Policy Implications

• Author: McNally, Sandra
• Pages: 30-33

30

include stereotypes of the field that are incompatible with the way that man y women

see themselves; negative stereotypes and perceived bias; and few role models for

women. In Cheryan et al.’s (2017) review of the literature, the ‘masculine culture’ of

computer science, engineering and maths is one of three overarching factors that

explain gender bal ance between STEM fields. The other factors are insuffi cient early

experience and l arger gaps in self-efficacy in these sub-fields. However, these other

factors are not independent of this ‘masculine culture’. For example, lack of early

experience does not lead to female under-representation in fields such as psychology

and nursing. According to Cheryan et al. (2017), it is only when a lack of early

experience is present alongside a perceived masculi ne culture that gender differences

are observed.

Cheryan et al. (20 17) draw on a very large range of references in psychology

and education and find that ste reotypes of people in various STEM field s correspond to

current patterns of gender disparities, with the most male-dominated fields being

associated with the most masculine traits. Correl ational evidence suggests that implicit

or automatic associati ons between STEM and males have negative consequences for

women’s science and maths interests and aspirations.

In brief, the stereotype is ‘male, socially awkward and focused on technol ogy’.

Such stereotypes are more promin ent in computer science, engineering and physics

than in biology, chemistry and mathematics and have been shown to correlate with

gender disparities in interests. In another paper, Cheryan et al. (2015) describe

computer science and engineering as stereotyped in modern American culture as male-

orientated fields that involve social isolation, an intense focus on machin ery and inborn

brilliance – al l of which are qualities typically more valued by men t han women.19 The

literature reviewed by Cheryan et al. (2017) suggests that women are less likely than

men to believe they fit these stereotypes and more likely to be deterred when the

stereotypes are salient.

It is very difficult to estimate the independent effect of ‘masculine culture’

because it is a broadly defined concept and difficult to change in an experimental (or

quasi-experimental) context other than specific aspects of this within field

experiments.20 But it is plausible that this culture underlies why the educational context

described in Section 7 matters so much.

9. Policy Implications

What are the policy impli cations arising from what we currently know a bout the

causes of the STEM gap? In this Section, I will reflect on implications from the literature

reviewed above, including ideas put forward in some of the papers reviewed. To facilitate

this, Table 1 summarises findings from some of the studies referred to in S ections 5-7.

19 As the focus of this review is on STEM fields, the emphasis here is on stereotyping within these

fields. However, as discussed by Cheryan et al. (2017) stereotypes that are incompatible with

traditional male gender roles may also help to explain gende r disparities in fields and careers in

which men are underrepresented.

20 For example, psychological studies include small-scale experiments that measure the effect of

encountering a stereotypical computer science student on women’s interest in computer science

(discussed in Cheryan et al., 2015).