«

»

Mar 11 2011

Clear gender bias in American Idol voting

If it seems like American Idol always turns into a sausage fest, it's probably because it really does. If you're a female singer, you have only slightly more than a 30% chance of reaching the top 2 under the current rules.

I compiled data from Wikipedia on the gender makeup of remaining finalists on all seasons excluding season 3 (more on that later). My feeling has been that women get hosed, but I didn't really think it was this bad:

Gender makeup of finalists
(note that season 1 had only top 10)

The average top 12 starts out at about 50% men, with slightly more due to season 8 having a top 13, the extra of which was a man (Anoop Desai), plus one extra man from voting in the semifinals. The average number of women for non-season-3 has never exceeded 50% once in the 9 complete seasons that have aired. The only 2 judges saves were both used on men, for Matt Giraud and Mike Lynch, neither of which made it to the top 3.

For all that, female winners (again, not including Fantasia, winner of season 3) constitute 3 of the 8 seasons, or about 37% of the winners, which isn't terrible. This is mostly on the strength, though, of Carrie Underwood, who in season 4 saw 3 out of 6 of the female competitors eliminated in the first 3 weeks; and Jordin Sparks, who saw 3 out of 6 eliminated in the first 4 weeks. Three seasons came down to two men, whereas no (zero) non-season-3 year has come down to two women.

About season 3: that year was rigged. The top 12 had 8 women, due to both Randy and Paula choosing a female from those who were not voted in by the public (Leah LaBelle and Jennifer Hudson[!]).

By the way, seasons 4 thru 6 I considered the best (and were the best rated, excluding season 7), and each had a final between a man and woman.

P.S. Season 8 had eight men, but women were eliminated very quickly. Removing this season from the data set does not appreciably alter the trend.