Understanding the Decision to Enroll in Graduate School: Sex and Racial/Ethnic Group Differences

Abstract

Women continue to receive fewer doctoral and first-professional degrees than men, even though women receive more bachelor's degrees. The underrepresentation of women holds even after allowing for time to complete an advanced degree. For example, women received 55% of the bachelor's degrees that were awarded in 1994-95 but only 44% of the doctoral degrees and 45% of the first-professional degrees that were awarded five years later in 1999-00 (NCES, 2002). (1) African Americans also represented smaller shares of doctoral and first-professional degree recipients in 1999-00 than of bachelor's degree recipients in 1994-95 (5.0% and 6.9% versus 7.5%, NCES, 2002). Hispanics represented a smaller share of doctoral degree recipients (2.9%) but a comparable share of first-professional degrees (4.8%) in 1999-00 than of bachelor's degrees in 1994-95 (4.7%, NCES, 2002). Bowen and Rudenstine (1992) offer several possible explanations for the lower representation of women, African Americans, and Hispanics among doctoral degree recipients than bachelor's degree recipients. First, non-U.S. citizens, the majority of whom are men, account for a higher share of doctoral degree recipients than bachelor's degree recipients. Second, the distribution of bachelor's degree recipients by undergraduate major field varies by sex and race/ethnicity, with women, African Americans, and Hispanics tending to major in fields in which smaller shares of bachelor's degree recipients enroll in doctoral programs, such as communications. Even after taking into account these sources of the gap, however, Bowen and Rudenstine (1992) conclude that women, African Americans, and Hispanics are less likely to receive doctoral degrees. They speculate that both lower rates of doctoral program enrollment and lower rates of persisting to doctoral degree completion play a role. Although researchers have examined sex and racial/ethnic group differences in undergraduate enrollment (e.g., Hurtado, Inkelas, Briggs, & Rhee, 1997; Jackson, 1990; Perna, 2000; St. John & Noell, 1989), few theoretically based, methodologically rigorous studies have explored the sources of observed sex and racial/ethnic group differences in graduate school enrollment using a nationally representative sample of students. Ethington and Smart (1986) examined sex differences in graduate enrollment using a comprehensive causal model but did not include race/ethnicity in the analyses, likely due to a small number of non-White individuals in the sample. Other research has been limited to describing the correlates of graduate enrollment for a particular group, such as Mexican-American female graduate students (Lango, 1995), African-American doctoral recipients in sports and exercise science (King & Chepyator-Thomas, 1996), doctoral students in higher education administration (Poock & Love, 2001), and newly enrolled graduate students at one institution (Malaney, 1987; Stiber, 2000). A variety of conceptual approaches have been utilized to examine graduate school enrollment, including a marketing perspective (Malaney, 1987; Stiber, 2000), Hossler and Gallagher's three-stage college-choice model (Poock & Love, 2001), Astin's theory of involvement (Lango, 1995), and Tinto's model of persistence (Ethington & Smart, 1986). This research contributes to our understanding of the sources of the underrepresentation of women, African Americans, and Hispanics among doctoral and professional degree recipients by examining one phase of the educational pipeline: the enrollment of bachelor's degree recipients in different types of post-baccalaureate educational programs. This study develops and tests a conceptual model that is based on an expanded econometric theoretical framework using data from a nationally representative, longitudinal survey of bachelor's degree recipients. Theoretical Framework Building on research that has examined racial/ethnic group differences in undergraduate enrollment (Perna, 2000), this research develops and tests the appropriateness of an expanded econometric framework for understanding sex and racial/ethnic group differences in post-baccalaureate enrollment. …