We (me+Nida Denson+Nick Bowman) have a new article in AERJ: “Does Socioeconomic Diversity Make a Difference? Examining the Effects of Racial and Socioeconomic Diversity on the Campus Climate for Diversity.”
I originally became interested in this topic writing a paper on class-based affirmative action in grad school, and have since kept an eye out for research on socioeconomic diversity. Most of this work has focused on the (still important) question of whether you can “get” racial diversity in a student body when you only use class preferences/other proxies for social class. There’s also a growing amount of work on the dearth of low-income students in selective institutions. It’s all important, but a few years ago I started wondering–does socioeconomic diversity in and of itself “do” anything? What does it do, does it do the same stuff as racial diversity, can it do it better? (a kind of scandalous question in the higher ed world, but one I think that needs to be addressed if we’re really covering our bases)
To backpedal a little, the reason I was asking these types of questions is because there’s a rich foundation of research on the benefits of racial diversity, whether it’s the effects of a racially diverse student body or students engaging with racial diversity (through things like classes, discussions, intergroup dialogue, interracial friendship, cross-racial interaction, etc). A lot of this work was pioneered by my adviser Mitch Chang and subsequent excellent work has been done by a slew of people, including my collaborators Nida Denson and Nick Bowman. This research has played a big role in arguments made before the Supreme Court in both the Grutter and Fisher cases, etc. In a nutshell, yes racial diversity matters–you need structural diversity (the actual racial diversity in #s) as a necessary pre-condition, and the engagement to see all of the lovely civic and educational benefits.
In contrast, questions re: socioeconomic diversity were focused on the point of entry (i.e., if we encourage socioeconomic diversity in enrollment, can we get racial diversity?) versus what actually happens once students get there.
So I started asking some questions and very quickly became stumped–how do you measure socioeconomic diversity in the first place? The intuitive answer might be–well, measure the % low-income students. But this # doesn’t give us heterogeneity/diversity, it just shows how financially needy a student body is. It’s an important measure, but not the one we were looking for. Enter Nida, my grad school cubicle neighbor and now research fellow at the University of Western Sydney (evidently life is really good for academics in Australia). With any marker of campus-level, demographic diversity you need institutional-level data, which we had in the IPEDS dataset. Nida came up with the brilliant idea of using IPEDS’ measure of “% students receiving financial aid” and recoding it into an upside down U-shaped variable, so institutions with the most financially homogeneous students (either 0% receiving financial aid or 100% receiving financial aid) had the lowest values, and institutions that had a more socioeconomically diverse student body (which we defined as having a more 50/50 split between students on aid versus students not-on-aid) received a higher level (as represented by the top of an upside down U). It’s definitely a limited variable but we did some testing and it was by far the best measure we could come up with given our data.
[In another paper coming out in Research in Higher Education, we found that our upside down U variable is correlated with students’ self-reported cross-class interaction, meaning that students who go to these schools where there’s more of a 50/50 split in the proportion of students who receive aid versus those who don’t are more likely to report interacting with students from different economic backgrounds. So limited as our variable is, we think there’s something there!]
We named our upside down U variable “structural socioeconomic diversity,” borrowing from the term “structural diversity” which is often used to indicate the racial heterogeneity of a student body. (Since we also use this term in our paper, we renamed it “structural racial diversity” for clarification) SSD for short.
Once we figured that out, we also found a way to capture engagement with socioeconomic diversity via a variable we call “cross-class interaction” (“CCI,” borrowing from the term “cross-racial interaction/CRI” which is often used in racial diversity work). Our dependent variables are actually CRI and engagement with diversity activities. So the big picture questions were whether SSD and CCI were predictors of CRI and engagement with diversity activities when other forms of racial diversity (e.g., structural racial diversity) were controlled for. Basically, would socioeconomic diversity (SSD) wipe out the effect of (structural) racial diversity?
Our abstract summarizes the findings—in a nutshell, structural racial diversity still matters and was not wiped out by SSD. However, we found that SSD had an indirect effect on the DV via CCI (cross-class interaction), suggesting that yes, socioeconomic diversity matters too. We saw distinct effects for both forms of diversity, racial and socioeconomic, and while they’re related, they’re not interchangeable.
So in a nutshell, race and class both matter—but not just at the point of entry during admissions, but in the experiences that students have once they actually come to college. It appears that yes, socioeconomic diversity and engagement across class lines can help bolster the overall campus climate for diversity by facilitating engagement with diversity activities and higher levels of cross-racial interaction. However, you still need racial diversity–so it’s a “both-and” and not an “either-or” regarding these two forms of diversity.