According to the CDC (2010), 46% of high school students have had sexual intercourse, and 39% of sexually active boys and girls did not use a condom the last time they had sex.  These rates are higher among African-American high school students. The consequences of these behaviors on future health and economic outcomes can be devastating (Hoffman 2008). Recent evidence suggests that the percent of African-American teachers in a high school improves aggregate student reproductive health measures (Atkins and Wilkins 2011).  In addition, minority teachers have been shown to improve educational outcomes among minority students (Dee 2004).  However, due to data limitations, scholars have only been able to speculate about the causal mechanisms underlying these findings.

Economic theory would suggest that higher expectations for future earnings should decrease risky behaviors because the opportunity costs of the behaviors rise.  Consistent with this theory, empirical evidence shows that higher expectations about college attendance reduce the number of sexual partners among adolescents (Cowan 2011). Thus, if minority teachers raise the expectations of minority students by serving as role models, we would expect to see both better educational outcomes and less risky sexual behaviors.

Contributing to this body of knowledge, our study examines the effect of African-American teachers on the sexual activity of high school students and the probability of a teen pregnancy, particularly among African-American students.  Unlike previous studies in this area, we use individual-level observations on a national sample.  Moreover, we are able to test whether the presence of minority teachers increases students’ expectations for the future and the degree to which expectations can explain the relationship between minority teachers and students’ risky behaviors.

For our analyses, we use data from the National Longitudinal Study of Adolescent Health (Add Health). The school-level structure of the Add Health data provide an excellent opportunity for examining these relationships as it includes both organizational data from the school and individual-level student characteristics and expectations.  Data are available for a rich set of control variables to help account for individual, family, and school-level characteristics.

Our regression models use each school’s percentage of African-American teachers and individual-level data on two binary outcomes and one likert scale outcome:  sexual activity in the last 3 months for girls (n=7091) and boys (n=6744), pregnancy that ended before age 20 for girls (n=7019), and a 8-point scale of students’ subjective expectation of graduating college.  The expectation scale ranges from “No chance” to “It will happen”.  There are 132 schools in our sample, which have between 0 and 100 percent African-American teachers (the mean is 12 percent). Because theory suggests that a critical mass of minority bureaucrats may be needed to influence outcomes (Meier 1993), we allow for a non-linear relationship between the percent of African-American teachers and outcomes.

Our preliminary results suggest that the percent of African-American teachers reduces the probability of sexual activity and increases college expectations, especially among African-American students.