Effect of Racial/Ethnic Composition on Transfer Rates in Community Colleges: Implications for Policy and Practice

September 2004

Most community college systems in the United States began with a primary mission of promoting transfer education (Dougherty, 1994). However, by the latter half of the 20th century community colleges had evolved into comprehensive institutions, offering a mix of vocational, remedial, adult education, and liberal arts programs (Bailey and Averianova, 1999). As the functions of community colleges expanded, the percentage of enrolled students transferring to a 4-year institution to pursue a bachelor’s degree declined; recent studies estimate the current national transfer rate to be between 20% and 25% (Bryant, 2001; Grubb, 1991).

While there is general agreement that transfer rates have declined, data limitations and controversy over how best to measure transfer rates make it difficult to define precisely the extent or causes of the decline, or to determine the degree to which the decline represents a problem requiring policy intervention. A natural consequence of the expansion of community college mission beyond transfer is a reduction in transfer rates. However, the decline in overall transfer rates is a problem if students face obstacles in meeting their educational goals, if whole classes of the population are underserved or underachieving, or if society’s need for an educated workforce and citizenry goes unfulfilled.

Defining and Measuring the Transfer Rate

Efforts to understand the issue of community college transfer are complicated by the variety of transfer patterns (Townsend, 2001; Townsend and Dever, 1999). In addition to the traditional vertical transfer to 4-year universities, community college students transfer to other community colleges and to private sub-baccalaureate institutions. Students still attending high school or already enrolled in 4-year institutions take courses at community colleges and transfer those units to 2- or 4-year institutions. In addition, there has been an increase in the “reverse transfer” of students who begin their education at a 4-year institution but later transfer to a community college.

Even if we restrict our interest to the issue of transfer from community colleges to 4-year institutions, debate over how to appropriately define the transfer rate causes a disagreement over which students to include in the “base.” The calculation of a transfer rate would seem to be relatively straightforward: the number of students who transfer to a 4-year institution divided by the number of potential transfer students. However, there are many possible specifications of this denominator. For example, should it include: (a) all entering students, (b) only students indicating an intent to transfer, (c) only students enrolled in a degree-granting program, (d) all students completing a specified minimum number of course credits, or (e) some combination of these possibilities? Previous studies have noted great disparities in transfer rates based upon how the denominator is defined. Bradburn and Hurst (2001) reported transfer rates ranging from 25% to 52% depending on how narrowly they defined the denominator. Nevertheless, useful conclusions can be drawn without coming to agreement about the best way to define transfer rate. Whether the absolute rate is in the 20% or 40% range is not the important point. What is important is how transfer outcomes are changing over time and how they differ across sub-groups of students.

Theoretical Constructs and Literature Review

Theories of social and cultural capital offer explanations for differences in transfer rates often found among various racial/ethnic populations. Bourdieu’s (1973) theory of cultural capital posits that social class affects educational attainment through parental education and family expectations, thereby perpetuating educational inequity. Cultural capital derives from the whole set of characteristics, including culture, attitudes and economic opportunities, to which children of a particular ethnic or socioeconomic group are more likely to be exposed (see Borjas, 1992). The notion of social capital extends the influence to include the support available through the wider social network and relationships outside the family. Students of underrepresented minority populations are less likely to have high levels of cultural and social capital due to the lower educational attainment and experience of their parents, other family members, and home communities, leaving them without the information and resources needed to successfully navigate the higher education environment. McDonough’s (1997) conception of organizational habitus may also contribute to an understanding of the lower transfer success of particular racial/ethnic groups. Organizational habitus is a set of perceptions and beliefs held by members of an organization, which in turn shapes the members’ attitudes and expectations. It may be the case that the organizational habitus of community colleges contributes to lower aspirations or performance among students of particular racial/ethnic groups and thus lowers rates of transfer to senior institutions. Many previous studies have attempted to identify the institutional and student factors associated with greater transfer success. Cuseo (1998) reviewed the literature on transfer, and found that institutional factors associated with higher transfer rates include a more “academic” curriculum, higher faculty involvement in transfer issues, more effective institutional research, better articulation with 4-year institutions, and substantial support and advising services for students. Other research points to the importance of a community college exhibiting a strong “transfer culture,” where the goal of transfer is given high priority among the faculty, staff and administration (Cohen and Brawer, 1996; Shaw and London, 1995, 2001). Focusing on student characteristics related to transfer, Grubb (1991) found that transfer rates are higher for males, Caucasians, students of high socioeconomic status, and those scoring higher on high school achievement tests and completing an academic track in high school.

Much of this research has involved case studies or other qualitative analyses of community colleges with high or low transfer rates. However, several studies have used quantitative modeling methods to study the factors associated with transfer success – the approach that we use in this research. Hurst and Bradburn (2001) used data from the Beginning Postsecondary Students Longitudinal Study (BPS) in logistic regression and Hierarchical Linear Modeling to examine student-, state- and institution-level effects on student transfer. They found higher probabilities of transfer for students who were younger, who came from a higher socioeconomic background, and who exhibited greater academic ability and higher expectations for their educational attainment. Enrollment in an academically oriented program and a greater number of credit hours was also positively associated with the likelihood of a student successfully transferring, as was the receipt of financial aid. The emphasis of a college on academic, as opposed to vocational, programs also positively influenced the likelihood of transfer.

Lee and Frank (1990) used regression analysis and data from the High School and Beyond (HS&B) survey to examine factors related to transfer. They found that students who transferred were of a higher social class, were less likely to be minority or female, were more academically oriented and successful in both high school and community college, and were less likely to be working while attending college. Students’ academic behavior and success in community college had the strongest direct impact on transfer, while the influence of ethnicity and social class were more indirect through their relationship to academic preparation in high school.

While Lee and Frank (1990) concluded that minority students have lower transfer rates, other research contradicts this finding. Bailey and Weininger (2002) analyzed institutional data for students in community colleges that are part of the City University of New York (CUNY) system. Their regression analysis included measures of students’ aspirations, socioeconomic status, academic preparation, and alternative commitments (e.g., work, childcare, etc.). The results indicated that African American and Latino students did not have a significantly lower likelihood of transfer than Caucasians after controlling for the other factors, although they were less likely to complete a bachelor’s degree after transfer. The difficulty in pinpointing the relationship between race/ethnicity and transfer rates in quantitative models, holding other factors constant, is very likely due to the high degree of correlation between race/ethnicity and other important causal factors such as socioeconomic status and, in particular, academic preparation. Research conducted by the U.S. Department of Education has found that the most significant predictor of persistence through the baccalaureate degree for all students, including those beginning their postsecondary studies in community colleges, is the degree of academic rigor of their high school curriculum (Adelman, 1999). This research demonstrated that African American and Latino students were significantly less likely to have completed rigorous high school curricula.

Download PDF