News for hackers. Stuff that matters.
Re:YouTube Comes To A 5-Star Realization: Its Ratings Are Us... by tel on Tuesday September 22, 2009 @05:49PM (#838042)
Though I don't have aggregate data, I don't think Amazon has this issue. People seem able to differentiate within the 2-4 range when there's an actual purchase involved.

Then again, Amazon doesn't account for the statistical uncertainty of votes, so it's a little odd as well.

Re:YouTube Comes To A 5-Star Realization: Its Ratings Are Us... by netsp on Tuesday September 22, 2009 @08:21PM (#838319)
Maybe it's because Amazon asks (and receives) written reviews as well. They make people think a little more. I assume there is still a bias towards the 1 & 5 stars though.
Re:YouTube Comes To A 5-Star Realization: Its Ratings Are Us... by Anon84 on Tuesday September 22, 2009 @09:46PM (#838450)
Not quite. See http://arxiv.org/pdf/0909.0237 Page 9 Table 4.

The closest I've ever seen to the "ideal" curve is the IMDB curve right next to it.

Re:YouTube Comes To A 5-Star Realization: Its Ratings Are Us... by netsp on Wednesday September 23, 2009 @12:13AM (#838585)
I love that on HN you can get as a response to a vague & half-arsed comment a reference and link to an academic paper.

Cheers.

Re:YouTube Comes To A 5-Star Realization: Its Ratings Are Us... by mynameishere on Wednesday September 23, 2009 @11:08AM (#839370)
http://74.125.113.132/search?q=cache:QCWrYt03Y8UJ:sloanwordp...

The econometric results reveal that the reviews for the majority of the products have an asymmetric bimodal distribution.

etc. I thought everyone knew this, and it makes sense. You don't review something if you think it's "okay". You review it if you love it or hate it. Rottentomatoes probably has a bell curve because the professional reviewers review everything, whether they want to or not.

Also, the reviews on youtube are quite useful, and a three-or-less star average indicates that something is labeled incorrectly (for copyrighted material, tv shows, movies, videos, and the like).