Umm... That report does
not say what you think it says.
Whoever did that excerpt was either utterly useless at statistics (and reading comprehension), or was trying to twist the results to try to prove a point.
The report was about harassment on Twitter in general, not GG. They were trying to get a very broad sample from all over Twitter, covering any abuse or harassment that was reported to them.
Since the whole GG mess was ongoing at the time of the research, the researchers decided to use the ggautoblocker list to see how much overlap their was between their sample, and people connected to GG, to find out how much it affected their results.
The result is that there was a fair bit of overlap - about 12% of the people in their sample were also on the ggautoblocker list. Most of the harassment they found was not related to GG.
Their finding: "Reports to WAM! constitute a much wider range of harassment than the GamerGate controversy alone".
That's not what the excerpt you posted claims at all. It interprets that as "only 65 out of 9,844 accounts on the ggautoblocker list engaged in harassment". Which is bunk - the study found nothing at all about the remaining 9,779 accounts on that list. Only that those 65 accounts were in both sample sets.
Here's another way to look at it. The 65 harassers who were also on the ggautoblocker list were responsible for 12% of the harassment on Twitter that these researchers found. So they obviously only noticed 0.6% of GamerGate. Therefore, if they'd found all of the GG-associated harassment and abuse, they'd end up with about with 18 times more harassment reports, of which 95% would have been associated with GamerGate.
That analysis is also complete garbage, but it's just as sound as the analysis in that excerpt you posted, and based on exactly the same faulty reasoning.
Many of the accounts on the ggautoblocker list are either bots, or duplicate accounts created specifically for the purpose of harassing people and discarded when they are no longer useful.