Thursday, April 12, 2012

An EP Study On The Scarcity Of Men And Women's Career Choices. Part II: The Four Sub-Studies

This post follows my first post below on the topic and looks at the study ("Sex Ratio and Women's Career Choice: Does a Scarcity of Men Lead Women to Choose Briefcase Over Baby?") in more detail. The quotes in this post come from the paper.

It discusses four separate studies. The first one is a statistical analysis of fifty US states. The last three are experiments carried out on small samples of female university undergraduates. The object in each of them is to ascertain whether women are more career-oriented when there is a scarcity of marriageable men. The last study also tries to establish whether hotter women (based on self-assessment) are less affected about such perceived scarcity than less hot women (based on self-assessment).

The experimental studies all conclude that a seeming surplus of marriageable men had no effect on women's expressed preferences between careers and family but that a seeming dearth of marriageable men did affect those choices, in the direction of more emphasis on career. The hot/not-so-hot study argued that the effect was driven by the not-so-hot women becoming more career-oriented when exposed to fictitious information about the coming end of men.

You might be astonished to find that I read through all the four studies and wasn't that bothered about them, given that I clearly am very bothered over the overall paper. That's because I'm pretty sure that both heterosexual men and women would behave differently if there was a very large imbalance of sexes, and that those behavior differences would apply mostly to the case where the numbers of the "opposite" sex are much smaller than the numbers of one's own sex. This is because such an imbalance makes it much less likely that the person herself or himself would find a life partner, and knowing that at a young age would affect one's economic and social decision-making.

What really got me going was the bit at the end of the four studies, a bit which essentially does not follow from the studies itself but appears almost pure propaganda or at least a cobbled-together argument for a study the authors did NOT carry out. It is that bit the popularization I read focused on.

Then to the four studies: The first one is a statistical analysis which uses data from fifty US states on the ratio of unmarried men to unmarried women between the ages of 15 and 44, an "operational measure" of the sex ratio in the "mating markets" and a bunch of variables the researchers chose to express career vs. family focus on women. These are the percentage of women in the ten highest-paying occupations for women per state, the average maternal age at first birth and the number of children a woman has. The study consisted of crunching out correlations between those variables and the operational sex ratio measure (apparently only in pairs). The conclusions:
These results provide qualified support for our prediction regarding how operational sex ratio is linked to women’s career choices and family planning. As the number of marriageable men in an area decreased, (a) the percentage of women in the highest paying careers increased, (b) the average maternal age increased, and (c) the number of babies born decreased. These findings suggest that the availability of mates may have important implications for women’s decisions to choose to pursue a high-investment career path and hold off starting a family.
Perhaps. But there are alternative explanations. The one which seems to be the most obvious one to me is that the authors assume a causal chain from fewer-marriageable-men to all the other variables, whereas the variable they use is a ratio of unmarried men to unmarried women. It may be the denominator that is the crucial variable here.

Young people in the United States move a lot. They go to college in other states than their own, they choose to find work in other states than the one they were born in. Come to think of it, young heterosexual women (men) could move from an area with a scarcity of unmarried men (women) to one where this is not the case. Thus, the implicit assumption in the study that people stay put in a particular place does not hold.

But more importantly, the results could follow from certain states being more amenable to high-level careers for women than other states, with more of the necessary industries, a more affluent client base and more universities and colleges which women need to attend first to get into those career paths. In short, the operational sex ratio may not push the results because of a scarcity of men. It may be correlated with the results because an "abundance" of young educated women have settled in certain states: those with the best career climates for women. And educated women marry later and have fewer children.

The authors call the support these results gave "qualified," perhaps because only six out of the ten top paying jobs had a statistically significant zero-order correlation with the operational sex ratio measure. My view is that the authors should have included more variables and analyses to test for the alternative explanation I outline above. I would also like to hear why women or men cannot move to find a partner inside the United States.

But the authors do acknowledge that the analysis in the first study cannot be viewed as proving causality running from the operational sex ratio to women's career choices. Hence the three experimental studies which manipulate the perceived sex ratio in various ways and then ask the study subjects (young female undergraduates) to answer questions concerning their views on career vs. family of this type:
To assess how sex ratio influenced desire for career versus family, participants responded to three items. Each item began with the following instructions: “Please indicate which is more important to you in terms of your future.” The three items were on 9-point scales anchored with the following labels: (a) having a family—having a career, (b) spending quality time with my future children—having a satisfying job, and (c) having a happy and well-adjusted family—reaching my full career potential. The order of the items was randomized. Responses were combined into a family versus career tradeoff index
The purpose of each study was disguised by giving the study subjects false information about the various parts of the study as being separate studies. The bit were the study subjects were fed fictitious information about the sex ratio was presented as a study about memory, for example.

I have the usual reservations about collecting psychological information presumably applicable to the whole world from small samples (some of these are very small) of American undergraduate students. Neither am I certain that the way the researchers manipulated the perceived sex ratio properly reflects how the study subjects would make actual career vs. family decisions. After all, if a heterosexual woman gets told oh-my-god-men-are-dying-out her first reaction might well be to become more career-focused. Or to run out and snap up a man. But of course men are not dying out.

Yet in the first experimental study which showed the study subjects three sets of eighteen pictures from the local dating scenes the ones who got the "scarcity of men" alternative saw either 12, 13 or 14 female faces in each group of eighteen. That's between 67 and 78 percent female. I would dearly like to know whether it was the women who were shown the 78% female pictures whose responses drove the results.

The other two studies did not use pictures but fictitious newspaper articles about the number of men vs. women at nearby college campuses. The articles came in pairs, one describing more men than women at the campuses, one describing more women than men. That according to the authors.

But the Appendix to the paper appears to contradicts that description. The articles are not just about a fictitious scarcity of men or women in nearby colleges. They are about a life-long lack of marriageable men/women in general. Here's the article those women read who were given the "few men" material:

Fewer Men for Every Woman for Today’s Students

By MORGAN K. JAMESTON, Senior Writer

There was once a time when the average college student could look around campus and expect to see an even number of males and females in a class. Those times are changing rapidly, however, according to new sociological research. Whether it’s in class, at work, or at the bar, college-aged women today should expect to see more women for every one guy.

The U.S. Department of Education recently released statistics of current enrollment patterns at national universities. The trends show that significantly more than half the incoming students across the country are women. “It’s astounding,” says Susan Rice, chief admissions officer for the University of Texas system. “College campuses are overflowing with young women.”

Across the universities of the Big 12, for example, many co-ed dorms have more women than men. “We’ve had to turn some of our boy’s bathrooms into girl’s bathrooms,” notes Taylor Bryan, a residential coordinator at Baylor University. “Whenever I walk around the dorms now, I always see some guy surrounded by a group of single girls.”
Interestingly, most students do not appear to notice the skew unless it is made explicit to them. At the University of Texas, for example, several students were asked to observe people around them for five minutes. Christina Jenkins, a first-year student, quickly noticed the trend. “Everywhere I looked, there were groups of women,” said Jenkins. “I was intrigued that there were so many single girls and so few men. I guess I need to get used to this.”

Demographers note that this trend will continue into the near future. “Looking at high schools right now,” observes Ryan Connick, a professor at Texas A&M, “it’s pretty clear that more women will be applying to college in the next few years.” Connick notes that this trend is a result of the number of males and females born in a given year. “We had a series of years a while back when more women were born. There is nothing wrong with this, but it will have an impact on people’s lives.”

The high numbers of women are likely to influence both the academic and the recreational lives of women and men. But it’s important to realize that the sex ratio is a lasting generational phenomenon. As the current generation of college students gets older, there will continue to be more women than men of similar ages. “When women graduate from college a few years from now, they will see the sex ratio follow them into their jobs,” points out Connick. “I wouldn’t be surprised if a woman ends up working in an office filled with single women and only a few men.”

Researchers across the country note that the sex ratio has looked different in the past and will likely look different again in the future. People who are college age right now, however, should expect to be surrounded by an abundance of women.

What do you think of that, huh? It's not just an experimental adjustment of the gender ratio in the local mating markets, my dears. It's a full-fledged the-sky-is-falling scenario. It states that there will be lots more women EVERYWHERE, all through the lives of the poor study subjects who read this:
Connick notes that this trend is a result of the number of males and females born in a given year. “We had a series of years a while back when more women were born. There is nothing wrong with this, but it will have an impact on people’s lives.”

Besides, there is no way in hell that the study subjects who read the reverse article on many men on college campuses would believe that or the rest of that story. It's too much in conflict with their daily experiences.

OK, I'm not that happy with the four studies, after all. But wait until we get to the third post on the bit added to the end of the paper, the one which does not follow from any of these studies, and you will see why most of this didn't look that bad to me.
All direct quotes in this post come from Kristina M. Durante, Vladas Griskevicius, Jeffry A. Simpson, Stephanie M. Cantú, and Joshua M. Tybur: "Sex Ratio and Women's Career Choice: Does a Scarcity of Men Lead Women to Choose Briefcase Over Baby?" Journal of Personality and Social Psychology, Online First Publication, April 2, 2012. doi: 10.1037/a0027949