Study 1a Data: 

Using data from Facebook, individual-level rows were created to represent unique ad impressions and imputed summary-level information in order to run the logistic regression. It is structured for a 2x2 logistic regression analysis using contrast coding. 

Column C encodes the context condition: Gain = +1, Loss = –1

Column D indicates whether anticipation text was present or absent: Present = +1, Absent = –1

Column E includes a contrast code for the interaction between these two factors. Column E is the product of Column C multiplied by Column D.

The data is prepared for dummy-coded logistic regression, where each row represents a unique particpant view.