For the institution and its regional campuses under review, the amount of outstanding student liabilities (i.e., monies owed by the student to the institution for educational-related expenditures not satisfied by the end of the term) grew at an alarming rate between 2000 and 2014, with some of the institution's regional campuses experiencing over a 400% increase in student liabilities between 2011 and 2014. This increase in student liabilities mirrors the trends in the rise of federal student loan borrowing and federal student loan default. The purpose of this study is to explore the relationship between various student characteristics and student liabilities owed directly to the institution at a multi-campus public university in the Midwest between the years 2011 and 2014. The period under review was selected due to its high increases in student liabilities. This study will further explore how these relationships differ between the university's multiple regional campuses and main campus. The students in the current study are students that have an outstanding liability to the organization of $100 or more, have been turned over to an outside collection agency for collections, and have had their student account balance written-off. The characteristics explored in the current study are many of the same characteristics shown in the current literature to be associated with federal student loan default rates. The characteristics are first generation vs. not a first generation student (i.e., FG vs. NFG), predominant campus location (i.e., main vs. regional campus), class status (i.e., freshmen, sophomore, junior, or senior), sex, underrepresented minorities vs. not an underrepresented minority (i.e., URM vs. NURM), federal Pell grant vs. not a federal Pell grant recipients (i.e., FP vs. NFP), federal student loans vs. not a federal student loan recipients (i.e., FL vs. NFL), private student loan vs. not a private loan recipients, and GPA. To explore the relationship between the variables and student liabilities, the following questions were addressed: (1) What are the FG/NFG, campus location, class status, sex, URM, grant status, and loan characteristics of students with outstanding liabilities at Midwestern University (MU)? (2) How does FG/NFG, campus location, class status, sex, URM, grant status, and loan characteristics of students with outstanding liabilities at MU, differ between students at the main campus vs. the regional campus? (3) What is the relationship between FG/NFG, class status, sex, URM, grant status, and loan characteristics of students with outstanding liabilities at MU? (4) How does the relationship between FG/NFG, class status, sex, URM, grant status, and loan status characteristics of students with outstanding liabilities at MU differ between the regional and main campus? The study was conducted in two stages. First, using SPSS basic descriptive statistics were presented to address questions one and two. Second, multiple regression was employed to investigate the relationship between the variables addressed in questions three and four. However, due to the violation of the assumption of normality needed for multiple regression, a second set of statistical analysis, the Chi-square test (x[superscript 2]), was deployed to further address questions three and four. Violations of normality can lead to over- or under-confident estimates of the precision of regression coefficients, and unreliable confidence intervals and significance tests. The result of the chi-square testing was considered statistically significant for all the variables tested. Overall, the variables of first generation students, freshmen, underrepresented minorities, Pell and federal loan recipients, and GPA had the greatest contribution to the chi-square test with all having a residual value greater than 2.0. These results are consistent with several national studies on the relationship between various student characteristics and federal student loan default rates. In addition, several of the results of the multiple regression analysis were considered statistically significant. However, these results are not reliable since the assumption of data normality was violated. [The dissertation citations contained here are published with the permission of ProQuest LLC. Further reproduction is prohibited without permission. Copies of dissertations may be obtained by Telephone (800) 1-800-521-0600. Web page: http://bibliotheek.ehb.be:2222/en-US/products/dissertations/individuals.shtml.]