Curious about Eliezer’s claim that his posts were voted up less than others, I did some statistical analysis on the scores of Less Wrong posts. I took the list of all posts as of midnight Mar 28, excluding posts with negative scores (which weren’t available to me), the ten most recent posts (which people haven’t had a chance to vote on yet), and the twenty oldest posts (from when the site was brand-new, and people weren’t around to vote or hadn’t established their criteria for voting), for a total of 93 articles. Of these, 20 consist primarily of a link and quotation, or are otherwise very short. Short articles received much fewer upvotes than full-length articles.
All articles (93): Mean 17.0, Median 14
Short articles (20): Mean 6.9, Median 5
Full length articles (73): Mean 19.8, Median 18
Articles by Eliezer Yudkowsky (21): Mean 17.4, Median 18
Short articles by Eliezer Yudkowsky (5): Mean 3.8, Median 2
Full-length articles by Eliezer Yudkowsky (16): Mean 21.7, Median 20
No short articles by Yvain
Full-length articles by Yvain (18): Mean 28.9, Median 25
Full-length articles by all other authors (39): Mean 14.7, Median 12
Curious about Eliezer’s claim that his posts were voted up less than others, I did some statistical analysis on the scores of Less Wrong posts. I took the list of all posts as of midnight Mar 28, excluding posts with negative scores (which weren’t available to me), the ten most recent posts (which people haven’t had a chance to vote on yet), and the twenty oldest posts (from when the site was brand-new, and people weren’t around to vote or hadn’t established their criteria for voting), for a total of 93 articles. Of these, 20 consist primarily of a link and quotation, or are otherwise very short. Short articles received much fewer upvotes than full-length articles.
The spreadsheet I used is at http://www.jimrandomh.org/misc/LWPosts.xls