Saverio Roscigno

Which population segments are more prone to believe in conspiracy theories? In a new study published in Socius: Sociological Research for a Dynamic World, UCI sociology graduate student Saverio Roscigno uses recent survey data to break down conspiracy believers by education, income, race and gender. He finds a noteworthy U-shaped relationship between conspiracy beliefs and socioeconomic status: conspiracy beliefs are common among the least and most educated, the low-income and the high-income, but less common among those who fall in-between. Further, the high-income, most educated group of conspiracy believers is made largely of white men. Below, Roscigno dives into his discovery and how it might guide future work.

What question were you trying to answer when you took on the topic of conspiracies?

At the simplest level, my central question was "Who tends to believe in which conspiracies?" Although there is a rich literature addressing this question in terms of psychological traits and political affinities - things like political extremism, narcissism, or "conspiracy disposition" - there's been comparatively little work focusing on the status attributes often of interest to sociologists (educational attainment, income, race/ethnicity, gender).

Past research has suggested that such beliefs are concentrated among less-educated, lower-income, and African American communities. The dominant explanation for this pattern goes something like this: groups that feel powerless use claims of conspiracy because they offer an explanation for the group's social position and reinforce group boundaries. In other words, groups that feel threatened use these claims because they serve a strategic function.

Just to avoid a common point of confusion, survey research typically defines a conspiracy as a "secret plot by two or more powerful people." So, unlike in much common usage of the "conspiracy theory" label, whether a particular belief is true or false isn't part of the conceptual formula.

What data sources did you turn to for this work? And what did they help you discover?

I analyzed a representative survey dataset used in political science, but in contrast to past analyses, I focused explicitly on status. I chose this dataset because it includes fifteen different conspiracy prompts - to name a few, claims about Jeffrey Epstein being murdered, hidden GMO dangers, and school shootings being perpetrated by the government. The benefit here is that I'm getting a much broader picture than if I had just examined one type of conspiracy belief (like those about COVID-19, for instance). I combined these prompts into a single measure in order to assess these phenomena in broad strokes.

When I looked at this metric across education and income levels, I found something surprising. Rather than a steady decrease as education and income increased, as past scholarship suggested, I found a remarkably persistent U-shape. So conspiracy belief decreased until the bachelor's degree level and then increased dramatically at the graduate degree level. The same goes for income: a steady decrease until the $150k mark and then a large jump. In short, conspiracy beliefs are common among the least and most educated, the low-income and the high-income, but less common among those who fall in-between.

I suspected that race and gender were part of the story, given that white men are overrepresented among the highly educated. I then remade the same graphs but excluded white men, and the U-shape vanished. What this means is that for other groups, the classic understanding seems to hold; there's a slight decrease in conspiracy belief as education and income increase. Yet for white men, it seems to be a different story entirely. White men with graduate degrees are remarkably conspiratorial compared to others, including other white men.

The second finding is that different groups of people tend to hold different beliefs about conspiracies. In other words, not all conspiracy claims are created equal. White male graduate degree holders, for instance, have a distinct affinity for the more taboo claims. 

Why is this work important?

Well, whether or not we like it, claims of conspiracy are all around us. They're on the left and the right, they're a regular topic of conversation online, and they're frequently discussed by politicians. It should be noted that these claims are fundamentally about inequality - or more specifically, about how power works in private - even though some of them are misguided. For this reason, conspiracy cultures offer rich insights into how people are making sense of a shifting and inequitable social world. At their core, these ideas surely raise questions about what is true and false; yet they also make us think about who is trusted, how culture and inequality interrelate, and about the diffusion and production of informal knowledge more generally.

What are the wider implications for this research, and what might be the next steps in this work?

At its core, this article addresses two key issues. First, it emphasizes that the relation between status and conspiracy belief is rather complex. Second, it highlights variation in conspiracy cultures; a conceptual point that had often been mentioned but had yet to be thoroughly examined relative to status.

If not all conspiracy beliefs are created equal, then it means that scholars need to be careful in how they measure and theorize these phenomena. If a survey strictly measures belief in COVID-related conspiracies, for instance, its findings may not apply to belief in the taboo claims mentioned above. At the most practical level, these findings suggest that scholars of the topic need to be explicit in aligning their concepts and theorization with their measures.

One thing I'd love to see is attention to informational access dynamics, in addition to the typical emphasis on attitudes. Even if one has a vague hunch that something is suspect, motivation alone is unlikely to crystallize into a distinct conspiratorial claim. Scholarship on rumor offers a strong launch point in considering these dynamics, as it focuses on the social structures that undergird the production and spread of "unverified information." There's also a burgeoning literature about online information-seeking strategies which I suspect will be relevant in understanding these access dynamics.

In a different vein, some crucial sociological questions might be better suited to qualitative methods. What do these beliefs mean to those who hold them? How do they shape everyday life and action? What are the key lines of variation within the conspiracy milieu, as understood by those within it? Such questions have received minimal attention, yet the few studies that do address these concerns are sources of great inspiration to me. I hope to make a meaningful contribution to this body of work in the coming years.

connect with us


© UC Irvine School of Social Sciences - 3151 Social Sciences Plaza, Irvine, CA 92697-5100 - 949.824.2766