Heapmalloc":1ik62sed said:
So they only looked at data from September to May and that's indicative of a trend? Wouldn't it make more sense to take a larger sample of before and after to see if this wasn't an anomaly? Why didn't they look at the sales data for a year before it was shut down and a year after? With such a small sample size I question the validity of their results. Heck it would have made more sense if they compared the same time period before and after...
I'm going to take a stab at answering your question. It all has to do with what their initial question was. If the hypothesis was, "if you take away one of the most popular piracy avenues people will start paying for the content." The longer you go out the more confounding variables you start introducing. I.e. different sites become better known. The raid on dotcom has a lot of... complications... that may turn people against media companies etc. That's just my guess.