Thread by @realscientists

Using quantitative research to study groups with intersectional identities has methodological issues

Much of this info will be taken from the article:

"When Black + Lesbian + Woman ≠ Black Lesbian Woman: The Methodological Challenges of Qualitative and Quantitative Intersectionality Research"

I do love that the article starts with an Audre Lorde quote

“... constantly being encouraged to pluck out some aspect of myself and present this as the meaningful whole, eclipsing and denying the other parts of the self”

many statistical models will analyze Black Lesbian Woman as an additive model:

Black + Lesbian + Woman

the method does not match the reality of the identity. Being a Black Lesbian Woman is multiplicative

Black x Lesbian x Woman

"A macro level analysis of economic inequality from an intersectional perspective demonstrates aptly how the social hierarchies of race, sex, and sexual orientation are mutually constructed in the lives of Black lesbians."

"Indeed, most of our statistical methods are implicitly additive, even when testing for interactions."

when statistically significant main effects exist, the probability of finding a significant two-way interaction or higher order interactions (3+ way interactions) decreases

"In an ANOVA for example, interactions are contingent on the size of main effects"

"Thus, when the effects sizes of main effects are large, the probability that no interaction effect will be found is greater"

"One of the foundations of intersectionality research is the premise that multiple factors uniquely combine to define an individual’s experience. For instance, being Black and lesbian confers a unique experience, above and beyond being Black or lesbian."

We need to critically analyze the methods we are using and the assumptions that come with them

We also need to implement more mixed-methods and interdisciplinary research. Across the humanities, social sciences, and STEM

These lines between disciplines are socially constructed and may just do more harm than good tbh