Individuals in the experiment were also asked about their support for allowing the children of undocumented immigrants to remain in the country, both before and after treatment. The responses to these questions are in the variables remain.pre and remain.post. Using a difference-in-differences approach, we want to calcuate the average treatment effect of the treatment on changes in subjects' answers to this question. First, create a variable in each subset that subtracts the remain.pre variable from the remain.post va