Intersectional synergies: untangling irreducible effects of intersecting identities via information decomposition

Abstract

The idea of intersectionality has become a frequent topic of discussion both in academic sociology, as well as among popular movements for social justice such as Black Lives Matter, intersectional feminism, and LGBT rights. Intersectionality proposes that an individual's experience of society has aspects that are irreducible to the sum of one's various identities considered individually, but are "greater than the sum of their parts." In this work, we show that the effects of intersectional identities can be statistically observed in empirical data using information theory. We show that, when considering the predictive relationship between various identities categories such as race, sex, and income (as a proxy for class) on outcomes such as health and wellness, robust statistical synergies appear. These synergies show that there are joint-effects of identities on outcomes that are irreducible to any identity considered individually and only appear when specific categories are considered together (for example, there is a large, synergistic effect of race and sex considered jointly on income irreducible to either race or sex).We then show using synthetic data that the current gold-standard method of assessing intersectionalities in data (linear regression with multiplicative interaction coefficients) fails to disambiguate between truly synergistic, greater-than-the-sum-of-their-parts interactions, and redundant interactions. We explore the significance of these two distinct types of interactions in the context of making inferences about intersectional relationships in data and the importance of being able to reliably differentiate the two.Finally, we conclude that information theory, as a model-free framework sensitive to nonlinearities and synergies in data, is a natural method by which to explore the space of higher-order social dynamics.

Revealing the Dynamics of Neural Information Processing with Multivariate Information Decomposition

“…This inequality has been recognized for decades; however, it was not until the seminal work of Williams and Beer [ 1 ] that it was recognized how to algebraically decompose the total joint mutual information into the particular contributions of the parts (and higher-order ensembles of parts). Since its introduction, this framework, termed partial information decomposition , has been widely applied in fields from theoretical and computational neuroscience [ 9 , 21 ] to climate modeling [ 22 ] and sociology [ 23 ].…”