I find it interesting how often we get questions about how to analyze data for an experiment that has already been run, as if careful consideration of the proposed analysis* needn't have been part of the original design of the experiment.

*(preferably with the involvement of a knowledgeable statistician who'll be likely to see why the analysis they're maybe thinking about might be a problem before they collect the data)

I see the same in person. I have been approached many, many times (when working on several different campuses at different times in different capacities) by researchers with completed studies, but only once to my recollection by one who hadn't yet collected data (and in that case partly because he'd come to me for advice on something else previously)

hey ssdecontrol

@ssdecontrol good point on that cars question

@whuber congratulations on passing 200 silver badges recently (and 400 bronze a little before that). That's quite an achievement

@Glen_b Thank you. At the rate you are going, you will surpass both thresholds by the end of the year. As you may recall, I pay more attention to indicators of quality than quantity (which is no criticism of you, because you do well in both categories), and badges are one such indicator. Let's continue to encourage everyone to vote generously and wisely (the two are not necessarily in conflict!) so that those who contribute good material can be recognized and rewarded.