Does Gender Composition in a Field of Study Matter? Gender Disparities in College Students’ Academic Self-Concepts

Abstract

Abstract Gendered field-of-study choice is a lively topic of discussion. The explanation usually given for the fact that women are still an exception in typically ‘male’ fields—particularly STEM (Science, Technology, Engineering and Mathematics)—employs domain-specific stereotypes regarding men’s and women’s ‘natural’ abilities in different fields. The central argument of our study is that domain-specific gender stereotypes help explain why few women enter such fields; however, they are not necessarily the driving forces behind the finding that female students who chose typically male subjects have weaker academic self-concepts than their male peers. If it were only domain-specific gender stereotypes that influence students’ perceptions of their abilities, we should find the opposite result in typically female fields of study and no differences in gender-mixed fields. Because existing studies often focus on the male-dominated STEM domain alone, research may have drawn the wrong conclusions. By comparing students in male-dominated, female-dominated, and gender-mixed fields of study, we ask: Does gender composition in the field of study matter for gender disparities in college (university) students’ academic self-concepts? Using data from 10,425 students in the German National Educational Panel Study, our results suggest that it is not only in male-dominated fields of study that women rate their own abilities to be poorer than men rate theirs; the same is true in female-dominated and gender-mixed fields. Therefore, domain-specific gender stereotypes regarding students’ abilities do not (alone) seem to drive gender disparities in STEM students’ perception of their own abilities. No matter what academic field we consider, female students generally exhibit weaker academic self-concepts; however, the gap is most pronounced in male-dominated fields.

Cite this work

No public copy is stored for this record, so only the metadata is cited.

Stored only when both papers are in the corpus — 12 citations exist upstream.

Papers that cite the same references — the relational fallback while embeddings are still backfilling.