Excellent questions.RobfromtheGulag said:Interesting article. Though in itself I fear it feels a little like troll food.
I wonder if you take into account the number of people who simply spot the troll and keep walking -- if you're taking responses into account you'd certainly want to note that the majority of the people responding are probably unstable in some measure themselves.
1: It's not easy to measure the number of people who ignore the trolls. However, if you own the forums, you can measure the number of people who clicked "quote" or "reply" to the trolling comment and then not post. This gives you a decent sample size of people who ignore the troll.
2: Another thing to consider is that there are some pretty standard commenting percentages. If a site's articles averages around 1000 views and 100 comments, you can measure the effectiveness of the troll's activity using basic analytics compared to a benchmark.
However, if you're doing this kind of analysis on a site you don't have access to the analytics to (which is most often the case) your data is much more subjective and inaccurate. This is the primary reason why trolls are tagged and followed to see if you can identify a pattern.
The key is to identify a behavioral and contextual pattern... one troll-ish comment is just one data-point. This is why I look at the patterns of comments and the types of troll-ish behaviors to research.