Thursday, September 20, 2012

The Mismeasure of Man, 21st Century Style

Mark Shea posted a reader email with a link to an article about a book that I haven't read. I read the article. Newspaper articles on research usually end up making scientists either gnash their teeth or add the lurid headlines to their conference powerpoints for a good laugh with their fellow nerds, so it's quite possible that Rosin's work was WAY better than this article would have me believe. The article...failed to affirm me in my womanly splendor, though it sure did try! Not enough data, article. But it's ok. I got all the data about my womanly splendor RIGHT HERE.

I found most interesting the story of how men and women responded differently to work in Alabama; the women learned different jobs, and the men "sat around" waiting for their old jobs to return. The conclusion is that the women are more adaptable.

I wonder what the specifics are, though.

Were the men and women working the same jobs for the same number of years? Or had the women already dipped in and out of the workplace multiple times between kids, and worked for multiple employers, while the men had been working at the same company for years? In that case, it would be the prior habit of employment flexibility that made the difference. Did the men have longer investment in their prior careers (courses, licensure) and are their jobs likely to return when the economy improves? Sinking money into courses might be a poor investment IF their business picks up again. New businesses fare better when started by women rather than men, but are the rates of starting new businesses the same for both sexes? If only the most gung-ho, business-minded women start new business, while a larger percentage of men with a broader range of skills start them, you could expect the same result, regardless of the "flexibility" of the entire population. Also of interest was the observation that men and women in finance (same career) who switch to a new firm (same action) end up with different results; most of the men do worse, and most of the women do better. I wonder what the reasons for their moves are. What if men switch because they are not fitting in very well and are low performers and the really high-performing men tend to stay with one company? The low performer is now the new guy at the company AND not too great at his job, and people are quick to be unimpressed. What if the women switch because they are doing amazingly well and get offered a higher salary/benefits elsewhere, while the low-performing women are not wooed and decide they are better off with their annual percent raises? That would mean that, rather than women being "more flexible," the workplace is treating high-performing men and women differently.

Imagine a graduating high school class in which both sexes are equally represented and have an equal range in intelligence, work ethic, and skills.

Case A: The women are more flexible and adaptable on average than the men.
Case B: The flexibility and adaptibility is the same for both sexes.

Now I throw in totally made up statistics.

1. A recession hits. Women and men have equal unemployment. 95% of the women go back to school and get different jobs, while 95% of the men wait patiently for their old jobs to return. Now I change the stats. 37% of the unemployed women go back to school. 30% of the unemployed men go back to school. The rest wait patiently for their old jobs to return (or find other jobs in the same field as before, with no extra training). Now make it 43:41.2 and no, you don't get to see the data distribution or the standard deviation. Now make it 7.5:5 who go back to school and over 90% of both sexes who take the dole until it runs out. How impressive (sorry, I would say "significant," but this is journalism we're doing here) does the statistical difference have to be to conclude something about the difference between men and women? In any of these cases, could we use these statistics to tell whether we are in Case A or Case B?

2. Men and women start businesses, and the women's do better (they make more money, they last longer, higher clientele, franchising, whatever). Only 5% of the women start new businesses, while 20% of the men do. Are we in Case A, Case B, or both? Let's say at 15% of both sexes start businesses, but the women all open ice cream parlors (with homemade flavors that change each week) and the men all open gun stores (and they sell that fancy ammo that bursts apart when it hits the human body). These men and women are in a very big town, so the ice cream and guns are spread out. Can we learn that women are better at starting new businesses than men if the ice cream stores are more popular and make higher profits? If so, can we also learn (from these business stats alone) because the women have succeeded because they are more adaptable and in tune with modern society? Let's say that 70% of the ice cream parlors fold in the year, and 80% of the gun stores fold. What if it's 90%:95%? What if it's 5%:95%? What if we are in NYC? What if we are in Dallas? What if we are in Nome, which has been populated by masochists? Again, do these statistics tell us more about the differences between the sexes-in-general or the differences in a town's appetite for ice cream and guns, and the entrepreneurs ability to discern this appetite? Discerning what other people want -- a classically girly trait -- hmm, maybe there is a book in here for me, too! Now imagine that each man and each woman starts a different business -- restaurants, stores, tech support, handicrafts, skilled labor. Does that change the conclusion? Do you just ignore this factor because it is so complex? Does the difficulty of analyzing the business factor mean that it isn't the primary reason for success or failure? What if it isn't the business type, but the way it's run? Will you be able to tell? Will those data teach you about the differences between the sexes? What kind of data would you need?

3. Women and men go into finance, and some of them switch firms. The women do "better" and the men do "worse." Let's say the women make more money. No, wait, they get promoted faster. Pick your favorite. The women get it. I've already hinted that the reason for switching could have little to do with innate flexibility and could be critical for understanding performance. Maybe the study "controlled" for the age of the person and the reason for switching -- and the financial situation of both companies at the time of the switch (imagine the difference between leaving to join a young firm, or to start one, or to go from a small firm to a huge firm, or to go to a competitor, or to leave right before a big layoff, or to leave at the peak of financial success and Business Weekly articles featuring oneself and the old firm). Forget all that. The study has "controlled" for it. Not sure how, because they will never tell us, even in the supplemental figures of the original peer-reviewed article, but they have. I swear. Now let's just think about the ratio of men to women in finance. If the finance world were evenly divided between the sexes, would the information about higher success in a new firm tell us that women are more adaptable? What if the financial world is more like 70% men, 30% women? What if it's 90:10? Can we, without ever interviewing the bosses who gave out the raises and promotions, figure out their reasons for doing so, based simply on income tax reports?

I get that Mark, being a dude, read this article and thought "Gee, another article about how great women are and how men just won't get with the program." I got that too, to an eyeroll-inducing degree. But I, being a lady scientist, and therefore privileged in this particular case, was WAY more intrigued by the mishmash of "science," "math" the lack of hard information presented, and the grandiose conclusions drawn from tests that did not set out to test adaptibility, rigidity, flexibility, enthusiasm, or really anything other than employment and finances.

What if a misogynist had interpreted these dry financial facts? Would he have written about men's fortitude and women's flightiness? Would he have implied that the women at new firms had new lovers (and all that extra money and promotions prove it, because look at how the men -- the normal, appropriate workers -- fare at new firms)? Would he have concealed the fact that women's businesses do better and reported that men start more businesses, concluding that they are bolder and take more risks (which of course pay off) and have a better head for business? Probably, because I've read a bunch of articles that have gone on and on about the bold risk-taking man and the reason for his greater success. You don't often hear articles about the bold risk-takers who end up failing and going under -- unless they are being compared to the sensible, flexible women who start SUCCESSFUL businesses. I wonder. How many of the men who grew up reading those articles like I did now feel they need to be BOLD and start NEW BUSINESSES and DROP OUT OF COLLEGE just like STEVE JOBS? And don't check themselves for Steve Jobs' skill with computers and his vast intelligence? Or themselves for actual business sense? Or the world around them for successful men who went to business school and took entrepreneurship classes? Maybe we timid little risk-averse ladies just are reaping the benefits of not getting set up for failure by articles, movies and books that showed the rewards of risk-taking without ever emphasizing the risk part other than as a fun obstacle. Men in the 70's presumably had just as much testosterone and adrenaline (mustaches HEYO!), but were better at staying in school and learning the skills needed to succeed in their desired goals. Of course, they got laid off a lot then, too. Maybe the men of my generation got suckered by the propaganda. Maybe the economy sucks, life is cruel, random bad things happen to people, and God blesses the just and unjust alike. Or women are just awesome.

Makes me think. Now women are being lauded to the skies. Pride goes before a fall. There could be a lot of girls and young women out there who will think, "Hm, I guess I am great at being adaptable and flexible, and that is a winning strategy. Time to quit my job? Time to shell out lots of money for extra classes that I might end up not finishing?" The virtues that come easily to us -- if indeed we possess them and are not merely deluded by flattery -- are not the only virtues we should focus on.

The question of why so many men are dropping out or hurting in their finances needs to be addressed, but I don't think that poorly drawn conclusions help.

When you have the idea that one sex is superior in character at all times and in all ways, it can lead you to conclude things about cause and effect that have not actually been tested.

It can lead you to conclude what you expected to find.