The rate of default on college loans in the United States is staggering and poses a major financial threat not only to the students who are unable to repay their student loans but the U.S. economy. For instance, in 2011, borrowers at nonprofit and for-profit 4-year colleges accounted for almost 50% of all federal loan recipients. College students took loans amounting to 106 billion in 2013 via the federal loan program. In effect, the amount funds that students owe different loan programs reached 1.2 trillion dollars with federal student loans accounting for 1 trillion dollars in 2013. The purpose of this quantitative non-experimental correlational and comparative research was to determine the existence, strength, and direction of relationships between loan defaults between graduates from for-profit and nonprofit 4-year colleges by examining the student factors of age, ethnicity, and gender between graduates. The data were analyzed using logistic regression to test hypotheses by selecting and arranging variables, age group, ethnicity, gender, and loan default, chosen from the connected datasets. The population consisted of students from for-profit and nonprofit 4-year college from each of the six regional accrediting agencies as defined by the U.S. Department of Education. The overall fit for this logistic regression model was statistically significant, Wald F (4, 197) = 2.445, p < 0.05. This model explained 1.7% of the variance (Cox-Snell R[superscript 2]) in whether students from for-profit institutions were in loan default. Of the 3 predictor variables, none were significant predictors of loan default status on their own. A positive relationship is seen between age in years and likelihood to default. The equation produced by this regression equation is: y = 0.7609x - 14.368 (where y = default rate; x = age of graduate). Analysis of gender and default rates indicated that females were more likely to default (3.51%) than males (3.05%), in the non-profit institution group, but the difference was not significant. Two different matrices of Pearson correlation coefficients were constructed. The matrices were calculated separately for the for-profit and non-profit institutions using the same dataset. An area that deserves further research is the possibility of loan repayment improvement based on adjusting a number of funds available to the student over the course of their educational career, extending the time frame in which loans may be repaid and experiment with developing a formula that takes into account the earnings of a student after a degree as been conferred based on national market data. However, the realization of the issues that influence loan repayment would be instrumental in helping to design a loan plan that is unique to each student and planned outcome, to encourage payment and ensure the program is sustainable. The disparity in student loan repayment was not by choice and chance but rather by different circumstances. [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.]