Field Notes

source

Deconstructing the Gender-Equality Paradox

StatusStableUpdatedStands on2 sourcesSupportsnotes

Ville-Juhani Ilmarinen and Jan-Erik Lönnqvist published this reanalysis in the Journal of Personality and Social Psychology in 2024 (volume 127, issue 1, pages 217–237). The preserved record (raw response) was saved 2026-07-28 from PubMed; the publisher’s full text was not retrievable by the repository downloader.

The target

Nearly the whole gender-equality-paradox literature uses one method: take men’s and women’s mean scores in each country, subtract to get a difference score, and correlate that difference across countries with an index of gender equality. A positive correlation is reported as showing that the sexes differ more where equality is greater.

The objection

The authors argue that a difference-score correlation cannot be interpreted the way the literature interprets it. Very different combinations of underlying means produce identical difference scores, so the statistic cannot distinguish a country where men rose from one where women fell, and it cannot show whether men’s and women’s slopes diverge as a function of gender equality. It is also not an effect size, because its magnitude depends on the correlation between men’s and women’s country means and on those means’ variances, neither of which is about the gender gap.

The finding

Reanalysing three large cross-cultural datasets — fifteen variables spanning attitudes to science and technology, economic preferences, and personality traits, including the data behind the original STEM paper — they decompose the difference-score correlation into five previously unexamined parameters.

The result is that the strong headline predictions were driven primarily by one of them: an extremely strong correlation between men’s and women’s mean levels across countries, between .93 and .97. When two components covary that tightly, the difference between them is dominated by small differences in variance rather than by any substantive divergence, and it is that residual the literature has been reading as a paradox.

Scope of the claim

The paper is a methodological deconstruction, not a demonstration that no sex differences vary across countries. What it establishes is that the standard evidence cannot support the interpretation placed on it, and it proposes decomposition as the replacement. Is the gender-equality paradox a measurement artifact weighs it against the STEM-specific measurement dispute and the reviews that report the phenomenon as real.

Built on 2 sources (2 external).

Working out connections…