| wowsa |
| ||
|
an article, with title similar to my subject line, recently appeared in the latest volume (april 2012) of "significance" a magazine published jointly by the american statistical association and the royal statistical society. the article tries to make a case that men's and women's track events can be combined, and that a "winner" can be declared in Olympic track events not by who crosses the finish line first, but rather by adjusting for how much better the finisher was than the average for their sex. an example is given for the 2008 3000m steeple, where the male gold medalist won in 8:10 and the woman in 8:59. Then, using some really questionable data to adjust (based on u.s. army observations of how long untrained men and women take to run 3k), the author calculates that the woman running 8:59 would win the gold for the new category of "performance" because her time is farther from the average of untrained women's times than the man was from men's. yikes. i'm surprised this got published! comparing men's and women's times is always going to be rife with controversy, as evidenced by the controversy around olympic qualifying and boston marathon qualifying times. i don't think this paper helps clarify the issues. i can't give a link as i think you have to subscribe to see it. if interested, google it and see if you can get the pdf. i'll be interested to hear what people think. |
| Today's Huck Finn |
| ||
|
This is equivalent to "women's running has less depth than men's running", which we've all known for a long time, and is actually true in every sport. This wasn't the sophistry I was expecting; it's actually even lamer. |
| Today's Huck Finn |
| ||
|
Oh, and trying to compare two different races simply by comparing times shows incredible ignorance as well. |
| wowsa |
| ||
yes, i agree. your point is one of several serious flaws with the proposal given in the article. that it was published in a statistics journal is a bit of an embarrassment to the profession. |
| word from a spinster aunt |
| ||
|
Yes |