Identifying as a ‘STEM person’ makes you more likely to pursue a STEM job – and caregivers may unknowingly shape kids’ self-identity

Employers in science, technology, engineering and mathematics – commonly called the STEM industries – continue to struggle to attract female applicants^[1]. In its 2024 jobs report, the National Science Board found that men outnumber women^[2] almost 3-to-1 in STEM jobs that require at least a bachelor’s degree and over 8-to-1 in STEM jobs that don’t, such as electrical, plumbing or construction work.

Despite women being just as academically prepared^[3] for many STEM roles as men, if not more so, and the fact that STEM jobs offer higher salaries and greater job security^[4] than non-STEM jobs, men continue to dominate this section of the workforce.

I am a social scientist who studies the relationship between education, identity and science^[5], and since 2019, I’ve led the Talking Science research and development group^[6]. One question we’ve sought to answer is why employers continue to struggle recruiting talented women to the STEM workforce.

Our team recently carried out a study where we discovered that how caregivers, especially mothers, talk about STEM topics may significantly shape their children’s interest in STEM careers^[7].

Are you a math person?

As a researcher, whenever I give a public talk I like to ask the audience, “Who here is not a math person?” Without fail, several hands shoot up faster than if I had asked, “Who wants free money?”

It turns out that most people are well aware of their own relationship to STEM fields and may see themselves as a math, science or “STEM” person, or, commonly, not a STEM person. Researchers like me call this kind of self-identification a “STEM identity,” and almost everyone has one^[8]. Although any given person can have a very high STEM identity or a very low one, most individuals fall somewhere in between.

Having a high STEM identity strongly predicts^[9] whether a student will choose to pursue a career in STEM. Research shows that if children don’t develop a high STEM identity by eighth grade, they are unlikely to ever pursue a STEM career^[10].

This finding raises the question: What childhood experiences shape children’s STEM identities?

Individuals come to identify with different groups by recognizing characteristics they share with members^[11] of those groups. In many cases, people learn about the characteristics of a group through direct experience. For example, elementary-age children often see teaching as a female occupation^[12] when they encounter mostly female teachers at their school. Most children, however, never spend enough time with a scientist to form a stereotype directly.

Children learn most of what they know about STEM professionals indirectly through depictions of scientists in their social environment^[13]. Once children have formed a stereotype in their minds, they then compare themselves to these stereotypes to determine whether they are, or could be, a STEM person^[14].

In the United States, five decades of the “draw-a-scientist” studies^[15] reveal that children asked to depict scientists overwhelmingly draw them as male – illustrating a persistent stereotype linking science and masculinity^[16]. While a growing body of research shows that in recent years gender-based stereotypes of STEM workers have decreased significantly^[17], STEM workforce employment patterns contradict this finding.

A missing explanation?

Since social stereotypes about scientists are becoming less gender-biased, our team realized that something else must be causing children to carry male-biased views of STEM into young adulthood. The Talking Science team believed that understanding why some women see themselves as STEM people and want to obtain STEM jobs held the key to understanding the gap between decreasing social stigma and the persistent lack of women in STEM.

To understand this phenomenon more deeply, our team interviewed 20 college students, 13 of whom identified as female. We intentionally selected these students because of their positive STEM identities and enrollment in college STEM programs.

During 60-to-90-minute interviews, we asked participants to list the various people who positively or negatively shaped their academic and professional interests. We then asked students to label each of them as either a “STEM person,” “not a STEM person” or somewhere in between. Finally, we invited each student to explain why they assigned each label.

The students mentioned 102 individuals – including parents, aunts, siblings, friends and teachers – as influential in shaping their STEM identities. Our team then assigned a gender to these individuals based on pronouns and other descriptors the interviewees used.

A gender gap clearly emerged. Women were only about 40% of those described as STEM people and 70% of the individuals described as not STEM people. This latter group almost always included our interviewees’ mothers.