Hacker News Reader

I agree, especially that the skew needs to be site dependent. For example the iPhone app store actually has the opposite problem where their rate on delete "feature", skews the average ratings way down. Reviews are heavily distributed along 1 star or 5 star ratings.

Would also be curious to see a comparison of different rating systems, for example i'd like to see what the distribution is on a site thumbs up/down rating site like digg vs a 5 star system.

View on HN

One data point is the netflix rating set from their contest. The system is 1 to 5 stars. The average movie rating was 3.4 in 1998 climbing to 3.8 in 2005.

Back to the original point -- if you used a formula with a default rating of 3 it would underestimate low-frequency movies. People don't view movies at random. They try to pick things they will like, which should be reflected in the statistical model.

Reply on news.ycombinator.com