| This dissertation research explored the possible patterns in the extreme financial strategies employed by full-time, full-year in-state students enrolled in 4-year public college education institutions in the United States. These financial behaviors included: failure to file the FAFSA, parental under-contribution, working full time, carrying credit card debt, and taking private loans.;The restricted database of National Postsecondary Student Aid Study (NPSAS):04, the most comprehensive national database available on college student finance, was used in this study. By using 6 filter variables, a nationwide sub-sample of 8,837 students enrolled in 294 4-year public higher education institutions and a 12-state-subsample of 4,125 students enrolled in 154 public 4-year higher education institutions were selected.;Nationwide two-level HLM files and three-level HLM files for 12-state-subset were created, and logistic and multiple regression models were run; however greater attention was focused on the 12 states that have state-representative data. This emphasis was necessary for two reasons: (1) higher education financial issues might be very different across states due to the differences in tuition levels and living costs, state student financial aid programs, state economic environments, and racial/ethnic and SES components of the population. Three-level HLM files for the 12-state-subset can separate state effect from individual and school effects; and (2) the 3-level HLM files built for the 12-state-subset study can handle the complex data structure of NPSAS: 04 better than the 2-level HLM files for the nationwide study, therefore making the findings more accurate.;The findings from 12-state-subset models indicated that student year, race, and family income all played significant roles in extreme financial behaviors, each associated with at least three extreme behaviors, while parental education level, primary language, gender, GPA, and unmet financial need were also associated with one or two extreme behaviors. Among extreme behaviors, there were significant inter-correlations as well. GPA was found negatively associated with overworking, taking private loans, and carrying credit card debt; however, the leading factors that affected GPA may be race, student year, and gender, and not extreme financial behaviors.;The conclusions are: (1) In financing public 4-year college education with the assistance of the student financial aid system in the U.S., parental contribution is playing an indispensable role for dependent students. Students' over-working and private loan-taking seemed to be, at least in part, making up for a significant amount of missing parental contribution. (2) Hispanic students, seniors, and students from lower income families were more likely to work full-time than other students. Students enrolled in highly selective institutions and/or in institutions with large enrollment were less likely to work full-time. Over-working students tended to carry significant credit card debt and had lower GPA than others, but they were less likely to fail to apply for federal financial aid. (3) Credit card debt was incurred most often among graduating seniors, full-time working students, and students with significant private loans. However, credit card debt might have more to do with a student's spending habits rather than real financial needs. (4) Private loans were complementary to Federal Stafford loans. Students whose primary language was not English were more likely to take private loans. Institutions located in suburban areas had the highest percentage of students with significant private loans. |
