Skip to content

Education, maternity leave, and breastfeeding

Abigail Haddad writes:

In today’s column, How We Are Ruining America, David Brooks writes that “Upper-middle-class moms have the means and the maternity leaves to breast-feed their babies at much higher rates than high school-educated moms, and for much longer periods.”

He’s correct about college grads being more likely to have access to maternity leave, but since this (and the cost of child care) translates to college grads being more likely to work and less likely to drop out of the labor force, I think he’s probably incorrect about this translating into ability to breastfeed. I used Census (ACS 2011-2015) data to look at typical hours worked per week in the previous year as well as current employment status among the mothers of infants. They both show the same thing: Education among the mothers of infants is positively correlated with number of hours typically worked per week in the previous year, as well as with currently being employed.

There is a leave gap, but it’s small in absolute terms. If you look at the sum of ‘not working’ + ’employed but on leave’ in the employment status graph (the ’employed but on leave’ group would be part of ’employed but not at work’) and assume that’s the group that’s not being prevented from breastfeeding due to their jobs, it’s the biggest among the less-educated group and the smallest among the most-educated group.

Below are my graphs and here is my workbook, and they’re also here along with the data. It’s possible that ‘hours worked’ is reflecting hours worked pre-baby – it depends on how old the baby is/how the respondent chose to answer. But it’s telling the same story as the employment status graph, and it’s tough to reconcile with Brooks’ story.

I’ve not looked at the details here but wanted to share with you.


  1. Tom says:

    This may be missing the forest for the trees. I thought the hypothesis is the superiority of breast milk over formula, not the act of breast feeding itself. Working mothers can still pump so the data doesn’t address the hypothesis I presented. I could be wrong about what’s claimed.

    • Abigail says:

      I do strongly suspect you would see that, among mothers who are employed, the more educated are more likely to have access to time and space for pumping, just as they’re more likely to have access to longer/paid maternity leave.

      Is that enough to outweigh the significant differences in overall employment? I don’t know, and given that we see substantial educational differences just in the rates of breastfeeding *initiation*, I think that this is barking up wrong tree.

      But, more generally, since Brooks mentioned maternity leave and not pumping, I also do not think he is making that argument.

      • Guido Biele says:

        The PDF you linked reports a positive association between education and breast feed (initiation and after 4 weeks). Are you suggesting that this reverses after 4 weeks, because more of the mothers with higher education have to back to work? This is possible, but to me it looks like the data that includes a breast feeding variable is consistent with a positive association of breastfeeding and education. Or am I missing something?

        • Abigail says:

          No, the first part of the claim about the association is absolutely correct. It’s the attributing it to “hav[ing] the means and the maternity leaves” which I think it’s pretty hard to get from the bigger data picture.

      • Tom says:

        “But, more generally, since Brooks mentioned maternity leave and not pumping, I also do not think he is making that argument.”

        I agree, that’s why I said missing the forest for the trees. Why respond to that particular claim without taking it as a good-faith argument on whether more educated mothers provide breast milk to their children at a higher rate? Your working vs stay-at-home by education makes the point that less educated mothers have more freedom to either breast feed or pump. So if anything the hypothesis would seem to be that the modern workplace makes it possible for all mothers to provide breast milk. In fact, the 3 questions in the linked questionnaire is whether the mother has breast fed or pumped milk so you can’t even identify solely breastfeeding!

        As you remark below, it is also correct to take issue with the “hav[ing] the means and the maternity leaves.” It is more to do with workplace accommodations. You also mention differences in breastfeeding (or pumping) initiation. If older, more educated mothers have access to insurance which pays for expensive breast pumping equipment or they can afford it themselves (and the ancillary equipment for storage, cleaning, etc.), then pumping is more accessible. If poorer mothers can’t afford pumping and dislike breastfeeding then that can explain the difference. Once again, I agree that Brooks’ claim is too specific to make such broad claims, but what about the bigger, and likely more nuanced, picture?

  2. Phil says:

    I assume most people on this blog have read Harry Frankfurt’s “On Bullshit” or at least know its main idea, which is that there is a distinction between lies, errors, bullshit. The important point is that the bullshitter doesn’t care what is true and doesn’t care. This is in contrast to the liar, who knows what is true and wishes to convince you of something else.

    As Andrew has amply documented on this blog, Brooks is a bullshitter. He doesn’t know the prevalence of breastfeeding among educated vs uneducated women (nor does he know the prevalence of pumping, or the average length of time children in different groups are fed breast milk, or anything else relevant). He doesn’t know and he doesn’t care. He needs to churn out columns, so he comes up with something that seems plausible to him and he writes about it. I presume he doesn’t write things he knows to be false, but he certainly doesn’t see any need to fact-check; perhaps he avoids doing so, because if he discovers that he is basing his article on a falsehood he might feel morally obligated not to write it, and then what would he do?

    Thank you, Abigail, for doing yeoman’s (or, rather, yeowoman’s) work to generate some relevant numbers for comparison. It’s a good reminder we _should_ care what is true and what is false. But I’m not sure it’s worth wasting your time with Brooks. You are shoveling against the tide. Brooks is never going to change his ways; his readers don’t seem to care that much of what he says is bullshit; and he is (still) employed by a company that buys ink by the barrel.

    • Andrew says:


      I wouldn’t quite characterize Brooks as a bullshitter; rather, I think he evaluates evidence in a kind of holistic way, detaching himself from the details of any particular argument. So if something sounds right, he considers it true in some deep sense. Recall his response when journalist Sasha Issenberg quizzed him on the notorious $20 dinner at Red Lobster: Brooks felt that his deeper point was correct and was annoyed that Issenberg was focusing on what Brooks considered to be essentially irrelevant details. My reaction was that, if Brooks didn’t think the false details were important, he shouldn’t’ve made them up in the first place. But Brooks presumably considers it his duty to convey the larger truth in a clear and readable manner, and if that has do be done with the occasional false statement, so be it.

      Anyway, I’m feeling charitable toward David Brooks, Michael Barone, and various others whose work I’ve criticized over the years, because their responses have been so civilized and moderate.

      Consider the following range of responses to an outsider pointing out an error in your published work:

      1. Look into the issue and, if you find there really was an error, fix it publicly and thank the person who told you about it.

      2. Look into the issue and, if you find there really was an error, quietly fix it without acknowledging you’ve ever made a mistake.

      3. Look into the issue and, if you find there really was an error, don’t ever acknowledge or fix it, but be careful to avoid this error in your future work.

      4. Avoid looking into the question, ignore the possible error, act as if it had never happened, and keep making the same mistake over and over.

      5. If forced to acknowledge the potential error, actively minimize its importance, perhaps throwing in an “everybody does it” defense.

      6. Attempt to patch the error by misrepresenting what you’ve written, introducing additional errors in an attempt to protect your original claim.

      7. Attack the messenger: attempt to smear the people who pointed out the error in your work, lie about them, and enlist your friends in the attack.

      We could probably add a few more rungs to the latter, but the basic idea is that response 1 is optimal, responses 2 and 3 are unfortunate but understandable, response 4 represents at the very least a lost opportunity for improvement, and responses 5, 6, and 7 increasingly pollute the public discourse.

      David Brooks is a pretty solid 4 on that scale, which isn’t great but in retrospect is like a breath of fresh air, given the 6’s and 7’s we’ve been encountering lately.

      Most of the responses I’ve seen, in academic research and also the news media, have been 1’s. Or, at worst, 2’s and 3’s. From that perspective, Brooks’s stubbornness (his 4 on the above scale) has been frustrating. But it can, and has, been much worse. So I appreciate that, however Brooks handles criticism of his own writing, he does not go on the attack.

      • Jordan Anaya says:

        From personal experience, I’ve found the prospect of hiding errors in my work similar to the proposition of cheating on an academic test.

        When someone came to me with with a problem with OncoLnc, I quickly did 1:

        However, the thought of doing 2 or 3 did cross my mind.

        Similarly, in med school we took unsupervised tests on our computers. Sometimes when I couldn’t remember an answer I would think about looking it up, invoking 5 (everyone probably does it defense).

        Although publicly admitting a mistake and not cheating might appear admirable, there’s another variable to consider.

        The OncoLnc issue wasn’t that big of deal, so I didn’t really mind doing 1. If all the data was complete nonsense then doing 1 would have been much more difficult.

        Similarly, it was easy to pass med school so there wasn’t any incentive (for me) to cheat. If I was on the verge of failing then cheating would have looked much more attractive.

        All I’m saying is when we judge people’s responses to errors, we should keep in mind the magnitude of the mistake and possible incentives to ignore it or trivialize it.

      • Phil says:

        Andrew, I will accept your judgment that Brooks is in category 4, but this doesn’t mean he isn’t a bullshitter by the Frankfurt definition…actually I don’t think it even bears on that. You’re looking at how Brooks behaves after he is caught in an error, which ignores why he made the mistake in the first place.

        The academics you have criticized who have responded in the 5, 6, and even 7 categories are not bullshitters. They care about what is true and what is false; they performed experiments or analyzed data to try to discover it or to try to support their theories. Their experiments may be poor and their analyses may be ill-judged, and they may behave terribly when criticized, but none of these facts suggest that they were indifferent to the truth or falsity of their claims at the time that they made them. Not wanting to accept that you were wrong, or not wanting to admit that you were wrong, are very different from not caring whether you were wrong. Indeed, in some of these cases I think the people in question care too much: it would be so painful for them to admit that they are wrong that they will go to any lengths to avoid admitting it, perhaps even to themselves.

        Brooks, though…I just don’t think he cares whether he’s wrong or right, and nothing you’ve said above seems to challenge that claim. Indeed, your ‘defense’ of Brooks (if I may call it that), that he feels that his ‘larger point’ is correct and that the veracity of the details doesn’t matter…isn’t that another way to say he’s a bullshitter? If someone makes an argument that is based on facts they made up, indifferent as to whether those made-up facts are correct or not, then those made-up facts are bullshit by Frankfurt’s definition. And I think it is fair to say that if someone generates a lot of bullshit, they’re a bullshitter.

        So I don’t think you’re really disputing whether Brooks is a bullshitter. He is. What I think you’re saying is, hey, there are worse things than being a bullshitter. And who could argue with that? Of course there are worse things than being a bullshitter.

        • Keith O'Rourke says:

          > Not wanting to accept that you were wrong, or not wanting to admit that you were wrong, are very different from not caring whether you were wrong.
          It does suggest being seen as right is more important than actually being right. If it is really important to be right(strictly speaking less wrong) then finding out where and how you are wrong should be a positive experience. However, it is a habit that is very hard to develop.

          Geoffrey Hinton raised an interesting case in one of his talks – roughly “the sun would revolve around the earth once, if the earth did not rotate, and this has to be added or subtracted to the 365.25 actual rotations of the earth. I can’t remember which and I don’t care. Knowing it has to be either added or subtracted is enough as I can look it up. Now this really annoys some people to no end.” Is that a 2.5 or 3.5?

  3. zbicyclist says:

    Brooks’ column is basically a quick book review of two polemical 2017 books by Reeves and Currid-Halkett. So it’s likely he’s just pulling sound bites from one book or the other and assembling them into a column. If he wasn’t writing for NYTimes, would anyone care?

    The basic premise is either obvious or silly. It’s obvious because in every civilization in every time, it is likely that people wanted their kids to succeed. In the words of Richard J. Daley, “If a man can’t put his arms around his sons and help them, then what’s the world coming to?”

    It’s silly when we try to tie it to something so specific as breast feeding without, evidently, having real information about the prevalence of breast feeding beyond 4 weeks (if I read Abigail’s link correctly). And, of course, we probably don’t have data over long time periods either, so we can’t say whether or not this is increasing ruination of America.

    • Abigail says:

      The first part of the claim that Brooks makes is correct.

      There’s less-recent data that looks at six and twelve months. It follows the same general pattern, although the non-high-school grad vs. high-school-grad ordering is a bit different, and you can see also that there are differences within the more-than-high-school group which the other CDC piece collapsed. But, yes, I don’t know of survey data that’s both over time and recent. There is a paper on why women breastfeed for shorter durations than they would like to, which could be interesting and speak to this, but I couldn’t find an ungated version. Which is to say, it’s not a topic I object to getting so specific about; there’s a lot out there. I just don’t think this is the way to do it.

  4. Terry says:

    Good analysis Abigail. Straightforward and logical. You cogently show that Brooks’s assertion about breastfeeding has almost no logical or empirical support or coherence. In a word, his assertion is vaporous. (Indeed, the entire column is vaporous, so I agree with the commenters here who say Brooks is a bullshitter because he’s not even trying to say something true and coherent.)

    The more you look at the breastfeeding assertion, the more vaporous it appears.

    <blockquote?Upper-middle-class moms have the means and the maternity leaves to breast-feed their babies at much higher rates than high school-educated moms, and for much longer periods.

    1. Notice how lazy this assertion is. He claims that having money and maternity leave necessarily translates into higher levels of breastfeeding and therefore higher income in adulthood. This is a big, big series of leaps. As Abigail ably points out, he ignores that these moms work more. (It leaves out a lot of other things too. For instance, higher-income mothers have historically avoided breastfeeding because it was seen as lower class. (I don’t know if this is still true.))

    2. By Brooks’s logic, breastfeeding should be highest among unemployed mothers on public assistance, so he should be ecstatic that the children of these mothers will zoom into the upper echelons of society because of this “leg up”. But that would clash with his theme, so it goes unmentioned. (Brooks’s assertion that upper-middle-class moms spend more time with their kids has the same problem.)

    3. There is no demonstration that breastfeeding has a significant effect on a child’s lifetime income. This is reminiscent of the priming literature that dreams of huge benefits from small tweaks.

    4. Brooks’s entire column clashes with recent studies that find that parental involvement has little effect on child development (absent serious neglect).

    5. There is no mention of heritability of psychological traits that are correlated with success. Any discussion of inter-generational income persistence without a discussion of inter-generational heritability is not a serious discussion.

  5. Adam says:

    I find the Brooks article frustrating, not just for the breastfeeding bit, but with how he switches between class and education as it suits him. The two are correlated, but not the same thing. It doesn’t feel innocent either; he’s doing it deliberately to trick the reader.

  6. Dan says:

    Betsey Stevenson and I did some research on this a while ago, and we find evidence consistent with Brooks’ hypothesis. See table 1 here:

    The bottom line is that in 2005-2006, college-educated moms were 27 percentage points more likely to breastfeed at birth an non-college-educated moms, and they were 14 percentage points more likely to breastfeed through 6 months, based on data from the National Health and Nutrition Examination Survey.

Leave a Reply