1. Archive

That's a glass window, not a glass ceiling

Like a catching a glimpse of a rainbow refracted through glass, experiencing a glass ceiling may all depend on where you're standing.

A new study by professor Laurie Morgan of the University of Michigan, Institute for Research on Women and Gender, indicates that the glass ceiling may not be as impenetrable as has been widely reported, and may be more like a glass window that's being slowly rolled down.

The glass ceiling is a gender-based barrier to professional advancement. It's the problem of women getting stymied or slowly losing ground in their careers, vis-a-vis their male counterparts. Morgan examined this phenomenon by analyzing two national surveys on the relative earnings of male and female engineers.

Morgan's findings, reported in the August 1998 issue of the American Sociological Review, show that women do suffer an earnings penalty, but it's not due to a glass ceiling phenomenon. Rather it relates to when each woman entered the profession. As the profession adjusted to women in engineering posts, successive entering classes of women experienced less discrimination across the term of their careers. For women who entered the engineering field after 1971, Morgan found that women were compensated on a par with men throughout their career. Equal pay for equal work.

You would think this kind of encouraging research would be touted by women's groups and the government as a great victory.

Not that I could tell.

Morgan's findings weren't mentioned anywhere on the Web site of the National Organization for Women. The non-profit research and women's advocacy group Catalyst, which specializes in glass-ceiling research, never heard of it. And you won't find mention of it on the Web site of the Women's Bureau of the Department of Labor. The site is too busy telling women how they are being shortchanged on the job.

In a DOL pamphlet titled "Worth More Than We Earn: Fair Pay for Working Women," which reads as much like a Cosmo quiz as a government pamphlet, the reader is asked: "Has Any of These Things Ever Happened to You?"

" _ You are a data entry operator getting paid $8,000 less per year than general laborers, who are mostly men. You think your jobs have comparable levels of skill and responsibilities."

(The report never suggests the woman apply for the general laborer job.)

" _ You are a child-care provider for the county government, and you get paid less than liquor store clerks who also work for the county. You think your job requires equal or more skill, effort, and responsibility than selling liquor."

(So, sell liquor.)

The pamphlet goes on to warn of the intractable wage gap between men and women, using a graph to show how it "widens during most of women's working years." Which suggests that as men advance in their fields, women lag further and further behind.

Morgan's research points up a flaw in this kind of statement and much of the glass ceiling research that has been done so far. You don't get a complete picture when you look only at cross-sectional data.

The snapshot of the current landscape doesn't tell you whether things are improving for later generations. For example, the Women's Bureau claim that working women's wage gap worsens over time is overstated. All the bureau can really say with its data is that older women suffer from a more severe gender earnings gap then younger ones.

To understand what's truly going on with the gender earnings gap, you have to compare and contrast multiple cohorts _ people with the same education level, experience, family status and race/ethnicity, who are from the same geographic region, professional specialty, industry and employer size _ over the years.

That's what Morgan did with engineers, nearly 16,000 of them.

She picked that profession because it's a major pathway to corporate and managerial careers and has a reputation for being hostile to women and their advancement. She figured this profession would be a good test for her hypothesis that the glass ceiling is really just a cohort effect that would wither away through attrition of the old guard.

And that's what she found.

"The determining factor in women engineers' earnings penalties is when they started their careers, not the length of their careers," wrote Morgan. She found for women entering engineering after 1971, the effect of being female on earnings "is essentially zero."

If concern over the glass ceiling is to have validity, more research using Morgan's techniques has to be done. If engineering has shaken out its misogyny over time, there's a chance that professions friendlier to women, like medicine and law, might have done so as well.

There's no question that there aren't enough people with XX chromosomes in corporate board rooms or occupying partnership suites, but if Morgan is right, they're on their way.