05 August 2013

News you can’t use: deflating a penis size study

Over on Facebook, I became aware of a press release that is making the rounds concerning penis size. It annoys me because it is prurient linkbait disguised as science.

Based on sales of sized condoms, the condom distributor has produced a ranked list of penis sizes for the American states and certain American cities. Oh no, Texas is thirty-fifth, putting the lie to “Everything is bigger in Texas.” (Cue snickering. No, I can’t resist the cheap joke.)

The problems are that there is no actual data anywhere that I can find, and that the rankings are meaningless.

The implication of this press release is that men in some places are better endowed on average than others, and that this is a real difference. But basic statistics says samples vary. I went to Random.org and asked it to generate 1,000 numbers following a normal distribution (bell curve) with a mean of 5.5 and a standard deviation of one. Then, I sampled 100 numbers from that list, ten times, and averaged them. These are the means.

5.648
5.639
5.585
5.583
5.563
5.484
5.466
5.424
5.424
5.374

And there you have a ranking (with one duplicate). But the ranks are meaningless because there is no difference in the population the samples were taken from. The average of the population is 5.5, always.

That the list is probably just noise is shown by the fact that the list of top states and the list of top cites don’t have much in common. The top state, New Hampshire, doesn’t have a single city in the top 10 list (though this may have to do with how they defined “city”). New Orleans is the top ranked city, but the state is listed seventh.

In short, you can’t draw any firm conclusions from this publicity stunt hiding itself as science.

Related posts

Friday Weird Science: Will this spoil the surprise?

Photo by Skakerman on Flickr; used under a Creative Commons license.

No comments:

Post a Comment

Comments are moderated. Real names and pseudonyms are welcome. Anonymous comments are not and will be removed.