This is not a POWERLIFTING or TRAINING related article,
but it is definitely PrettySTRONG appropriate to
acknowledge that the wage gap is closing! *** Hell yeah! ***
By Shaun Gallagher – FORBES
Gina Trapani, the founder of Lifehacker, has created a site called Narrow the Gapp, which has this headline at the top:
”Women in the U.S. make 81 cents to the dollar men earn doing the same job.”
The 81:100 claim is based on 2010 data from the Bureau of Labor Statistics. Specifically, it compares the median weekly earnings of full-time wage and salary workers over age 16. For men, it’s $824. For women, it’s $669, or 81 percent of $824. But it’s not for doing the same job. That’s the figure for all men and all women, not for men and women in the same industry or job type.
So let’s alter that statement to make it less wrong:
”The median female wage in the U.S. is only 81% of the median male wage.”
Now, this is a downright crude measurement of the wage gap — perhaps the crudest possible, based on the available data.
So let’s calculate a more meaningful estimate, using the same data that Narrow the Gapp relies on. Instead of simply comparing the median paychecks of all men and all women, let’s use a method that allows us to compare like work against like work, and that also takes into account the relative impact of the wage gap on women. If we examine all of the occupations for which a wage gap can be computed based on available BLS data and weight them according to the number of women in each occupation, we find that the weighted median wage gap, by occupation, is about 86.5:100. That’s 5.5 percentage points different than the 81:100 estimate, and it closes up nearly 30 percent of the estimated wage gap. (If we take the weighted mean instead of the weighted median, it puts us at 84.8:100, which is 3.8 percentage points different and closes up 20 percent of the gap.)
And yet the cruder measurement is the one that is almost always used.
Moreover, when this statistic is cited, it is almost always implied that the gap is attributable entirely to gender discrimination. Both Narrow the Gapp and the National Women’s Law Center’s “Women Are Not Worth Less” campaign imply as much.
But common sense alone should suggest that at least some of the gap is attributable to things other than discrimination against women — for instance, differences between men’s and women’s career choices, work experience, salary-negotiation skills, and even the number of hours worked per week.
Yet we don’t have to rely on common sense alone to conclude that not all of the pay gap (based on the BLS numbers) is attributable to gender discrimination.
And what the research shows is that most of the gender pay gap — about 60 percent of it, in fact — can be attributed to factors other than gender discrimination, such as choice of industry, choice of occupation, years of work experience, and union status.
That’s based on 2010 U.S. Senate testimony from Heather Boushey, a senior economist with the Center for American Progress Action Fund, who cites an analysis by labor economists Francine Blau and Lawrence Kahn.
Now, it may be true that seemingly free choices, such as a person’s choice of occupation, might actually be affected by systemic gender discrimination — such as when young women are subtly (or not so subtly) discouraged from pursuing well-paying careers in historically male-dominated industries. But let’s not conflate that with direct workplace discrimination, in which a woman is paid less than comparable male colleagues simply because she is a woman. Both are problems that need to be confronted and addressed, but each deserves its own consideration; the better we can distinguish them, the better we can extinguish them, so it does no good to characterize one as the other.
And by the way: The other 40 percent of the wage gap, the research says, can’t be explained based on the data available … but that doesn’t necessarily mean that all 40 percent is therefore attributable to gender discrimination. It just means we don’t know how to isolate exactly what the other causes are.
But, just for the sake of argument, let’s assume that we can attribute all of that 40 percent to gender discrimination.
If your goal is to highlight that gender discrimination exists in the workplace, then the statistic you should be using is:
“Women workers earn 7.6 cents less per dollar than men because of apparent gender discrimination.”
Those 7.6 cents reflect 40 percent of the 19-percentage-point difference based on the crude estimate. If you’re willing to instead take 40 percent of the more meaningful wage gap measure presented above, you’ll drop it down to 5.4 cents, thus resolving 72 percent of the implied gender-discrimination gap.
The statistic you should not be using is the 81:100 claim, and if you do use that statistic (even though you should not be using it), you should not imply that the gap is entirely or even mostly attributable to gender discrimination. To do so is beyond purposefully misleading — it’s purposefully lying.
I say this as someone who believes that gender discrimination is a real problem in workplaces in the U.S., and who believes that no one should be paid less simply because of their sex.
I say this as someone who supports legislation to prohibit employers from retaliating against workers who divulge their salaries (which can help root out gender discrimination by allowing women to compare salaries with their male colleagues).
In short, I say this as someone who shares the goal of eliminating unjust discrimination against women in the workplace … but who is concerned that the inaccuracies and unfounded inferences made by movements like Narrow the Gapp and the NWLC ultimately undermine the credibility of the cause.
They know that 60 percent of the BLS-derived wage gap is attributable to causes other than direct gender discrimination — heck, the NWLC links to the research on its own site — yet they continue to act as if that 60 percent is a result of workplace discrimination.
If there’s anything I’ve learned as the creator of a site devoted to generating not-quite-unassailable statistics, it’s that you can’t always infer what you might think you can infer from a set of numbers.
Fortunately for everyone, the stats I generate don’t have the potential to shape public policy. We must take the utmost care, however, to ensure that statistics that do have that potential are as credible and as meaningful as possible.
Journalist-turned-programmer Shaun Gallagher is the author of two forthcoming books: “Experiments on Babies: 50 Science Projects You Can Perform on Your Kid” and “Correlated,” based on his site Correlated.org.