It's obviously difficult to discuss the merits of a study that's totally fictional, but from what we were presented in the episode, it seems that the authors did a thorough examination of the data and came up with a statistical correlation between witnessing gun violence and then committing gun violence, which is fine, but then just arbitrarily leapt to the conclusion that the statistical correlation between the two meant that there was a causal relationship between the two.
Novak should have been able to rip the study to shreds just based on that.