Ceci and Williams' new paper
Dear friends,
In Amy Bug's posting and the discussion that followed, she brought up the fact that the National Academy of Sciences report, Beyond Bias and Barriers, has been criticized, most notably by Ceci and Williams at Cornell. They have a new paper just released in Proceedings of the National Academy of Sciences in which they again say it's not really discrimination at any step they have been able to detect in an exhaustive analysis (two books worth and more). But it's choices women make, many of which are rooted in children and child care issues. The conflict of fertility peak and tenure stress peak. And the lack of affordable on site day care in most academic settings (have you checked out the waiting list at your local facility, if you have one?)
Ceci and Williams argue that it's not the bell curve of intelligence, as John Tierney of The New York Times would like us to focus upon, but it's the family-unfriendly atmosphere that prevents women in science's success. I think they have good evidence, and yet I have experienced anti-woman bias myself and seen it affect younger women recently, fighting it without success. So I am not willing to bury bias at the crossroads yet.
But I think their emphasis on child care and child friendly policies is good and I want to emphasize how important that is. It is a part of the subtle message, what you need does not get to the top of our priority list.
Cheers,
Laura
In the years of my career when I felt the most at risk, I was unmarried. I think it is convenient to point to child care as THE issue, but it's not at all. But by pointing at that, we can conveniently ignore the often-subtle but still distinctly present other forms of bias. They aren't as uniform or obvious as child care and family, so harder to pinpoint.
For example, we know that academic woman are less likely to be parents. Then those academic women who are not parents, should not show any evidence for disadvantage in salary, awards, labspace, teaching, etc. Try telling THAT to Nancy HOpkins.
I've been struggling with how to interpret this paper (as I think most commentators have, even the most enthusiastic advocates). But these comments acknowledge the general critiques, which is good.
Laura, regarding the prevalence of democrats in the behavioural sciences, you may find my most recent post of interest: http://www.nature.com/scitable/blog/cognoculture
Hi friends,
Several of you have emailed me to complain that I was "too easy" on this article. Ceci and Williams have done a meta analysis, i.e. they didn't do any of the studies, just reanalyzed the data from others' work. So the main conclusions are based on the 'weight of the evidence' and particularly strongly based on the 2009 report of the Committee on Gender Differences in the Careers of Science, Engineering, and Mathematics Faculty (NAS Press, Washington, DC). This report focused on tenure track faculty members at the most productive R1 universities, so it's not a random sample of all university faculty experiences by any means. And it tested for transitional events rather than feelings and attitudes, so that any discrimination people felt behind the scenes was pretty much not taken into account. That such things might contribute to a hostile atmosphere did not seem to occur to the authors. As I mentioned in my original posting, they didn't ask whether what women need is likely to get to the top of the priority list at their universities. Thus, their over-emphasis on women's choice, when what was offered to choose between was not optimal for women.
cheers,
Laura
Yes, H, the Summers proposal is still circulating out there and it's as hard to kills as a hydra. John Tierney is taken with the greater variation in men's intelligence, but never asks, is there any evidence that the male full profs in physics and math have intelligences that are in the high fringe of those measured? Very problematic analysis he provides. And he must really believe it because he throws it in even when not so relevant, e.g. in his recent article about the prevalence of Democrats among Social Scientists.
Cheers,
Laura
For heavens sakes, don't dump this Stereotype Threat stuff on us, about "the bell curve of intelligence". Exposure to unproven statements like this decrease the performance of underrepresented groups!
One of Ceci and Williams main points was that hiring and P&T data (for math-intensive science fields) don't support the widely held view about bias as a hindrance to women. They recommend that structural and institutional bias is likely a much more potent force--it's obviously much harder to measure than the hiring/P&T data. Here are two sources that characterize the study best:
NatureNews: http://www.nature.com/news/2011/110207/full/470153a.html
The Guardian: http://www.guardian.co.uk/science/blog/2011/feb/08/sexual-discrimination-women-scientists-institutional
Childcare is something easy to pinpoint in a survey. And it also the easiest to remedy. Having affordable, high-quality childcare for women in science from the graduate level on would be a great help.
BUT I think the issue is more than getting good childcare. It is more a question of work/life balance once you have those children and need to juggle a demanding academic position with being the kind of parent you wish to be.
It is not a coincidence that women with children in academe fair worse than single women. The fact that men with a spouse do BETTER than those without sums up the issue. Men in academe thrive with a wife and even children. Women do not. And that is what we have to remedy to make the job attractive to more women.
I think Melissa has a good point. If we could prove that African Americans always got jobs and were promoted in spite of put downs and racist remarks, no one would be saying we should ignore the put downs and racist remarks. So good treatment for women should remain a priority even if women can grin and bear it and succeed, that's my position.
But where to put money, I'd agree with Ceci and Williams that more money should go to child care leaves, to child care onsite and cheap and abundant in academia, more like the biotech firms and Patagonia do it.
Their analysis looks quite good, I agree, and if bias is not effective in blocking the progress of women, we can still deplore it and see how it's impacting the well being of women and perhaps making their opting out an easier, more welcome choice.
But I think the variation between fields, such as Computer Science in different countries and the difference between biology and physics in the US, should make us think there's not just this one issue in play. If every senior professor in field X has felt discrimination, if there's only one woman in physics or engineering, it's not just child care, which is generally the same policy across the whole institution.
MKS
I think rather than saying bias is dead, Ceci and Williams' analysis has shown that it is not effective in blocking the progress of women in science. Instead, they are opting out. If we want more, we need to address the issues of family that make them want to leave.
I run into bias every day. I have no idea how Ceci and Williams found out it's dead, but they can't have looked carefully.
