First-year dropout rates are higher than those in the following two years; type of aid used in the first year has an apparent effect on first-year students not indicated by the later analyses. Students relying on loans only or familial aid only have higher dropout rates than students using loans in combination with other types of aid. This phenomenon appears to affect levels of SES differentially. Low SES students have higher dropout rates when relying exclusively on loans or on family aid than do high SES students.