Community college students represent approximately half of all postsecondary students in the United States. The purpose of community colleges is to provide open-door access to higher education for students who otherwise would not have the opportunity. Earning an associate degree or a certificate translates into higher lifetime earnings compared to a high school diploma. Despite lower barriers to entry, retention remains an educational and economic concern. Nationally, only 60 percent of students persist at any two-year institution and only 40 percent earn a credential or degree within six years of enrollment. In the first chapter, I discuss the extensive retention literature. I categorize the literature into five factors and discuss them individually. The factors include ability, aid, institution, instructors, and peers. I review evidence for each factor's impact on student academic outcomes in the higher education setting and direct attention to identification techniques that uncover the causal effect of these factors on academic success. In the second chapter, I use a restricted student-level data set to identify what best predicts passing an English or math course. I implement a nonparametric model averaging technique to identify what covariates are relevant and which combination of models best predicts passing. I find that models focusing on ability and neighborhood characteristics best predict passing an English or a math course. In the third chapter, I examine the effect of math remediation on academic outcomes using the same restricted student-level data set. I find that the effect of math remediation on earning an associate degree or certificate is insignificant on average. My contribution to the literature is twofold: compliers and changing the assignment rule. When conditioning on students who comply with their course placement assignment rule, I find math remediation increases graduation rates by between 3 and 19 percentage points. I find that the current math remediation assignment rule is not optimal. A one-point increase in the math remediation threshold would have no adverse effects on academic outcomes; in fact, increasing the threshold could improve persistence and graduation rates for marginal students. [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.]