🎆Proud to announce the publication of our meta-analysis on race/ethnicity in candidate/conjoint experiments 🎆
📝 with @LizaMugge and Daphne van der Pas, we ask:
Does race/ethnicity impact how voters assess political candidates❓
The results might surprise: we do not find voters have a significant negative bias against racial/ethnic minority politicians. But racial/ethnic minority voters do prefer politicians of their in-group. A thread... 🧵
We stand on the shoulders of giants. Some we will mention in this thread.
First, @aecoppock and Susanne Schwarz already published an amazing meta-analysis on gender in candidate experiments.
https://doi.org/10.1086/716290
We do the same for race/ethnicity.
We re-analyzed the data (N=305632) of all publications from 2012-2022 with candidate experiments randomizing race/ethnicity.
We also synthesize all theoretical frameworks and outline three schools of thought:
1️⃣ Unjust stereotypes - voters are less likely to vote for racial/ethnic minority candidates because of prejudice and discrimination.
We do not find any support for this 👇
2️⃣ Useful stereotypes - in low-information elections voters use shortcuts, cues or heuristics to make meaningful voting decisions.
Our meta-analysis couldn't fully unravel this issue, but my follow-up research did make a start 👇
Much more to come on this topic, stay tuned 😇
3️⃣ Shared identification - do racial/ethnic minority voters prefer to vote for their in-group?
🎉YES! 🎉
We find overwhelming support that this is the case, on average 👇
If the first impressions voters have of political candidates do not explain underrepresentation, what does?
We build on the canonical work of @PippaNorris
and
@jonilo
outlining demand and supply-side explanations for political underrepresentation of minoritized groups.
Our meta-analysis focusses on demand-side explanations and we find no evidence that the general population prefers candidates with majority over minority races/ethnicities.
Does that mean that we point to supply-side explanations?
It is not that simple.
Recent work by Rafaella Dancygier et al. shows that candidate supply also does not explain it:
https://doi.org/10.1111/ajps.12553
And it might be more likely explained by strategic considerations on the part of party selectors or online hate campaigns discouraging the retention of minortized politicians in office – shared identification effects aside.
One major caveat, however, is that very little of this data is based on European research. Stay tuned for research on these questions from Europe. There is much more in the work.
Read this and much more in the article. You can also find all the replication materials in the online appendix.
There are many many people to thank! My co-supervisor @Hakhverdian and also @OrlySiow and @joostvanspanje for giving feedback on early versions of the paper.
THANK YOU!
Even though conducting a meta-analysis is a lot of work, I can really recommend starting your PhD with it. It really helped me get a firm grasp on the literature and master the methods I’m still using. The online guidance and resources by @daniquint also helped a lot!
To get the most comparable results I ploughed through the original datasets of dozens of publications.
I would like to thank all the researchers who shared their replication materials, such as @NickCarnes and many others who are not on Mastodon.
Sharing data in research is not only a cornerstone of #OpenScience, it also enables researchers to build on each other's work, as this meta-analysis is an example of.
Sanne van Oosten