Empirical Analysis of Intersectionality

Dubrow, Joshua Kjerulf.  2008.  “How Can We Account for Intersectionality in Quantitative Analysis of Survey Data?  Empirical Illustration of Central and Eastern Europe.”  ASK: Society, Research, Methods 17: 85-102.

To account for intersectionality in quantitative analysis of survey data, I compared and contrasted two approaches.  The unitary approach is a useful starting point to understanding the influence of master demographic categories, but can never lead to discovery of emergent categories.  Of the multiplicative variants of intersectionality theory, the categorical approach is the most user-friendly, as it allows for the greatest array of statistical procedures.  This more relaxed version of intersectionality theory calls for analyzing mainly interaction terms.  To avoid misspecification of the model, the constituent elements of the terms must be included into analysis. 

 
For quantitative analysts wanting to account for intersectionality theory with existing survey data, interaction terms are the best way to measure intersections.  Dichotomous variables are the most straightforward. Because intersections can be quite complex, great care must be taken with the interpretation of main effects and higher and lower order interaction terms (Baumoeller 2004). 

 
Doing intersectionality research with cross-national survey data raises a difficult question: In intersectionality’s search for the effects of “categories of difference” (Hancock 2007: 63-4), what role does cross-national comparison play? As intersections are inseparable from their societal contexts, country level effects should be considered.  I addressed this issue by considering country level processes that generate disadvantaged categories within social groups.  In practical terms, I selected countries based on common political, economic and social characteristics, such as level of democracy, economic development, and historical and cultural heritage.  I also analyzed separate effects for each country on the response variable, with surprising results.  This cross-national study reminds that each intersection has time and space specific consequences.

 
Criticisms of the additive approach fail to give proper credit to quantitative analysis as a practical way to address the impact of disadvantaged identities.  Intersectionality theory rightly advocates the complexity of individuals.  But to understand this complexity, there must be ways to determine which identities are advantaged and which are disadvantaged, in what contexts and to what extent.  As illustrated here, quantitative techniques make possible such accounting that not only allows for valid comparisons across countries, but also among types of very complex intersections.   

 
Thus, although the majority of intersectionality research is done using qualitative methods, intersectionality theorists should embrace quantitative techniques to develop the intersectionality paradigm.  Large survey data sets, especially cross-national ones, provide opportunities for intersectionality researchers to provide empirical support for their theoretical statements and generalizability of their findings.  We need to stop wondering whether quantitative analysis of survey data is appropriate for accounting for intersectionality.  The challenge now is to strengthen the bond between intersectionality theory and quantitative techniques.