News for hackers. Stuff that matters.
Re:Why the Facebook Experiment Is Lousy Social Science by qu4z-2 on Sunday September 28, 2014 @08:00PM (#8381244)
Regarding your point about the percentages, the OP states that between 10-90% of positive or negative posts are removed, depending on the group. Looking at just the positive posts for a moment, if around 50% of the posts in the news feed are classified as positive (46.8%), surely removing 10% of those is precisely the 5% of posts you claim is different. Note that the percentage removal is based on the uid, so the range of posts removed would be between about 5 and 45%, depending on the poster. The 4.68% quoted is for a user on the extreme low end of the removal range, and is percentage of total posts, not positive posts.

I don't see how this is different from the claimed 10-90% of positive posts. Similarly for negative posts. Am I misunderstanding something?

Re:Why the Facebook Experiment Is Lousy Social Science by gwern on Monday September 29, 2014 @02:38PM (#8384387)
The OP may or may not have misunderstood it depending on how cleverly you parse what he wrote, but he presents it in a way that no reader would be able to understand, compared to the extract from the study which actually does specify the final numbers so one understands that it was a small intervention.