Cross Validated is a question and answer site for people interested in statistics, machine learning, data analysis, data mining, and data visualization. Join them; it only takes a minute:

Sign up
Here's how it works:
  1. Anybody can ask a question
  2. Anybody can answer
  3. The best answers are voted up and rise to the top

The mean is not good in this case, because there are galleries that have an artist with a high rank and several other artists with way lower ranks. I'm thinking about doing a weighted mean, but I don't know how to split the weights to the ranking ranges. Any ideas? Thank you.

share|improve this question
This really depends on what you are trying to get with the rankings and the nature of the artist rankings. I am assuming that there are known "artist ranks", but are these 1, 2, 3 .... (and who's number 1? Da Vinci?) or "great", "near great"... Or what? And do you want to weight for number of paintings? Does having 1 Van Gogh make up for having 100 by notalent noname? And – Peter Flom Oct 18 '12 at 10:36
The ranks are numbers (1, 2, 3 and so on). Number 1 is Andy Warhol actually. I don't have the number of paintings, so they don't weigh in. I don't have to computer the ranks of the artists anyway. I have them already. – elbear Oct 18 '12 at 18:28
up vote 0 down vote accepted

How about simply summing the qualities of the artists, where ranks are probably not the best measure for the quality, but if that's all you have...

Isn't it the case that a gallery with ten soso artists is better than one with five soso artists?

Also think about whether a gallery with ten works of a single artist is better than a gallery with five works of the same artist (if you have access to this information).

share|improve this answer
This is what I ended up doing. Thanks. – elbear Jan 16 '13 at 13:06

Maybe just quote the top one and say how many of their works are there. The quantity of art is quite important isn't it?

Summary statistic of all ranks: if you have outliers at the top you may want to use a robust measure. Look up robust statistics on wikipedia. These include the median and trimmed means, or Winsorization.

share|improve this answer
I don't have a number for their works. I'll look at your other suggestions. – elbear Oct 18 '12 at 18:30

Your Answer


By posting your answer, you agree to the privacy policy and terms of service.

Not the answer you're looking for? Browse other questions tagged or ask your own question.