2.1. Method 2.1.1. Participants and procedure
We recruited 418 participants (42.4% female;
M
age=29.2%,
SD
=11.0) from Amazon’s Mechanical Turk website (
http://www.mturk.com)tocompletesurveyquestionsonline.Thesamplewas restricted to respondents from the United States. The keyquestions regarding trolling and other online behaviors wereembedded in a larger battery of personality questionnaires. Partic-ipants received monetary compensation ($0.50) for their time
3. Study 2
A limitation of Study 1 is that we asked participants to selecttheirfavoriteactivityfromalist of options. Thisnecessitatedacat-egorical index of trolling that likely underestimated the effects.Hence in Study 2, we assessed enjoyment of each commentingactivity (including trolling) on separate continuous scales. To ruleout the possibility that overall Internet use explains relations withtrolling, we also included a question about total time spent onlinefor use as a control variable.
So it was pretty much inevitable that people who were already likely to be heavy internet users (Based on usage of MTurk) selected trolling as their favorite activity online were likely to go for the humorous responses that lead to "zomg narcissism and sadism" results.
I find the study to be flawed, and the results as useful as Brady's statistics on anything gun-related.
Edit:
Sorry, I read the actual study, not just a summary on a pop-psychology website
http://www.scribd.com/doc/231541019/Buckels-Trapnell-Paulhus-Trolls-Just-Want-to-Have-Fun
Second edit: Please don't let this sway your view of me as a narcissistic, sadistic *******, that's completely unrelated to my internet behavior.