Primary Outcomes (explanation)
The primary outcomes will be analyzed using two approaches. First, individual survey questions will be analyzed independently. Second, we will construct indices for each primary outcome category (beliefs, behaviors, views and policy views) and sub-categories by aggregating related survey items that have been standardized to ensure comparability.
Additionally, we will examine heterogeneity in treatment effects across different ML-determined types, gun-safe ownership (imputed or stated), as well as prior knowledge of Byrna. Furthermore, we plan to explore the heterogeneity of the endorsement treatment effect, particularly with respect to Fox News viewership.
For the MPL question, we will analyze results both for the full sample, as well as for a restricted sample who demonstrate comprehension of the task and consistency in their responses. We will also conduct additional estimations of nonlinear models for the MPL, comparing the likelihood of having a positive willingness to pay (WTP) and performing ordered logit regressions.