Ummm...downmodded for what? If the study doesn't claim the relationship to be causal (downloading causes buying), then its uninteresting as it doesn't support

Ummm...downmodded for what? If the study doesn't claim the relationship to be causal (downloading causes buying), then its uninteresting as it doesn't support anyone's position. There was no indication that the study controlled for an individual's interest level in music (overall amount consumed). So, if we presume the study to simply examine the correlation of two binary variables (downloaded_at_least_1x and bought_at_least_1x), the results could be explained if most of those polled did not download or buy music or did both. Am I thinking about this correctly?