Girls in STEM: Why the Gap Opens in Secondary School and How Teachers Can Close It

Girls in STEM: Where the Gap Actually Opens, and What the Research Says About Why

A contributing voice from educational research | Specialist in education equity, school improvement, and teacher professional development

Girls and boys show comparable mathematical and scientific ability through primary school, and by Class XI, girls are under-represented in science streams. The research on this gap makes a specific, important distinction worth stating precisely: this is documented as an ability gap that isn’t there combined with a confidence and expectation gap that is. Stereotype threat research, one of the more foundational findings in social psychology, offers a specific mechanism for exactly this pattern, awareness of a negative stereotype about one’s group’s ability in a domain can itself impair performance and shape self-assessment under evaluative conditions, independent of actual capability.

Key facts

• Stereotype threat, a foundational concept in social psychology most associated with researchers Claude Steele and Joshua Aronson, describes how awareness of a negative stereotype about one’s group’s ability in a specific domain can itself impair performance and self-assessment in that domain, particularly under evaluative, high-stakes conditions. This offers a specific, well-established mechanism for why a genuine ability gap can be absent while a participation gap opens.
• Research on teacher and parent expectations has found a documented pattern internationally: adults often hold lower expectations for girls’ mathematical ability even when actual performance between boys and girls is equal, and these differing expectations have been found to shape girls’ own self-assessment of their ability and, downstream, their course and stream choices.
• This creates a precise, important distinction worth stating explicitly: the research supports an ability gap that does not exist at the underlying capability level, alongside a confidence, expectation, and self-efficacy gap that does exist and widens through adolescence, a different, more specific claim than a general “girls fall behind in STEM.”
• The bulk of the specific research cited here, stereotype threat, expectation effects, comes from international, primarily Western contexts, and this piece cannot verify current, India-specific statistics on the Class XI science-stream gender gap or India-specific intervention studies without a working search tool this session.

The gap is specific: not ability, but what surrounds it

If the underlying research is right that mathematical and scientific ability itself doesn’t differ, then the Class XI gap isn’t a capability problem waiting to be fixed with more content instruction. It’s a confidence, expectation, and stereotype-threat problem, which means the intervention that would actually work looks different from more rigorous maths teaching alone.

This distinction changes where a teacher should focus. If the gap were genuinely about ability, the response would be remedial content support. Since the research instead points to confidence, expectation, and stereotype-threat mechanisms, the more relevant response involves how a teacher frames evaluation, whose ability gets visibly doubted or encouraged, and whether girls receive the same expectation of success boys typically do, not additional content drilling.

Five principles for closing this specific gap

1. Diagnose whether a specific girl’s disengagement from science or maths reflects an actual skill gap or a confidence and self-assessment gap. Given the research finding ability comparable through primary school, a teacher noticing declining engagement in a specific student should check self-assessment and confidence before assuming a content deficit.
2. Monitor your own expressed expectations for differential patterns by gender. Since expectation effects are documented even among well-intentioned adults, a teacher genuinely committed to equity benefits from actively checking whether encouragement, difficulty of questions posed, and assumed capability are actually equal in practice, not just in intention.
3. Reduce stereotype-threat-triggering framing around high-stakes evaluation specifically. Stereotype threat research has found that how a test or evaluative situation is framed can itself affect performance; being deliberate about framing, rather than assuming evaluation is neutral, is a specific, research-backed lever.
4. Build sustained relationship and visible confidence-building specifically around STEM self-efficacy, not just content mastery. Since the gap this research identifies is substantially about confidence and expectation rather than capability, relationship-based encouragement targeted specifically at self-efficacy in these subjects is addressing the actual documented mechanism, not a generic motivational add-on.
5. Start by tracking your own classroom’s pattern of who gets called on, who receives harder follow-up questions, and who is expected to struggle, since these micro-level patterns are exactly where documented expectation effects tend to operate, rather than assuming the gap opens somewhere outside your own classroom’s daily practice.

Frequently asked questions

Is it actually true that girls and boys show equal mathematical and scientific ability, or is that an oversimplification? This is well supported at the level the research addresses: through primary school, underlying ability shows minimal or no meaningful gender difference in the broader research base. The gap that opens later is documented as substantially about confidence, expectation, and stereotype-threat mechanisms, not a later-emerging capability difference.

What is “stereotype threat,” and how does it specifically explain this pattern? It’s a foundational social psychology concept describing how awareness of a negative stereotype about one’s group’s ability in a domain can itself impair performance and self-assessment under evaluative conditions, offering a specific mechanism for how a real capability can coexist with declining confidence and participation as stereotype awareness increases with age.

Does this research apply directly to the Indian context, or is it drawn mainly from elsewhere? The foundational mechanisms, stereotype threat and expectation effects, come primarily from international, largely Western research. This piece cannot confirm current, India-specific data on the Class XI science-stream gender gap or India-specific intervention outcomes without a working search tool this session, and a reader should verify Indian-context specifics directly before treating the mechanism as automatically identical in scale or expression here.

What’s the most direct thing a teacher can check in their own classroom this week? Whether encouragement, difficulty of questions posed, and assumed capability are actually equal by gender in daily practice, not just in stated intention, since documented expectation effects tend to operate at exactly this micro level of classroom interaction.

What to do:

1. Name one girl in your class who seems less confident in maths or science than her actual performance would suggest, and write down one specific thing you will do differently for her this week, not a general encouragement, but a named action for a named child.
2. Search DIKSHA or ask your Block Resource Coordinator for any resource specifically on girls’ STEM engagement or stereotype threat in your state context. If nothing exists, note that gap; the absence of a resource is itself useful information.
3. Spend five minutes this week tracking your own pattern of who you call on, and with what level of assumed difficulty, by gender, not teaching, just watching and noting.
4. Tell one colleague about the distinction between an ability gap and a confidence and expectation gap, and discuss whether your school’s current approach to this issue targets the right one.
5. At the end of the week, ask yourself what you would try differently next week, and what you would look for, such as whether a specific student’s self-assessment of her own ability shifts, to know whether it is working.

Note on sourcing: web search was unavailable during this session, so current, specific citations and India-specific statistics could not be verified live. The core research described here, stereotype threat (Steele and Aronson) and documented teacher and parent expectation effects, reflects well-established, broadly known findings in social psychology and education research, though drawn primarily from international contexts. A reader preparing this for publication should verify current Indian-context data and specific citations before treating any specific claim here as directly confirmed for the Indian setting.

Support resources: NIPUN Bharat FLN assessment tools (via your state SCERT or DIET); NCERT teacher guides; Azim Premji Foundation open-access research; DIKSHA platform content in 36 languages; Tele-MANAS mental health helpline: 14416 or 1800-891-4416; CHILDLINE: 1098.