Panel Paper: Do Women Influence Other Women to Enter STEM Fields? Evidence from the Case of Female Teachers in Bogota, Colombia

Thursday, November 7, 2019
Plaza Building: Concourse Level, Governor's Square 12 (Sheraton Denver Downtown)

*Names in bold indicate Presenter

Dario Maldonado1, Olga Dulce Salcedo2 and Fabio Sanchez1, (1)Universidad de los Andes, (2)Government of Colombia


Both internationally and in Colombia, there is a salient underrepresentation of women majoring in science, technology, engineering and math (stem). This paper analyses the correlation between the proportion of female teachers in stem fields in high school and the enrollment in stem majors in Bogotá during 2008-2014 by high school graduates in those schools. We first show that, in Bogotá, the gap in probability that male and female high school graduates will enroll in a STEM program is of 30 pp (with women being disadvantaged). This gap is such that female students belonging to the highest decile in math results in the Colombian national standardize result have a probability of enrolling in a STEM higher education program which is lower than the male students in the lowest decile. We then ask whether having female teachers in STEM areas in high school is correlated with a lower gap. Using a linear probability model, we show that an increase of two standard deviations in the proportion of female stem teachers reduces STEM enrollment gaps at the university level by 10.8 percent. We consider an additional hypothesis: that the gender gap and its correlation with gender composition of STEM faculty varies with students’ results on standardized math tests but find no evidence for this hypothesis meaning that the gap in behavior, related to STEM majoring, does not change with high-school math results. For our analysis we use administrative data (N=103092) including all high-school graduates that took the national standardized exam and enrolled in a higher education program between 2008 and 2014. The data allows to know the school in which each student finished high school and some school level variables (including teacher characteristics), the program or major in which the student enrolled and some socio economic characteristics of the individuals.

Full Paper: