Levels of support for President Trump in 2016 were a major predictor of whether schools opted to reopen for in-person learning this fall, according to a new study from researchers at Boston College and the University of North Texas.
The working paper, published by the Annenberg Institute at Brown University, concludes that “politics, far more than science, shaped school district decision-making.” Analyzing reopening plans for 10,000 public school districts, political scientists Michael T. Hartney and Leslie K. Finger found that the percentage of students attending classes in person this fall was strongly correlated with the share of the vote that Trump received in the surrounding county in 2016.
Notably, the number of coronavirus cases being reported each day by mid-August did not have as strong a correlation with a school district’s reopening plans as did the area’s political leanings, the analysis found. Districts where Trump won only 40 percent of the vote were 17 percent more likely to stick to virtual instruction, compared with districts where he won a 60 percent majority.
The other major predictor was the size of a given school district, with larger districts likely to have stronger teachers’ unions, the researchers say. Even after controlling for other factors, such as the number of coronavirus cases being reported each day, those districts were far more likely to start the school year remotely.
Another factor that correlated with a district’s decision to reopen was the number of Catholic schools in the area. The researchers theorized that public schools would be more likely to offer in-person classes if they risked losing students to parochial schools, which are typically more affordable than other private schools and therefore accessible to a larger number of families.
That hypothesis bore out: The density of private schools in a given area did not correlate with whether schools chose to hold in-person classes, but the density of Catholic schools did have a statistically significant relationship with that decision.