Social Science Statistics Blog

Posts of measure zero?

The original post still seems to be accessible from posts (on other blogs) that have already linked to it. Don't know if that's a concern.

I think probability of measure zero is now really zero, at least on this site.

OK, but let's just be clear about what seems to be going on here. As far as I can tell there have been at least THREE separate posts by David Kane on the Lancet study on Iraq. Two of the posts were still up at the time that Gary King put up this announcement. By the end of the evening only "one post on this topic had been left up, the only post by Kane that did not seem to provoke commenters to question his methods or arguments.

So the probabiliy of vanishing posts has been established to be non-zero, but are we also beginning to discern somewhat of a pattern to the attrition bias?

Would you be kind enough to explain what you mean by "distracting some folks here from our mission to make the world safe for quantitative methodology". Surely these posts did discuss statistical methodology, even if mostly reminding your guest poster of his ignorance? Didn't someone mention libel? Was that the reason the posts vanished?

David Kane accused the authors of the study of fraud. His only evidence, of course, being that the numbers they got were far higher than they like. Now, in the world of propaganda, that's not only acceptable, but standard practice. However, in academia, one had best have proof. Therefore, his post was yanked.

Hopefully the people who run this program will reconsider the involvement of Mr. Kane with this program.

I'm very late to this, but I'm really surprised you removed the post. The fraud claim was unsupported. But was it so beyond the pale as to be censored? Statisticians must have very, very mild workshops ... hopefullly not because of an imperative not to offend.

The BBC has put up a quite constructive forum for debating the Lancet study's methodology, and what was wrong with it. It covers a lot of material that might have been constructively debated here, but wasn't:

http://news.bbc.co.uk/2/hi/uk_news/6045112.stm