Gender differences in teachers' assessments and blind test results – evidence from Uruguay
Resumen:
This paper analyzes the existence of gender bias by public school teachers in Uruguay when grading students in the third and sixth years of primary level. The econometric strategy consists of estimating the effect of gender on the course score (non-blind outcome) when controlling by blind test scores and other relevant characteristics. We do not obtain evidence about a bias in the third year. However, we find an average bias in favor of girls in the sixth year, which responds to biases in the middle of the distribution of abilities (the extreme abilities are not gender-biased when assessed). The average results are robust to several checks. We rule out that sixth-year biasis mainly driven by statistical discrimination or explicit beliefs on talent gender stereotypes.
Este trabajo analiza la presencia de sesgo de género por parte de los docentes de escuelas públicas de Uruguay al calificar a los estudiantes en tercer y sexto año de primaria. La estrategia empírica seguida consiste en estimar el efecto de género en la calificación del curso (resultado no ciego) con las obtenidas en pruebas estandarizadas corregidas por docentes que no observan las características de quienes las realizaron. No encontramos evidencia de la presencia de sesgo en tercer año. Sin embargo, encontramos la existencia de un sesgo promedio a favor de las niñas en sexto año, el cual responde a sesgos más marcados en el centro de distribución de habilidades. Los resultados promedio son robustos a varias comprobaciones. Descartamos que el sesgo en sexto año sea principalmente impulsado por discriminación estadística o creencias explícitas sobre estereotipos de género en el talento de los estudiantes.
2024 | |
Gender differences Discrimination Stereotypes Teacher grading Blind-test Education Diferencias de género Discriminación Estereotipos Calificación docente Pruebas ciegas Educación ECONOMIA |
|
Inglés | |
Universidad de la República | |
COLIBRI | |
https://hdl.handle.net/20.500.12008/43686 | |
Acceso abierto | |
Licencia Creative Commons Atribución - No Comercial - Sin Derivadas (CC - By-NC-ND 4.0) |
Sumario: | This paper analyzes the existence of gender bias by public school teachers in Uruguay when grading students in the third and sixth years of primary level. The econometric strategy consists of estimating the effect of gender on the course score (non-blind outcome) when controlling by blind test scores and other relevant characteristics. We do not obtain evidence about a bias in the third year. However, we find an average bias in favor of girls in the sixth year, which responds to biases in the middle of the distribution of abilities (the extreme abilities are not gender-biased when assessed). The average results are robust to several checks. We rule out that sixth-year biasis mainly driven by statistical discrimination or explicit beliefs on talent gender stereotypes. |
---|