Welcome to CSPI
Welcome to the Center for the Study of Partisanship and Ideology (CSPI)!
Our goal is to help elevate political and social discourse on topics of public interest. While a normal think tank starts with a worldview and seeks to support research from that perspective, our process is different. We look at the intellectual landscape, and ask where the discussion can benefit from insight from the social sciences.
There is no “ideology” to our think tank as conventionally understood. We have supported, or are in the process of supporting, the research of conservatives, liberals, libertarians, and people who are non-political. To the extent that we are driven by an animating concern, it is pushing back against trends that make science and rational discourse more difficult or impossible: namely motivated reasoning, usually driven by concerns of political correctness, career interest or partisanship.
We are also motivated by unease about what is happening in the academy. This goes beyond common complaints about bias or political censorship. Those are certainly problems, but even setting them aside, the entire model of how academia works tends not to be conducive to the accumulation and dispersal of knowledge among a wide audience.
To see why, let’s say you have a theory about why Trump overperformed with Hispanics in 2020 relative to 2016. In the age of the internet, anyone can explain their theory, gather data that would support or refute it, and then put it on a blog or in a Twitter thread. Others can look at the theory and the evidence presented for it, and see if it makes sense.
This is not how it works in academia! If you are a graduate student or professor looking to advance your career, you may begin by coming up with the theory and gathering the data, just like the layperson. But that’s simply the beginning of the process, and the easiest part. If you want to get your argument into a top journal, you must extend your analysis to some 10,000 words, whether they are necessary or not. You need a healthy number of citations, at least several dozen. To know what to cite, you need experience; reviewers will potentially reject your paper for not placing the argument into some broader literature even if others can easily do so and it contributes useful knowledge to the world. Political considerations will play a large role in the citations you use, and the paper will be reverse engineered so that what’s written fits with the works cited. Some things you may consider are
Who are my reviewers going to be? This is very important, as not citing their work may offend them.
Do I myself have previous papers I can cite? This will help get my citation count up.
Are there particularly important people I can cite to add intellectual gravitas to the paper? Is the gender and racial makeup of those cited representative?
Skill at this game means being able to get the right number of citations, along with the right kind, and reverse engineer a text that will make it look like the process is based on some kind of intellectual merit. This will include non-sequiturs and digressions that are irrelevant to the question being considered. In addition to length and citation requirements, you also need to write the paper in “academic speak,” which involves pretending both that your argument is more complicated than it really is, and that it has broader applications than it actually does. For example, if I’m writing a paper on why Hispanics moved towards Trump in normal English, I might begin with the first two sentences.
“Why did Hispanics move towards Trump in the 2020 election? Here, I present data showing that the shift can be tied to the president’s handling of the economy.”
That won’t work in an academic paper! This might be more appropriate for the American Political Science Review.
“Under what conditions do minorities support a culturally dominant political movement? Relying on 2020 precinct- and county-level data, I show that American Hispanics moving towards Trump can be explained in ways that are consistent with the sociotropic model of voting.”
Both these statements are saying the same thing (this is not to imply that this particular argument about Hispanics and Trump is correct, it is just an example!). The first, however, is written in a way that is intended to communicate knowledge, while the second seeks to conceal it to all but a chosen few. Making your argument broader (the paper being about “minorities supporting a culturally dominant political movement”) allows you to add citations from papers on unrelated topics, like say one on indigenous voting patterns in Mexico in the 1980s, though of course in the conclusion you will caution that one cannot apply the lessons of Hispanics and Trump too broadly, and add that more research should be done. It would have been better to simply avoid the broader claim in the first place, but “placing your paper within a literature” helps both to add citations and more words, which is the point of this exercise.
None of this is cost free. In the experience of many academics, the nonsense parts of writing papers–the signalling, the needless complications, the overbroad claims that you then qualify or walk back, finding the right citations and getting them in the right format, and expanding the length of the paper–take more time and energy than the part that actually involves the production or communication of knowledge.
Once you’ve taken your simple argument and made it into an academic work, you can expect it to go through a process by which it will become a published, peer reviewed paper available for public consumption in a year or two. It’ll likely be paywalled at that point, available only to those with a university affiliation or willing to pay some exorbitantly high fee.
Every step in this process creates barriers to entry. Ideally, this means discussion is happening among a community of scholars best situated to judge one another’s work. Often, they’re just playing the citations game, and walling off potential criticisms about the entire value of a particular research agenda.
These are serious problems that go beyond issues of “cancel culture” or “political correctness.” Barriers to entry make it difficult to acquire knowledge, and stop the spread of what is actually learned. To a certain extent, this is unavoidable. The term “sociotropic model of voting”, referring to people voting on the overall state of the economy, is actually useful in order to distinguish it from self-interested voting, in which people vote based on their own economic circumstances. Other aspects of the academic model, like the peer review process itself, are completely unnecessary. It makes absolutely no sense to delay the publication of an important work for up to two years, during which a paper avoids critiques from the larger public, to satisfy the concerns of 2-3 reviewers as likely to be as concerned with citations as the substance of the argument.
Because the academic model works so poorly from the perspective of bringing knowledge to the attention of the wider public, poor analysis from pundits, often lazy or politically motivated, tends to fill the gaps.
Our first report, The National Populist Illusion, takes on several ideas that are popular in political discourse but not supported by data. While the casual observer sees Trumpism as a fundamental break with the Republican past, the report argues that from the perspective of policy and voting behavior we see more continuity than change.
At CSPI, we hope to support new and innovative work, while at the same time being a bridge between what the academy has learned and the culture of the broader educated public. In our highly polarized times, we hope that data and dispassionate analysis can have more of a role to play in political discourse. The effects of this will not be felt on our politics overnight, but in the long run we believe that data-driven research can gain more of a share of the conversation at the expense of more partisan or ideologically-driven analysis.
You can also sign-up to receive updates through Substack.
As a 501(c)(3), all donations to CSPI are tax-deductible. In the coming weeks and months, we are planning to have Zoom events in which our fellows give talks on our research, and also smaller, more intimate gatherings, to which we will invite our top donors. All contributions are greatly appreciated, and will allow us to continue to support the most relevant and interesting social science research being done today.