intersectionality and the dimensional curse

Ok, so this guy on Twitter asks me:

Ben, I’m here to help. So here’s the deal: intersectionality is an approach to inequality in sociology that focuses on how inequality emerges from multiple, overlapping categories. Rather than being a set of specific empirical claims, intersectionality is about how you think about inequality in general. You think about which of your multiple social categories is relevant in a particular institutional environment and how they combine in particular ways.

What is the curse of dimensionality? Roughly, it means that when you have data with lots of dimensions, you get goofy mathematical problems. For example, if you need to solve a problem by considering different combinations of things, then lots of dimensions is bad. Why? The more dimensions, the slower the solution. The wiki has other examples.

The link between sociological intersectionality and the curse of intersectionality? Not much, I’m afraid. In practice, most sociological research on intersectionality may focus on two or three forms of status (race and gender, for example). So you don’t get the sorts of things mentioned in the wiki. If you have a larger number of categories (race, gender, class, religion, nationality), then interaction effects will all be statistically insignificant unless you have a really massive data set. But most people don’t have data to really sort through this anyway, so it is moot.

There’s a family resemblance, but that’s it.


50+ chapters of grad skool advice goodness: Grad Skool Rulz ($4.44 – cheap!!!!)
A theory book you can understand!!! Theory for the Working Sociologist (discount code: ROJAS – 30% off!!)
The rise of Black Studies:  From Black Power to Black Studies 
Did Obama tank the antiwar movement? Party in the Street
Read Contexts Magazine– It’s Awesome!!!


Written by fabiorojas

September 4, 2018 at 4:01 am

Posted in uncategorized

2 Responses

Subscribe to comments with RSS.

  1. Thanks for taking the time to write this. What I am interested in is how adding extra dimensions to intersectionality means, inevitably, that you will have a harder time finding people that match your intersections of identity.

    Intersectionality recognizes unique experiences, which is good, but also makes it harder to build unity in large groups, which feels like an underrated problem that is not addressed. Like your point that more dimensions requires exponentially more data for statistical significance, finding people with similar intersectional backgrounds requires searching through exponentially more people.

    Or: Intersectionality works against unifying societal myths beyond very small groups. I suspect it inherently leads to factionalism.


  2. Ben, I think your comment makes sense, but I don’t think it is very relevant to real experience. Regardless of whether we call it intersectionality or not, people have the experiences they have, and already experience the difficulties in finding people whose experiences and backgrounds match theirs. The creation of the idea of intersectionality didn’t suddenly make people more interested in knowing others who are like them. And it didn’t suddenly make it harder to build unity. Rather, it made some people realize the divisions that already existed and think a little harder about whether their movements were actually attuned to the experiences of the people who might potentially be part of them.



    September 9, 2018 at 1:33 am

Comments are closed.

%d bloggers like this: