To replicate their analysis we will use the BCH_DD_data.dta dataset. This dataset contains the main data to run the difference in differences analysis. Each observation is the average net price of a gas station in a time period. The time periods are before/during/after the spill. The columns/variables in the dataset are: a. random_id. This is a randomized anonymous ID for each gas station b. time. This is a variable indexing time. 0 is the pre-spill period; 1 is the months during the spill; and 2 are the post-spill months. c. during_spill. This is an indicator for the months during the spill (May September 2010)