For every data set, the seven models in the candidate class are fit to the data; SIC, SICf, SICp; and SICfp are evaluated; and the favored model for each criterion is recorded. Over the one thousand data sets, the selections are tabulated, summarized, and reported.