Girls Do Better

abstract

In this article we analyse gender gaps in educational outcomes in South Africa using four nationally-representative datasets - Progress in International Reading and Literacy Study (PIRLS), Trends in International Mathematics and Science Study (TIMSS), Southern and Eastern African Consortium for Monitoring Educational Quality (SACMEQ) and Matriculation – over the period 1995-2018. We show that girls outperform boys at the mean in all subjects and all grades, including in Mathematics and Physical Science, as well as in the school-leaving exam, Matric. Pro-girl gaps at the primary-school level are large and statistically significant with Grade 4 girls an entire year ahead of Grade 4 boys in reading outcomes. Similarly, large differences can be found in the most recent (2018) Matric microdata. We show that the received wisdom in South Africa, that males outperform females in Mathematics and Physical Science in Matric, is partly a function of higher rates of male dropout in high school, leaving a stronger cohort of males to write matric. In 2018, for every 100 females in matric there were only 80 males. If one compares an equal number of males and females in matric, girls do unequivocally better in all 13 subjects when looking at average performance. When looking at higher levels of achievement (60%+) girls do better in nine subjects and boys do better in two subjects (Mathematics and Physical Science). However, girls are still less likely to fail these two subjects when compared to their male counterparts. Given these results and the theory that human capital contributes to employment and earnings, we see the need for further research in the South African context for why the female advantage in education does not translate into a female advantage in the labour market. What are the general, and specific, constraints for why women experience inferior labour outcomes in the world of work?

ORCID

Nic Spaull http://orcid.org/0000-0002-2970-5229

Nwabisa Makaluza http://orcid.org/0000-0002-9265-6642

Notes

1 The same individuals are observed over multiple points in time.

2 This is looking specifically at full-time candidates taking at least seven subjects. Of the 512,735 candidates, 284,643 are female (DBE, 2019a:50).

3 These figures are taken from the National Senior Certificate (NSC) Technical Report for each year, published by the Department of Basic Education.

4 The results we report here are the final mark allocated to a candidate for that subject in matric. That is to say it is inclusive of a continuous assessment component. The standards authority Umalusi, together with the Department, have a practice of excluding continuous assessment marks if they are more than ten percentage points different to the final exam mark.

5 The analysis of Matric 2018 is an extended analysis of a previous RESEP project analysing this dataset for Tshikululu Social Investments for the “Maths Challenge” project.

6 To provide an intuitive interpretation of a standard deviation, a 0,5 standard deviation increase is the same as moving from the 50^th percentile of the distribution to the 69^th percentile of the distribution (assuming the distribution is normal).

7 Note that none of the differences in Figure 3 are statistically significant. It is possible that this is because the assessments in 1995, 1999 and 2003 were too difficult for South African Grade 8 students, given that they were aimed at students in high-income contexts. From 2003 onwards South African Grade 9 students wrote the Grade 8 test. The ‘match’ between the difficulty of an assessment and the ability level of the test-takers is one of the attributes contributing to the reliability and validity of the results.

8 For youth aged 18-24 years, those with matric have a narrow unemployment rate that is three percentage points lower than those without matric (looking at 1995-2011 using the October Household Survey 1995-1999, Labour Force Survey 2000-2007, and the Quarterly Labour Force Survey 2008-2011 (Spaull, 2013:38).

9 Other reasons for higher male drop-out rates may include higher job search prospects or enrolments into TVET institutions. Although these options are not investigated in the paper, the question of why males are more likely to drop out poses an interesting question for further research.

10 We also calculate the differences for the other 10 official languages. Looking at the non-comparable group the difference ranges from 3.2 - 4.2 percentage points, and for the comparable group it ranges from 5.2–7.7 percentage points.

11 See Appendix B for the exact number of males and females in each category.

 

Additional information

Notes on contributors

Nic Spaull

Dr Nic Spaull is currently a Senior Research Fellow at the Research on Socioeconomic Policy (RESEP) group at Stellenbosch University in South Africa. He has recently returned from Paris where he was a Thomas J Alexander Fellow at the OECD, and before that a Visiting Scholar in the Graduate School of Education at Stanford University in the United States. Nic has a PhD in economics and has published numerous journal articles on education focusing on assessment, accountability, literacy and education policy in South Africa. Nic is currently working on the Funda Wande project which is developing a certificate to teach Foundation Phase teachers how to teach early grade reading in African languages, funded largely by the Allan Gray Orbis Foundation Endowment. He advises numerous NGO's, policy-makers and grant-making bodies, and also regularly updates his website (nicspaull.com) with new research and articles he finds interesting.

Nwabisa Makaluza

Dr Nwabisa Makaluza is a research associate at the Research on Socio-Economic Policy (ReSEP) at Stellenbosch University where she obtained her PhD in Economics. She is the Head of Research at Funda Wande. The research unit at Funda Wande aims to generate evidence on early grade reading for use by policy makers, academics, and the general public. She has worked on training programs for the Department of Basic Education in South Africa and the Namibian National Planning Committee. She has also written policy briefs for the Growth and Economic Opportunities for Women program funded by the IDRC.