Association for Public Policy Analysis & Management

More students interested in earning a bachelor’s degree are opting to first attend a local community college before transferring to a baccalaureate granting institution (Goldrick-Rab, 2010). Even after completing their lower division courses to transfer to a four-year college, some high performing students who are eligible to attend a selective college enroll at a more broad access institution. While Dowd et al. (2008) calculated the number of community college students who eventually transferred to a selective four year college, they were unable to estimate the number of eligible high performing students. Bastedo and Flaster (2014) argue that estimating the number of students who undermatch is difficult because the United States higher education system lacks a uniform standard for classifying the selectivity of a postsecondary institutions. Moreover, Bastedo and Flaster note that researchers have different ways to operationalize a high performing student with some using a combination of GPA and SAT scores.

In order to find a uniform standard for classifying the selectivity of a university and identifying a high performing student, I bound my research within California and leverage the states’ higher education policies. Under the California Master Plan on Higher Education (Coons et al., 1960), California’s public baccalaureate granting system is composed of the selective University of California institutions (UCs) which includes nine undergraduate serving colleges and the broad access California State University institutions (CSUs) which has twenty-three campuses. The Master Plan stipulates that the UC campuses are responsible for conducting research and conferring doctoral degrees while the CSU campuses focuses on undergraduate teaching. I exploit a program called the Transfer Admission Guarantee (TAG) that exists exclusively between California’s community colleges and the state’s selective public institutions that make up the UC system. The TAG agreement provides a written promise for admission to one of six participating UC campuses as long as students meet a GPA requirement and complete 60 transfer units (Transfer Admission Guarantee).

To estimate student enrollment choices, I drew upon restricted administrative data from the California Community College system (CCC) that includes the population of all enrolled students. The dataset includes student transcripts and individual level descriptors such as GPA and socioeconomic status. The dataset also includes institutions that community college students transferred to. I can identify students eligible to sign a TAG agreement for guaranteed admission to at least one UC campus. I can also estimate TAG eligible students who actually transferred to a UC campus or undermatched and chose to attend a college within the broad access CSU system.

Of the high achieving students eligible to transfer to a UC campus, I find that only 17.7% actually took this offer and enrolled. Moreover, I find that high achieving community college students are more likely to attend a broad access college within the CSU system than a UC campus. I find heterogeneity in the college choices made by high achieving community college students with students of color less likely than their white counterparts to attend a selective college.