The period effect: the effect of menstruation on absenteeism of school girls in Limpopo

Author(s): Chloé van Biljon Cobus Burger

The period effect: the effect of menstruation on absenteeism of school girls in Limpopo

Stellenbosch Working Paper Series No. WP20/2019

Publication date: December 2019


This paper will aim to answer three questions: (1) Are girls absent from school during their periods? (2) If so, how large is the effect of menstruation on absenteeism? (3) Do the effects differ by socio-economic status (SES)? A large body of research examines the barriers that girls face to schooling, yet little is known about menstruation in particular as an obstacle for school attendance. The few existing studies indicate that menstruation does have repercussions for girl’s school attendance. This paper contributes to the literature by using a large provincial dataset to estimate the influence of menstruation on the school attendance of girls in Limpopo. The data, school administration data of the Department of Basic Education (SA-SAMS data), is collected quarterly from schools as part of the Data Driven Districts (DDD) initiative, which resulted from a partnership between the Department of Basic Education and the Michael and Susan Dell Foundation. The DDD programme aims to provide access to high quality, visualised education performance data across the country. Research on Socio-Economic Policy (ReSEP) has been asked to undertake some analysis of this underlying data to illustrate its potential use for research. The SA-SAMS data includes detailed data on absenteeism for most schools in Limpopo, one of South Africa’s poorest provinces. Reasons for absenteeism are not reported, and it is therefore unclear when absenteeism is menstruation-related. In this paper, we develop a structural model to identify whether there are patterns in older girls’ absenteeism that could be explained by menstruation. The model is estimated with maximum likelihood methods and is applied to two control groups: girls before they have reached menarche, and boys. The results of the model are compared across these three groups and by school socio-economic status The results indicate that menstruation causes absenteeism for young girls (12-13 year olds in the poorest 60% of schools and 10-11 year olds in the richest 40%), but that older girls do not have a higher probability of being absent during their menses. These results imply that encountering menstruation for the first time presents challenges for girls in relation to school attendance.

JEL Classification:

B54, C55, I20, J13, J16


school attendance, education, menstruation, South Africa


PDF (2 MB)