Explaining Cross-country Preparedness and STEM Participation

• Author: McNally, Sandra
• Pages: 20-22

20

Canada but not in Ireland. When nursing is included in the Irish STEM measure, gender

gaps become much smaller. Indeed, within Ireland, there is considerable heterogeneity

between STEM subjects with regard to the importance of educational preparedness for

explaining the gender gap. The sub stantial gender gap in engineering i s mostly

explained by course-taking in upper secondary education and on grades, wh ereas the

substantial gap in technology is mostly not explained by these measures. There is no

gender gap to explain in scienc e. Another interesting finding for Ireland is that there is

a stronger role for comparative advantage at the top of the ability distribution. This has

also been found for the US (Riegle-Crumb et al., 2012).

Boys and girls have also been found to react differently to past performance with

regard to subsequent choices regarding S TEM course-taking and majors (Kahn and

Ginther, 2017). But this tends to depend on subject. For example, in a study about

Israeli high school course choices, Friedman-Sokuler and Justman (2016) found that

girls react more strongly to prior grades in biology and chemistry when making subject

choices. On the other hand, boys react more strongly to grades in computer science and

physics. Both in Ireland and in the UK, boys have been f ound to react more strongly to

comparative advantage in English and maths with regard to their STEM choice than girls

(Delaney and Devereux, 2019a; Aucejo and James, 2016).

It is relevant to consider gender difference in the variance of reading and maths

scores as well as the average score. In an analysis of international test score data,

Machin and Pekkarinen (2008) show higher variance in test scores for boys than for girls

in most OECD countries. In maths, this comes about because boys are more likely to be

found in the upper part of the distribution in maths (i.e. more of them are ‘higher

achievers’ in maths). In reading this comes about because more boys are in the lower

part of the distribution (i.e. more of them are ‘lower achievers’ in reading). To the extent

that ‘higher achievers’ in maths are more likely to study STEM, this could help explain

the gender difference we observe i n who takes STEM subjects in tertiary education

(Kahn and Ginther, 2017).

However, di fferences in prior achievement between girls and boys are not the

whole story. For most S TEM subjects, a significan t share of the gender gap in tertiary-

level educational choices cannot be explained by educational preparedness. For

example, Delaney and Devereux (2019a) find that there is a nine percentage point gap

in the propensity for males and females to choose STEM courses at tertiary level, even

for persons wh o have i dentical preparation at the end of secondary school in terms of

both subjects studied and grades achieved.

4. Explaining Cross-country Preparedness and STEM

Participation

A number of studies relate measures of gend er inequality at a country level to

the gender gap in educational performance and more recently to the probability of

entering STEM at tertiary level. One of the best-known such studies is by Guiso et al.

(2008). They cl assify c ountries ac cording to several measures of g ender i nequality –

including the World Economic Forum’s Gender Gap Index (GGI) which reflects economic

and political opportunities, education and well-being for women. They use data on

educational achievement from PISA 2003 an d correlate the gender gap in performance

with various measures of gender inequality. They find a positive correlation between

gender equali ty and the gender gap in maths. These results suggest that th e gender

gap in maths, though it historically favours boys, disappears i n more gender-equal

societies. Th e sam e cannot be said for how boys score in maths compared with how

boys score in reading. Boys’ scores are always higher in maths than in reading and

although the difference between boys’ maths and reading scores varies across countries,