Albany, Times Union, "Major Inaccuracy Found in Influential Statistic," Albany Times Union,

 

Albany, Times Union, "Major Inaccuracy Found in Influential Statistic," Albany Times Union,
17 May 1996, p. A4

(AP) It was a jaw-dropping statistic widely influential in the movement to change America's
divorce and child-support laws.

Eleven years ago, sociologist Lenore J. Weitzman published "The Divorce Revolution," her
groundbreaking study of California's no-fault divorce system. In it, she reported women's
households suffered a 73 percent drop in their standard of living in the first year after divorce,
while men's households enjoyed a 42 percent rise.

The figures have been quoted hundreds of time in politician's speeches and court rulings.

There's only one problem.

Her figures are wrong.

Richard E. Peterson, a New York sociologist who reanalyzed Weitzman's data from computer
and paper records archived at Radcliffe College's Murray Research Center, found a 27 percent
decline in women's post-divorce standard of living and a 10 percent increase in men's - still a
serious gap, but not the catastrophic one that Weitzman saw.

Weitzman, a professor at George Mason University in Fairfax, Va., acknowledges her figures
were wrong. She blames the loss of her original computer data file, a weighting error or mistake
by a Stanford University research assistant.

But "I'm responsible - I reported it," she says.

Peterson checked Weitzman's conclusions because they were so much at odds with what other
researchers had found and because they conflicted with some of her own data. For several years
after her book's publication, she did not make her data available; she said there were errors in the
master computer data file she wanted to correct first.

American Sociological Review will publish Peterson's research and Weitzman's response in
June.

Publicity around Weitzman's findings was similar to what occurred after a 1986 study that said if
a college-education white woman has not married by age 40, she has a 1-in-100 change of getting
hitched. That figure was discredited, but not before a Newsweek story declared a 40-year-old
woman had a better chance of being killed by a terrorist than getting married.