How Education Did (and Did Not) Accentuate Partisan Differences During the Ebola Outbreak of 2014-15

Abstract

Education exacerbates partisan gaps in scientific knowledge and attitudes. However, previous findings about the extent and symmetry of this moderation have been mixed. As a conceptual replication of previous research, this study examines whether education asymmetrically moderates the Democrat-Republican gap in attitudes about Ebola virus disease (EVD) and policies to combat EVD. Weighted data from a survey of 1,461 non-institutionalized adults drawn from a probability-based panel were collected during the 2015 EVD epidemic. The survey measured seven attitudes: fear of personal infection, estimated severity of Ebola, suspicion of exposed Africans, suspicion of exposed Americans, Western government preparedness, support for low-intensity interventions, and support for high-intensity interventions. Knowledge about EVD was also measured. As in prior studies, highly educated Democrats uniquely diverged from other respondents in some attitudes. However, in the other attitudes, there were main party and education effects but no evidence that education was a moderator of partisan differences. Overall, education moderated partisanship when attitudes were affect-laden and targeted toward immediate threats, but not when attitudes were policy-oriented.

Abstract

Background

Research suggests that highly educated left- and right-wing citizens are more politically polarized from one another than their less educated counterparts. In the U.S., attitudes toward climate change match this pattern—college-educated Republicans (right-wing) and Democrats (left-wing) have uniquely strong right- and left-leaning attitudes toward climate change. However, little research has been done on non-environmental topics to confirm that this effect is robust. In addition, previous research is unclear about whether the effect of education is stronger among Democrats or equally strong among Republicans and Democrats.

Why was this study done?

I conducted this study to examine whether the polarizing effect of education, found in earlier studies on climate change, could be replicated on a non-environmental topic. The West African Ebola epidemic of 2014-15 was the non-environmental topic that I examined. This epidemic was politicized by the Republican party in 2014. I also sought to examine whether there was an asymmetrical effect, with a strong contrast on the Democratic (left-wing) side between highly educated and less educated citizens, but a weak contrast on the Republican side between highly educated and less educated citizens.

What did the researchers do and find?

I analyzed data from a survey of 1,461 U.S. residents, gathered during the West African Ebola epidemic. This survey measured seven attitudes: fear of Ebola infection, estimated severity of Ebola, suspicion toward exposed Africans, suspicion toward exposed Americans, estimated Western preparedness for future outbreaks, support for low-intensity interventions and support for high-intensity interventions. The survey also measured knowledge about Ebola. Highly educated Democrats uniquely diverged from less educated Democrats in three attitudes: fear of infection (slightly), suspicion of Africans, and suspicion of Americans—they were less fearful and less suspicious than other subgroups. This was an asymmetrical effect. I found no evidence of an equal and opposite effect on the Republican side. In other attitudes and knowledge, there was little evidence of an interaction between party identification and education.

What do these findings mean?

These findings challenge the idea that education moves both left- and right-wing citizens in opposite directions. The findings suggest that, in some situations, there is a marked contrast between highly educated and less educated Democrats, but not between highly educated and less educated Republicans. (Some prior research reached similar conclusions.) My findings also suggest that polarization through education may be most likely to occur in targeted attitudes, such as suspicion toward persons, but not in attitudes toward policy.