February 28 2013
Reporting Misleading Stats Isn’t “Analysis”
Carrie L Lukas
Over at the Huffington Post, Jillian Berman reports on an “analysis” provided by Ariane Hegewish of the Institute for Women’s Policy Research on the how much time women spend “toiling away for free as their male colleagues get paid.”
Hegewish provided the Huffington Post with this insight:
Since American women workers make on average 77.4 percent of what their male counterparts make in a year, that means they have to work 22.6 percent more days to make as much money as men, Hegewisch wrote in an email to The Huffington Post.
Hegewisch's analysis assumes American employees work 260 days per year (a full 52 weeks without vacation). In that time, women are working about 59 days for free that their male colleagues are getting paid for.
I suppose that’s an impressive bit of multiplication, yet I wish that before she started employing 4th grade math, she had actually analyzed the Department of Labor data behind that 77.4 percent statistic on which her equation rests.
Just a tiny bit of digging would have revealed that the Department of Labor data isn’t actually comparing how much women earn compared to their “male counterparts.” DOL merely reports and compares how much the median full-time working woman makes compared to the median full-time working man. DOL doesn’t control for industry, hours worked, education, year of experience, specialty and many other factors that affect earnings and would really provide an insight into how much two similarly situated coworkers earn.
Hegewisch wouldn’t have had to turn to crazy right-wingers like me for this insight. If she consulted her fellow liberals at the AAAUW she would have found in their analysis of the wage gap that controlling for such factors eliminates the majority of the gap.
Numerous studies have done similar calculations, and estimates of the remaining “unexplained” gap differ – some are as high as 12 percent of the gap remaining unexplained while others find the gap disappears almost entirely (you can read lots more on this in Women's Figures). One should note, however, that just because a gap remains and is unexplained, that doesn’t mean it’s necessarily due to discrimination.
For example, many women who plan to have children make career decisions years before having kids to opt for positions and companies that will be a better fit when they have a family. That’s tough to capture in statistics. Some research also suggests that women are less likely to negotiate salary offers and ask for raises, which could also impact their earnings throughout their lives. That would also explain some of the wage gap and the good news is, that’s something that we can address by encouraging women to be more proactive and teaching our daughters to be comfortable in discussing salaries with employers.
A discussion like this would have been actual analysis. Using misleading statistics to generate eye-popping—but totally bogus—estimates of how much women are being short-changed isn’t.