Study on glass ceiling is worthless science
© St. Petersburg Times
Benjamin Disraeli was thought to have said there are three kinds of lies: lies, damned lies and statistics. A new report, A New Look Through the Glass Ceiling: Where are the Women?, written by U.S. Reps. John Dingell, D-Michigan, and Carolyn Maloney, D-New York, is the third.
The glass ceiling is a form of sex discrimination where women are prevented from reaching upper-level managerial positions regardless of their professional achievements. Last month, the liberal representatives presented their pack of statistics on the issue at a star-studded press conference featuring Lorraine Bracco, the leggy therapist from The Sopranos. "The glass ceiling has hardened, rather than shattered, since 1995," they announced, pointing to research conducted by the GAO on the status of men and women managers in 10 industries.
"This startling new data shows that in seven of 10 industries studied, the wage gap between male and female managers widened between 1995 and 2000," said Dingell in a press release. "It's compelling evidence that the glass ceiling remains a powerful obstacle to women in the work force, and it suggests things may be getting worse."
Oddly, while Dingell's comments and the title of the report refer to the glass ceiling, much of the report is centered on a perceived growing wage gap between male and female managers and not barriers women face in advancing their careers. Maybe that's because the GAO study provided good news there, concluding that women held 48 percent of all managerial positions across the industries studied -- the same healthy percentage as in 1995.
But encouraging statistics about women's employment aren't worth broadcasting. They don't stoke the victim machine like a screaming title: "Where are the Women?"
Dingell and Maloney's report made media headlines for a day, cementing in people's mind that women are still fighting a losing battle for equality at work. Yet a full reading of the report paints a far murkier picture.
To be able to point to sex discrimination as a cause for unequal pay between full-time male and female employees, you have to control for the non-sexist factors that influence compensation. The study used data from the Bureau of Labor Statistics' Current Population Survey and was able to control for level of education, age, marital status and race. But two of the most important factors in determining earnings: years of experience and one's level of managerial responsibility, were not held constant. So, the study ended up comparing the salaries of CEOs with that of clerical supervisors, and the earnings of managers with 30 years experience to those with five.
The results were worthless as social science but valuable as propaganda. It turned out that women managers in 2000 earned 73 cents for every $1 earned by a man, distinctly less than the figures from 1995 when women earned 86 cents for every man's $1.
But even as Dingell and Maloney denounced this rise in sexism, they had to acknowledge their conclusions were little more than speculation. "Clearly, more detailed research is called for to understand whether and how women's advancement is stalled," admitted the authors. Notice the word "whether."
There were other admissions as well. On combining work and family, the report said women managers are far more likely to interrupt their career by choosing part-time work or leaving the work force for some period to raise a family. This, of course, would have a substantial impact on the relative earnings of full-time male and female managers, but the study did not control for these factors.
Another problem with the study was that the statistics were based on sample sizes so small no valid conclusions could be drawn. The data gleaned for the entire entertainment industry was based on the experience of only 164 people. Similarly, the communications industry's analysis was based on only 186 managers. Neither the GAO nor Dingell and Maloney bothered mentioning this.
Broader studies suggest that earnings differences between full-time male and female managers have actually been shrinking in recent years. A study done by the non-profit Employment Policy Foundation analyzing 59,000 managers (but again not controlling for job experience, tenure or level of responsibility) found that raises for women in management over the past five years were 40 percent larger than those for men. Other studies have found when you compare cohorts and control for lapses in employment, such as for child rearing, the earnings gap between men and women virtually disappears.
To advance their political agenda, Dingell and Maloney used meaningless numbers to make it appear as though rampant on-the-job sexism still exists. The women's movement is clearly fighting to stave off irrelevance and it is not beyond using lies, err, I mean statistics, to do it.
© 2006 • All Rights Reserved • Tampa Bay Times
490 First Avenue South St. Petersburg, FL 33701 727-893-8111
Times columns today
Susan Taylor Martin
From the Times Opinion page