21

The White House claims that:

On average, full-time working women earn just 77 cents for every dollar a man earns.

Is it the case?

gerrit
  • 17,636
  • 17
  • 84
  • 137
Quora Feans
  • 2,637
  • 1
  • 19
  • 29
  • possible duplicate of [Is the gender pay gap evidence of discrimination?](http://skeptics.stackexchange.com/questions/5159/is-the-gender-pay-gap-evidence-of-discrimination) – rjzii May 24 '14 at 16:00
  • 1
    This all comes back to the gender pay gap which is highly controversial and the answers are dependent upon what you are asking. – rjzii May 24 '14 at 16:00
  • the simple answer is "yes" -- a good question needs to be more detailed about why women are paid less. – adam.r May 24 '14 at 16:07
  • The web-site makes many claims. To answer this properly, we need to know exactly what claim is being made. – Oddthinking May 24 '14 at 18:05
  • 2
    I've re-opened based on @Articuno's edit, with the proviso that this is a rather uncontroversial variant of the claim - the more contentious issues occur when people try to factor out confounding issues like experience, age, training, hours per week, seniority, etc. – Oddthinking May 25 '14 at 17:58
  • I chose this one because it seemed like it most closely matched the plain-language interpretation of the original question (although more specific). –  May 25 '14 at 18:00
  • @Oddthinking Do you think it might make sense to note the time-frame of this claim? – rjzii May 25 '14 at 19:07
  • Did they mean that a women will get less money for the same job? or that the average salary of women is lower than the average salary of men? – SIMEL May 25 '14 at 20:01
  • 3
    @IlyaMelamed They mean something closer to the latter. The sentence includes no controls for same job or same anything. –  May 25 '14 at 22:02
  • This is not a duplicate. This question asks if there *is* a pay gap (and should be pretty easy to answer plenty of studies on it.) – DJClayworth May 26 '14 at 03:40
  • 8
    I think the question needs to be rephrased, because it doesn't really matter whether the claim is true or not. What the average reader *wants* to know (and will deduct from this statement) is whether a woman is earning the same as a man doing the same job with identical qualifications. The *average* pay is easier to determine, but it does not answer the real question. – Twinkles May 26 '14 at 12:12
  • 2
    @Twinkles Every notable claim deserves examination on this site. [Andre the Giant's hand](http://skeptics.stackexchange.com/questions/20654/is-this-andre-the-giants-hand-holding-a-can-of-beer), and yes, also this simple one. You can always ask a separate question if you have a more nuanced claim examined :) –  May 28 '14 at 05:56
  • See also: http://politics.stackexchange.com/q/2689/2130 – Bobson May 28 '14 at 12:47

2 Answers2

19

It is not as accurate to claim that on average full-time working women earn just 77 cents for every dollar a man earns. A more accurate claim using the proper mathmatical nomenclature would be ...

In 2011, the median full-time year-round* working female age 15+ earned 77 percent of what the median full-time year-round working male age 15+ earned.

*[25] A full-time, year-round worker (page 20 pdf, labeled page 12) is a person who worked 35 or more hours per week (full time) and 50 or more weeks during the previous calendar year (year round). For school personnel, summer vacation is counted as weeks worked if they are scheduled to return to their job in the fall.

The White House website is publishing that statistic that also shows up in their Fifty Years After the Equal Pay Act (FYAtEPA) report.

This progress notwithstanding, in 2011, the average woman still earned only 77 cents for every dollar earned by men.[1]

The source of the claim, according to the FYAtEPA, comes from the Census Bureau

[1] U.S. Census Bureau. “Women's Earnings as a Percentage of Men's Earnings by Race and Hispanic Origin.” Historical Income Tables, Table P-40. (2011). http://www.census.gov/hhes/www/income/data/historical/people/.

Table P-40 notes that the 77.0 percent ratio is a comparison of the median full-time female to median full-time male.

(Based on median earnings of full-time, year-round workers 15 years old and over as of March of the following year. [...])

The median and average(mean) are not the same thing, but sometimes the word average is used when a median value is reported for a statistic, because median values have a central tendency and are less skewed by exceptionally high/low values. The more commonly used "average" (mean) full-time working woman's earnings from the Census Bureau, Table P-37. Full-Time, Year-Round Workers by Mean Income and Sex: 1955 to 2012. Using those figures from 2011, you would determine that a full-time year-round working woman earns on average 72.2 cents (49040 / 67913) for every dollar a full-time year-round working man earns.

There is a wide range of criticism over the 77 cent figure. Some reporters even noted the White House's own pay gap, where the average women earns only 91 cents of what the average man working at the White House earns.

This comparison of median/mean earnings is on a broad level and does not control for many factors that can be significant in explaining earnings differences. One factor is that women on average work fewer hours then men do.

Among full-time workers (that is, those working at a job 35 hours or more per week), men are more likely than women to have a longer workweek. Twenty-six percent of men worked 41 or more hours per week in 2012, compared with 14 percent of women who did so. Women were more likely than men to work 35 to 39 hours per week: 12 percent of women worked those hours in 2012, while 5 percent of men did.

There are more women in occupations with lower-pay, and less women in occupations with higher pay.

While women are more likely than men to work in professional and related occupations, they are more highly represented in the lower-paying jobs within this category. For example, in 2009, professional women were more likely (nearly 70 percent) to work in the relatively low-paying education (with $887 median weekly earnings) and health care ($970 median weekly earnings) occupations, compared to 32 percent of male professionals.

In 2009, only 7 percent of female professionals were employed in the relatively high paying computer ($1,253 median weekly earnings) and engineering fields ($1,266 median weekly earnings), compared to 38 percent of male professionals.

Women on average have less work experience than men (page 37).

on average, women at every educational level and at every age spend fewer weeks in the labor force than do men. The differences between men and women in labor force attachment are much smaller among those with a college degree or more education.2

Malady
  • 165
  • 1
  • 7
user1873
  • 8,931
  • 42
  • 81
  • @Zonata, that is what I am here for. I probably should mention that **median** values are often used in statistics because although they have other drawbacks, they do a good job of removing outliers from a set of values. The White House could use median female/male incomes, or average incomes, but the real problem with the **pay gap** is that it isn't an indication of discrimination which is commonly what it is used indicate. if you want to eliminate the pay gap, you are going to have to pay teachers more or engineers less, make men work less hours or women work more, etc. – user1873 Jun 01 '14 at 21:33
  • @user1873 "*if you want to eliminate the pay gap, you are going to have to pay teachers more or engineers less, make men work less hours or women work more*" - do you have a citation for that? –  Jun 04 '14 at 05:02
  • @user1873 Which of them says that the only solution to the pay gap is to pay teachers more or engineers less or make men work less hours or women work more hours? –  Jun 04 '14 at 05:45
  • Your analysis of median vs mean is misleading. First off, the word "average" can refer to median, mean, or mode. So the median *is* an "actual average". You then present a hypothetical where mean is more representative than the median--an extremely bimodal distribution with very few data points in the middle. But pay distribution is not distributed like this. The gap in mean pay is probably larger than the median because of a small number of men earning very large salaries. (Mean may in fact be better than median in this case, but not for the reasons you provided.) – Kip Jun 04 '14 at 16:18
  • @user1873 If you're using the "etc." to mean "and anything else that might cause the pay gap", then the sentence doesn't actually tell us anything. You're basically saying ""if you want to eliminate the pay gap, you're going to have to remove the causes of the pay gap". Did you have a stronger point than that? –  Jun 04 '14 at 18:26
  • 1
    @Kip, corrected. Those arguments while accurate are not central to my answer. The biggest issue with the 77%/72% figure is that it is a broad statistic that doesn't factor in any of the reasonable explainatioins for the wage gap (hours worked, type of work, work experience, etc.), and tries to imply that this wage gap is an issue that should be addressed – user1873 Jun 04 '14 at 18:28
  • @user1873 Good edit! Regarding "the biggest issue", the 77%/72% figure is just a number and doesn't/can't try to imply anything. *People* might try to use it in arguments, but that isn't what this question is about. It's just *is there a gap of this size?*. If the central point of your answer is not whether women earn 77 cents for every dollar a man earns, then your answer is straying off-topic. –  Jun 04 '14 at 18:31
  • @user1873 *Do you think it is off-topic to explain the difference between the average man/woman?* - Yes, for *this* question. If the further implications that you want to address are in fact notable, then that can be addressed in a separate question/answer. That would be totally on topic on this site. The central point of any answer here is to dig into the evidence behind the claim in the question. It can be a *secondary* point to give a bit of context, and to give *a lot* of detail about another issue, that's best left for a separate question. –  Jun 04 '14 at 20:34
  • I suppose I could remove the details, but change the more accurate claim, "In 2011, the median full-time year-round* working female age 15+ that on average worked fewer hours, had less experience, in careers with lower pay earned 77 percent of what the median full-time year-round male age 15+ earned." That should satisfy everyone. – user1873 Jun 04 '14 at 21:12
  • I feel like the opening sentence could use rewording. You say *it is not as accurate to say* but never answer “as accurate *as what?*” The next answer supplies the answer, but the end of the bolding strongly separates the two. I think delaying the emphasis on *average vs. median* is the primary culprit—read alone, that sentence suggests that the notion of a gap itself is “less than accurate” rather than that the wrong word has been used. – KRyan Feb 05 '18 at 22:24
1

If the wording were "paid" rather than "earned" then the answer is clearly "yes", and while both the left and right agree that there is a gap, there is disagreement on the magnitude, as evidenced by user1873's answer.

However, if the word "earned" remains, statistically, this is unlikely the majority cause. Thomas Sowell outlines the general case here: occupational choice, etc.

The vast plurality of the cause of the gap is experience. This has been detected by many studies over the decades.

There is some residual gap that cannot be explained by conventional causal statistics.

Some are trying to attribute the remaining gap to women's risk aversion relative to men's.

  • 1
    Even if you do insist on attributing the gap to women's risk aversion, assuming that can actually be determined, you don't state why a willingness to take more risk "earns" more money. –  Oct 07 '15 at 12:44