White Americans’ Reactions to Racial Disparities in COVID-19 — CORRIGENDUM

This corrigendum corrects errors in the published version of my letter, “White Americans’ Reactions to Racial Disparities in COVID-19.” After reviewing my code, I discovered an error in the description of the variable, negative stereotype endorsement. Negative stereotype endorsement is a variable that captures the extent to which respondents endorsed stereotypes of African Americans as less hardworking and less intelligent than white Americans. Negative stereotype endorsement is coded to have three levels: 0 (endorsed neither), 0.5 (endorsed 1 of 2), or 1 (endorsed both). However, there was no error in the code used to create the variable.

I have also revisited the way I incorporated the survey weights for survey-weighted generalized linear models. As a result, there is a minor change to Table 1. The interaction between exposure to the racial disparities information and negative stereotype endorsement for the outcome variable, visit parks without any restrictions is now significant at the 0.10 level for a one-tailed test instead of at the 0.05 level. The significance of the results for all other variables remains the same. I have included the amended version of Table 1.

Table 1. (corrected): The Influence of Racial Disparities Treatment and Negative Stereotype Endorsement on COVID-19 Opinion

Note: ^∗p < 0.10; ^∗∗p < 0.05; ^∗∗∗p < 0.01; one tailed p-values. Entries are logit coefficients, with standard errors in parentheses.

In addition, I have revised Figures 1, 2, and 3 to display 84% confidence intervals, as originally described in the article. Upon review of the code, the original figures inadvertently displayed 85% confidence intervals. The corrected estimates are slightly different, but the pattern of the results and the statistical significance of the results did not change.

Figure 1. (amended): The Effects of Racial Disparities Information (Racially Prejudiced Whites)

Note: These estimates, with 84% confidence intervals, were generated from the models in Table 1.

Figure 2. (amended): The Effects of Racial Disparities Information (Non-Racially Prejudiced Whites)

Note: These estimates, with 84% confidence intervals, were generated from the models in Table 1.

Figure 3. (amended): Difference in Average Treatment Effects: Prejudiced vs. Unprejudiced Whites

Note: These estimates, with 84% confidence intervals, were generated from the models in Table 1.

I have also made updates to the Supplementary Materials, which are in the updated Dataverse. I have amended them to include the unweighted average treatment effects for the interested reader. Finally, I am including a link to the online Pre-Analysis Plan here: https://osf.io/u54ge.

A discussion of deviations from the Pre-Analysis Plan is included in the updated Supplementary Materials, which is in the updated Dataverse.