University retention and males

Carleton University admits more male students than females. But it graduates more female students than males. Why? What, if anything, can and should we do about it? (I don't know.)

The public access data is here. (Datacubes is a lovely tool, but it takes a little time learning how to use it.)

You can see there's a difference from the public access data. I can also get access to Carleton internal data. Except I can't seem to access it from home, so what follows comes from memory:

1. Among the 20 or so Ontario universities, only Carleton, Waterloo, the smaller University of Ontario Institute of Technology, and the tiny Dominican University College, admit more males than females.

2. The exact numbers vary year by year, and across cohorts and faculties, but at Carleton we nearly always see lower continuation and graduation rates for males than females. Cumulative differences of 5 or 10 percentage points are not uncommon a few years after admission. The difference is big enough to matter.

3. The Faculties of Arts and Social Science, and Public Affairs (Law, Criminology, Political Science, Economics, Journalism and Communication, etc.) have a bigger male-female difference in continuation and graduation rates than the Faculties of Science, Engineering and Design, and Business. (FASS and FPA also have more female than male students.)

4. Unsurprisingly, the biggest drop in cohort numbers comes in the students' first year. But surprisingly, the male-female difference only starts to appear after second year. (Perhaps I should double-check those numbers, because they are surprising, and don't match what Ross Finnie and Theresa Qiu found as quoted below.) [Update: I checked, and my memory was correct. But Ross and Theresa say that more female students than males switch universities, so if a lot of females switch after first year, this might reconcile my results with theirs, because my data can't distinguish those who switch universities from those who drop out.]

An econometrician who works at the University tells me that the male-female difference in retention rates disappears once you add the students' grades to the equation. But that seems open to more than one causal interpretation.

Carleton is certainly not alone. Ross Finnie and Theresa Qiu (big pdf) have much better data than me, and have done a much better analysis than I could do, and they find the same pattern in Atlantic Canada:

"Men leave at considerably higher rates than woman at the university level: 17 versus 13.8 percent in the first year for the 17 to 20 group, 14.4 versus 10 in the second year, with a cumulative difference of 28.4 percent versus 21.9 by the end of year 2. “What’s the matter with men?” is thus seen to be a relevant question with respect to persistence rates as well as access rates – i.e., going on as well as getting into PSE.

The implications of these findings are important. Not only do men enter university at substantially lower rates than women (e.g., Finnie, Lascelles, Laporte (2004), Finnie and Mueller (2008), Frenette and Zeman (2007)), they are also considerably less likely to continue on in their studies. Gender differences in final graduation rates – i.e., the numbers actually obtaining degrees – are, therefore, skewed even further than the access rates we have previously been looking to would indicate.

Women’s switching rates are, conversely, a bit higher than men’s. This means that when we put leaving and switching rates together, which (again) is the “quit rate” from the perspective of individual institutions, the true gender differences in persistence in PSE (i.e., after allowing for switchers) are understated. The benefit of being able to include switchers in our analysis, as is possible with the PSIS data, is again clear. The reasons for these different gender patterns represent an interesting topic for further research." (p. 35)

Universities care about retention and graduation rates. They care because they don't get money from students who quit. They care because their government funders tell them they ought to care. And they care because they want their students to do well. Retention has been on the agenda at Carleton for as long as I can remember.

But I can't remember sex differences in retention being on the agenda.

I am pretty sure that if it were the other way around -- if women's retention rates were worse than men's -- we would be hearing a lot about it, and there would be lots of demands that somebody do something.

What explanations would you give, and what policies would you recommend, if it were the other way around?

Maybe it's time for the Goose Sauce Gambit. Let's do the gender switch. Are you saying it's part of boys' essential nature? Are you blaming the victim? Or are universities a hostile environment for boys?

Me, I don't know. It could be lots of things. Some of which aren't a problem, or are problems we can't do anything about. But sometimes I walk around campus, and try to imagine I'm not familiar with what I'm seeing and hearing, and try to see it all through fresh male eyes and ears. Would I feel welcome in that professor's office, given the posters on the door?

Sensitivity training?

Comments

[Eric Pedersen]

My question would be: of the men who are dropping out, how many of them are in ending up either in a trade school or some sort of industrial job*? Thinking back to friends who graduated high school with me, the guys who didn't go to college or dropped out are largely in the oil patch or related trades, whereas the women are more in (I would guess) lower paying white collar jobs.

In economics terms, there are two reasons men would drop out (or not enroll) : higher costs relative to women, due to whatever factors, or lower costs of quitting, ie higher opportunity costs to staying in. I think to tease those apart you'd actually have to track down outcomes of men and women who drop out vs. stay in university.

Miles Corak had an interesting post that I think touched on very similar ideas recently: http://milescorak.com/2013/01/26/the-success-of-women-in-politics-signals-that-gender-discrimination-is-still-a-problem/

[marcel]

2 comments:

1) This issue (differential retention) has been slowly getting more attention in the US over the past 5-10 years, and not just at the college level. Boys, esp. those from working class and poor households, are less likely to graduate from HS than girls. A variety of approaches are underway around the country in attempt to addresss this. Most involve changing the nature of school, not the nature of boys.

2) The distinction between essential nature and hostile environment strikes me as a bit of a red herring. It can easily be both, and my impression is that in dealing with situations where 1 sex/gender is performing less well than the other in some institution, people are increasingly willing to consider that both possibilities hold: an environment may be "hostile" (or at least a poor fit for 1 s/g or the other) because of issues about essential nature.*

If the goal is to change the situation -- in this instance, males' lower school completion rates -- and those who who are being held responsible are those right on the scene, -- school authorities -- then quick fixes to the broader culture are out of the question. It is likely to be much easier to adapt the institution to the s/g rather than the other way around. We are not talking about trying to prepare males to bear or nurse children (oh yeah, we have figured out a way for men to nurse children) or women to inseminate other women (oh yeah, that one's been solved too!)

*The extent to which essential nature is biological or merely culturally conditioned is not one that I have any desire to explore nor any expertise.

[K]

Nick,

I'm assuming that Carleton doesn't have any kind of affirmative action for boys? I.e. these boys who drop out had the same entrance qualifications as the girls who don't? And if so, when you say that dropping out correlates with (university?) grades, that means that boys with the same qualifications coming in are underperforming girls in terms of grades?

[Nick Rowe]

Eric: I'm trying to think this through: Assume that boys on average had higher opportunity costs and so lower net benefits of going to university than girls. So fewer boys than girls go to university. Would that necessarily affect retention and graduation rates? Because, conditional on having chosen to go to university, all boys and girls would have positive expected net benefits to university. See what I'm trying to say? Even if the probability distributions of net benefits for boys is to the left of girls, the truncated subsets of those distributions might not be. (Can somebody please say this more clearly?)

marcel: From my quick Googling last week, it looks like this question is more on the agenda in the US than in Canada, (except the US is 10 times bigger than Canada, so you would expect 10 times more hits on any topic). Some US lessons must apply here too, but I wondered how comparable the two countries are in terms of funding for university students, job markets, etc. (The Brits seem to be a bit ahead of us too.)

I think I agree with you on the "it can easily be both" nature/environment/interaction bit. And we look at the things we can change, and then ask whether policies to change them are good policies. Not all differences are bad, and not all policies that can reduce those differences are good. (But it's funny how I hear nothing about how to encourage boys to take arts or social work, while the "shortage" of girls in other subjects is seen as a problem we must do something about. If there are more boys than girls in X, there logically must be more girls than boys in Not-X. If one is a "problem" then, by definition, the other is too. Everyone is somewhere.)

[Nick Rowe]

K: I'm almost certain that Carleton does not have an affirmative action program for boys. (I can't rule it out in certain small limited access programs where they look at things other than just grades, but Social Work, for example, which looks at more than grades, admits hardly any boys.) We have the same high school GPA admission cutoff for boys and girls, conditional on program (the cutoff varies a bit by program, and the sex mix varies by program too).

Boys have lower high school grades than girls, and I haven't checked the data, but my guess is that the average boy admitted to Carleton would therefore have lower high school grades than the average girl. And I should check the data, but my guess is that that difference would continue for university grades. But why boys get lower grades seems to me to be part of the same puzzle, rather than being an explanation of the puzzle.

[Eric Pedersen]

Nick, re: higher opportunity costs already factored in. I thought about that when writing my comment before. My question would be : doesn't that same argument equally apply to people dropping out because of an unfriendly environment for men? Wouldn't people factor that into their decisions to go to school? Really, people who drop out are ones who've decided they miscalculated either the net benefits of a university degree, the net costs of getting the degree, or the net benefits of a non university job. If the potential benefits from the next best option are closer to the benefits of an university degree for men, I would still expect the net dropout rate to be higher, even conditioning on men having decided to go to university.

[Jim Sentance]

I think the better alternatives for men without a degree are the main factor driving differences in enrolment rates, so that's probably the place to start looking for differences in retention as well.

Going to University is a learning experience in more than one way - you're probably also learning if it is the best thing for you to do. So even if as you say, conditional on having chosen to go to university, all boys and girls have positive expected net benefits to university, they're getting more information on that as they go along, and a higher drop out rate for the boys could still be a reasonably expected outcome.

[Nick Rowe]

Eric and Jim: OK, that seems to make sense. It depends on what new information arrives after coming to university. But the difference does switch sign at the PhD level, where, IIRC, there are more men than women.

[Jim Sentance]

Both more men and a higher drop out rate for women?

[Jon]

Larry Summers was on to something when he discussed variance of ability. Men have more, women have less. The means are pretty much equal. Now think about education as an exponential or linear process. If you grow at rate r, this compounds or adds over time. So suppose that students have characteristic r +/- e. Everyone admitted to the school is about a level y at the time (lower they wont be admitted, higher they go elsewhere). Now let the process evolve. Everyone with mean r or higher is obviously fine. So what matters for the dropout rate is growing below r. If admissions was redone at the second year, not as many would be reaccepted. This covers more of the men.

Those with high r stay for graduate school and maybe become professors if they have really high r. Eventually the process starts to favor men because the weak portion of the distribution is continually discarded at each level.

[Nick Rowe]

Jim: eyeballing the Carleton data, there are more men than women in the PhD programs, and more men than women graduate with PhDs. The sex ratio in program looks roughly the same as the sex ratio graduating. I don't have access to admissions data at home, so I can't say who has the higher dropout rate. But if you think of Masters and PhD as just the fifth and sixth and seventh etc years at university, it is clear that more men than women are continuing into the sixth and seventh years.

Jon: that makes sense to me. But roughly half the population go on to post secondary, so if variance were the whole story, you would think that the numbers of men and women would be about equal in first year, and that the ratio of men to women would slowly rise thereafter. Whereas the ratio is less than 1.0 in first year, falls in second and third year and fourth year, rises back up for the MA, then exceeds 1.0 for the PhD.

[Jon]

Nick: At the time of admission everyone in the cohort is about at level L. Suppose 90% eventually graduate. Then you need to be at or above the 10th percentile of r within that cohort to stay in the program to completion. Now most of the women are comfortably clustered around the 50th percentile of r. So most stay in the program. Nearly everyone in the bottom 10% is a guy and drops out.

What assumption have I made? That the difficulty of the material grows a rate consistent with graduating 90% of the matriculating cohort.

Now we have applicants who need to get into a master's program. Say the top X% are suitable, how many men and women do you get? Well that depends on the variance and the accumulated growth and where those distributions intersect Lx--the level of proficiency that can be assessed to qualify the top X% for the program.

So the reason the ratio of men to women doesn't slowly rise continuously is because the r needed to stay in the program is set a t a level that retains a reasonable overall percentage, but qualifying for the program gradually depends on having a higher and higher r because only those people can reach the qualifying level at the correct age.

[Determinant]

Engineering was and still is a male-heavy programme, and I went to one of the other, female-majority-overall schools.

So control for how large your engineering programme is to the overall school.

Sex-discrimination in favour of men? Now I have heard. everything. In the Public Service of Canada, the one that matters in Ottawa, the Equality Act still says that women are an Equity Group with preference after all other merit criteria are met, but there is now an overall female majority in the Public Service of Canada and in fact more women then their overall percentage of the workforce.

One could argue that that particular Equity provision has run its course.

Where's Frances when you need her? Paging Frances, Frances, we have a Gender Economics thread on WCI and its going on without you!

[Neil]

I do think Eric Pederson is correct, but it goes beyond just the high wages available in male-dominated trades and heavy labour jobs. The higher drop-out rate is likely just the flip side of pro-male discrimination in the workforce. Even amongst the white collar workforce, women have to be more qualified in order to get the same opportunities.

As for your question about whether this matters for retention rates, I think it does. "Because, conditional on having chosen to go to university, all boys and girls would have positive expected net benefits to university." Both men and women attending university expect positive net benefits, but if the women's expected benefits are greater than the man's, the threshold at which a struggling student will decide to drop out is going to be lower.

Unrelated to what I said above, but I have my own experience in dropping out of university, and I'm not sure its in your students' best interests to work harder at retaining them. For most of us with middle class upbringings, university is seen as being just as mandatory as high school, and so at age 17 we dutifully pick a degree and sign up for it. Then even after realizing that it's not the right choice, we still feel bound to complete the degree because anything less makes us a failure.

About 2 years - and many thousands of dollars - after realising that I wasn't cut out to be an engineer, I finally gave in and dropped out. Best decision I ever made. Sure I made less money when I first entered the workforce, but after 5 years, I had caught up and found work that enjoyed in the process.

If you want to do what's best for the students that you're losing, you should focus on offering part-time education options. Once you've dropped out and gotten a job, you're no longer in the parent-supported financial position that makes full-time studies a viable option for youth. Right now, the options for part-time studies are very limited: you'd better want a business degree.

[hix@hix.net]

We have a similar situation in Germany, just that it starts in the school system already. There is at least one study which suggests women get better grades for the same achievment level.

[Nick Rowe]

Jon: OK. You are assuming men have higher variance in growth rates. I thought the assumption was that men have higher variance in levels.

Determinant: " In the Public Service of Canada, the one that matters in Ottawa, the Equality Act still says that women are an Equity Group with preference after all other merit criteria are met, but there is now an overall female majority in the Public Service of Canada and in fact more women then their overall percentage of the workforce."

And in a few years, as the older boomer males retire, the PSC will become even more female, because that's who they have been hiring. I have heard the same thing from a (female) friend who works in the PSC. I have never heard anything said about this in the MSM. Is there a conspiracy of silence? How come this never appears on the agenda? Or did I miss it? If I were a younger male trying to get a job in the PSC (if I were you) this would make me very pissed off.

Neil: "The higher drop-out rate is likely just the flip side of pro-male discrimination in the workforce. Even amongst the white collar workforce, women have to be more qualified in order to get the same opportunities."

How would you respond to Determinant's comment? If men were more discriminated against in those jobs requiring university credentials, such as jobs in the Public Service, which are a big deal for a university located in Ottawa, then you would expect to see fewer men than women complete university, especially for those degrees which the Public Service normally hires. Which is what we observe.

Again, suppose the facts were the other way around. Suppose fewer women than men went to and graduated from university. What would you say? Would you say that this is because women are discriminated against in the labour market, and so fewer women decide getting the degree will be a good investment?

hix: I think it's the same in Canadian schools. I think the only difference between Germany and Canada is that (from what I hear) Germany has a better apprenticeship system for the trades. So the boys all go and learn how to build BMWs, which is what boys love doing anyway.

[Nick Rowe]

BTW, I noticed something about how Ross Finnie and Theresa Qiu frame the question (or, more accurately, their reflection of how the question is normally framed, because they put it in quotation marks):

"“What’s the matter with men?”"

If it were the other way around, and women had lower retention than men, would we frame the question as "What's the matter with women?"?

You bet we wouldn't frame it that way. We would frame it as "What's the matter with universities?".

And anyone who framed it as "What's the matter with women?" would get a critical response.

[Patrick]

Nick: Have the numbers for males changed much over time, or are they just looking bad relative to women?

I'd say that any place that's full of young women running around in yoga pants and crop-tops is going to be hostile to young male higher brain function.

[Nick Rowe]

Patrick: From memory; the data don't go back far enough to show any big changes, except that both male and female retention rates have improved as we increased admission standards. The length of data gets truncated a bit because we only get a couple of years data on the more recent cohorts (i.e. the data tables are triangular).

I can't resist doing the Gender Switch on you:

"I'd say that any place that's full of young men running around in tight pants and T-shirts is going to be hostile to young female higher brain function." ;-)

[Shangwen]

Speculation: is this explained partly by universities expanding programs that appeal more to women, thus creating a self-reinforcing cycle? I don't know about education, law, or engineering, but health care programs have certainly expanded more rapidly relative to arts and sciences, and these are female dominated except. It wouldn't surprise me if gynomagnetic programs (for so we should call them) grow faster than other areas, given that such topics are often political hot potatoes. The whole STEM rallying cry does not really seem to have found real traction in funding.

[Jim Sentance]

I think the general pattern is that male participation rates plateaued some time ago, while women's have continued to rise.

On the Phd thing, Larry Summers' theory might be a factor, but I think again you have to take context into account. The Phd is largely about the academic market, a career doing research and teaching, and my perception is that the time pressures of academia are as much a disincentive for women as for example the time costs of getting to be senior partner in a law firm or other such jobs. Childbearing hits them disproportionately at a time when time is of the essence. I think there's still a large enough differential in the calculation about the desirability of an academic life in favor of males to drive a good part of the differential in numbers.

[Patrick]

Might be true. FWIW my wife (with poli sci degree and an interest in gender issues) seems to think so.

Anyway, for men there is some evidence:

[link here NR]

I think men seek to create male only spaces for good reasons.

[Jacques René Giguère]

Last year, the Musée du Chateau Dufresne in Montreal
[link here NR]

had an exposition about schooling in Montreal in the XIXth century. One of the main problem was that boys were leaving in the 3rd or 4th grade while girls would go on to 6th grade...
The solutions proposed were, among others,separate classes for boys, wearing school uniforms, having more sports and replacing female teachers with men so as to provide role models...

[Jon]

Nick writes: Jon: OK. You are assuming men have higher variance in growth rates. I thought the assumption was that men have higher variance in levels.

Yes, but isn't it fair to say that leads to variance in levels?

[Determinant]

And in a few years, as the older boomer males retire, the PSC will become even more female, because that's who they have been hiring. I have heard the same thing from a (female) friend who works in the PSC. I have never heard anything said about this in the MSM. Is there a conspiracy of silence? How come this never appears on the agenda? Or did I miss it? If I were a younger male trying to get a job in the PSC (if I were you) this would make me very pissed off.

Neil: "The higher drop-out rate is likely just the flip side of pro-male discrimination in the workforce. Even amongst the white collar workforce, women have to be more qualified in order to get the same opportunities."

How would you respond to Determinant's comment? If men were more discriminated against in those jobs requiring university credentials, such as jobs in the Public Service, which are a big deal for a university located in Ottawa, then you would expect to see fewer men than women complete university, especially for those degrees which the Public Service normally hires. Which is what we observe.

The short answer is that there was no plan, none, of what to do when the Employment Equity Act actually met its targets. It appears that the justification will switch from "women are underrepresented in the Public Service" to "Women are underrepresented in the workforce in general and we must be exemplary." The Act will not be repealed, I think.

I am a young male looking for a job in the federal public service, and there are a few things which can stack the deck in your favour. Federal hiring is a two-step process, you have to meet Merit Criteria first, that is the job requirements. Then, and only then, may Employment Equity criteria be considered and that only moves you up or down in the ranking. Most hiring is based on pools, so that means where you rank in the pool, but the pool will be emptied before they hire again.

The huge leg-up to give yourself at the Merit and Testing stage is to learn French. All positions in Ottawa are designated bilingual and the onus is on the Department to prove otherwise. Ottawa is also the centre of the universe for Second-language French training, there is an active and competitive market for language services. The biggest complaint in PS hiring is that they can't get enough bilingual candidates. When I say bilingual, I mean after the lessons you'll be able to read La Presse or Le Devoir without a translator, tell me what you read, and write an editorial reply to an article in French.

That is why when I applied to be a meat inspector two week ago, I got taken out of the pool and tested because I have good SLE results already. I can therefore work anywhere in that job and would not be limited by language profile. That was lightning-speed, and all because of my French results.

Second, the other advantage if you're adventurous is to apply to Quebec positions as an Anglophone. Montreal Federal PS jobs are all designated Either/Or, you can perform your job in English if you choose (Hehehe). All civil departments have under-representation of Anglophones in Quebec. Anglophones from other provinces who want to move to Quebec are even rarer. Frankly, the line is shorter and you may get to the testing stage just because you increase the "representative" factor. It makes the hiring committee look good.

A certain agency in Montreal currently has my bootprint on their rear ends because I dared go past the testing stage and they frankly couldn't handle an outsider getting that far, let alone an English-speaker. Too bad, I have the law on my side, and that's all that matters.

[Determinant]

The SPAM filter ate my comment! I am not SPAM!

[I just checked the spam filter, and it's not there, so I think Stephen may have already fished it out. NR]

[Determinant]

I was erudite! I was topical! I may have given Jacques pause!

[Nick Rowe]

Shangwen: good point. I wonder if more women came to university because universities offered more programs that women liked, or if universities offered more programs women liked because more women came to university?

But see Jacques Rene's point too. It may go back a long way. I think I remembered reading in Larks rise to Candleford (the book that Frances once blogged about about a small English village around 1900) where the boys all left school early to go work on the farms, while the brighter girls stayed at school longer. I read something recently (forgotten where) that I think said in the US more girls went to college than boys, because the girls wanted to become teachers. It changed post WW2 with the GI Bill, and is perhaps merely reverting back to how it always was.

[Determinant]

Until the 1960's you didn't need a university degree to teach in the public school system. You finished high school, went to a Normal School for two years and then started teaching. My grandfather started in the one-room schoolhouse he graduated from two years before. Ontario closed most of its rural one-room schools in the early 1960's.

[marcel]

I cannot find it online, but I recall reading the following some years ago. When it was founded (or refounded) in 1893, the University of Chicago, was one of the early leaders in what was then called co-education. After a few years, the faculty seriously considered restricting admission to men because (a) women were performing significantly better in (at least) many classes and (b) it was thought that this was significantly discouraging men leading them to perform even worse. The problem of relative college performance of the two sexes/genders is, apparently, not of recent origin.

[steve]

I think you should consider how university has changed over time. I started as a physics major, then changed to chemistry to go to med school. There was about zero graded homework. You got graded on tests and labs. You mostly worked alone, except for having a lab partner in some chem classes. My son is currently a physics/math major. All homework is graded and significantly influences the final grade. (Fluency with social media helps a lot of you want to perform well I am told.) Tests dont matter as much. Working in groups is very common. Perhaps this does not favor males.

Steve

[hix]

Interaction with attractive women lowering brain function sounds amusing, but without further information this is just one specific case of performance anxiety. So do men have more performance anxiety? And why?

[Chris J]

I think Determinant is right. Break the data into faculties and track retention vs completed year for men and women.

Hypothesis: Engineering admits students knowing many will not make it out of first year. These are disproportionately men.

[Nick Rowe]

Chris: I have already looked at the data broken down by faculty. That's what point 3 in my post was saying.

[Lord]

The reason no one is concerned about the lack of males in female dominated fields is they are all lower paid. This is about the plums. This may also result in more boys dropping out as lesser fields are unattractive to them, while girls that can't make it in higher fields can still find an accommodating position in a lesser field.

[Nick Rowe]

Lord: Maybe, but pay isn't the only thing that seems to matter to students, when they are looking for a subject. Try the gender switch:

'The reason no one is concerned about the lack of females in male dominated fields is they are all boring and nerdy. This is about the plums. This may also result in more girls dropping out as lesser fields are unattractive to them, while boys that can't make it in higher fields can still find an accommodating position in a lesser field.'

Plus, graduating from university seems to improve one's pay.

[Lord]

It is the only thing that matters to those outside the field and to status in general. If more girls did drop out as a result, their fields would be better paid and their status would rise. Graduating improves pay, but mostly among the plums. While it does so in general, it may not do so commensurate with the effort in lesser fields. Not graduating is certainly riskier, but boys are greater risk takers. If they can't hit the jackpot one way, they are more likely to seek another.

[Bob Smith]

Gender differences in risk-taking behaviour might explain the higher drop-out rate. If men are more willing to take risks than woman, they might be willing to forego the relative security provided by a post-secondary degree to pursue riskier, but potentially more rewarding, opportunities (starting up their own business, heading out to Alberta to work in the oil patch, etc.). Certainly, there's no shortage of examples of famous males who dropped out in pursuit of the big score (Bill Gates, Steve Jobs, or the annual multitude of would-be future NBA stars - although whether any difference is due to differences in risk prefence or the absence of opporunity for woman is an open question. The latter is probably the explanation for why dropping out of college might not be as attractive for a female basketball player).

Shangwen: "I don't know about education, law, or engineering, but health care programs have certainly expanded more rapidly relative to arts and sciences, and these are female dominated except."

That's certainly true of law as well. Mind you, I'm not sure that those could be characterized as "gynomagnetic" (great term, btw), since at least in medicine and law, the dominance of woman is a relatively recent phenomenon (and one that seems to correspond more to opportunity than anything else) - 30-years ago those fields would have been considered "phallomagnetic". It could just be that woman, on average, are better than men at the skills that make for good law and medical students.

[Nick Rowe]

I'm a bit uncertain about the risk preference argument. You might argue that going to university and getting a degree is a risky investment, with uncertain costs and an uncertain payoff.

[Bob Smith]

"You might argue that going to university and getting a degree is a risky investment, with uncertain costs and an uncertain payoff."

You might argue that, but that's never the way it's presented to high school students ("You need a university degree to get a good job. University graduates make lots of money"). It may well be that, in fact, a university degree is a risky proposition than, say, becoming a carpenter, but it's the perception that matters. For the high school graduates who expect to be making $90,000 a year by the time they're 30 (I can't find a working link, but I'm sure you'll recall that survey last year), a university degree probably doesn't present as a risky proposition.

[Nick Rowe]

Bob: but isn't it males who are supposed to have those very optimistic expectations? If the men who didn't go to university, or who dropped out, were all off making other even riskier investments, I could see it. Sure, there are some who are trying to be the next Bill Gates, but how many are working as truck drivers, auto mechanics, or just sitting in the 'rents basement playing video games and watching porn, and working just enough hours to buy the basics?

[Nick Rowe]

I'm thinking back to steve's comment above.

I didn't used to give out assignments in Intro Economics, just tests and exams. But a lot of the students said they really really wanted assignments, because they knew that if they did the busywork they would get a guaranteed grade, unlike those risky tests and exams. Just like in high school. And all the OISE types said that assignments were a good thing because of continuous blah blah whatever.

As a student, I would have been bored to tears by these stupid assignments. And they have almost no worth as a way of evaluating student's knowledge of economics; except insofar as that is correlated with a willingness to do busywork. So I resisted as long as possible. But eventually I gave in, and added 4 little assignments, worth 10% total, to keep the students and OISE types happy.

I have just looked at the names of the students who didn't hand in assignment #3. Mostly male names.

The boys never complain about having to do these stupid assignments. They know that if you go to skool you have to do stupid assignments to please teecher. They just don't bother to do them. I would have done the same. I didn't bother doing a psych assignment in first year, because it bored me too.

So, I think I may be part of the problem. I think I might scrap those assignments next year.

Is OISE to blame?

[Chris J]

Apologies Nick. In future I will read a second time before commenting.

[Patrick]

Nick: FWIW, that jibes with my experience. My classes in engineering the mid 90's tended to have about 30% or so on the homework assignments. I slogged through as much of the assignments as I could manage on my own and usually did pretty badly, thus dragging down my grades. But I was usually way above the class average on exams.

The clever students got together, divvied up the problems, and all got an easy 30% boost. It was more than enough to make-up for what they gave-up on the exams. Also, I think the profs knew that most students weren't doing the assignments alone, so they tried to compensate by making them longer and/or harder than they would have otherwise been.

Don't know what they do these days, but I suspect social media has only increased collaboration.

[K]

Nick,

"As a student, I would have been bored to tears by these stupid assignments"

Then don't give them stupid assignments. Give them tough assignments.

You can't learn to solve difficult problems unless you do lots of them. Reading the text book doesn't work, and essentially nobody will do tough work (figuring out how to solve a problem and doing difficult math) unless the teacher makes them. I remember some kids who didn't do the assignments (worth 10-30%) or "collaborated". By the second year they were lost. My undergrad was in physics but economics is no different.

You can't learn deeply without hard work. If that makes boys drop out they weren't cut out for a degree in a tough field and are probably better off driving a truck at least for a few years.

[Nick Rowe]

K: maybe.

Yesterday I gave a lecture on money, talking about the sort of hard stuff I talk about on this blog. In many ways it was a fairly crappy lecture. I wasn't clear enough, and I didn't present my thoughts in a very ordered way. But some of the boys really woke up, and started sticking their hands up. They wanted to argue. "But isn't paper money backed by the country that issues it?". "How can you say that the Bank of Canada controls monetary policy, when the Bank of Montreal produces a large part of the money supply too?". "What is QE and is it a good thing?" And some of them wanted to continue arguing after the class. Nobody said "Will this be on the exam?"

[Scott P Bacon]

I think Eric hit the nail on the head earlier when he wrote "how many [men] are in ending up either in a trade school or some sort of industrial job*?"

In my experience as an undergraduate at Carleton, every one of the 20-odd 'dropouts' I knew left because they had an awesome opportunity with a local start-up, co-op, fully paid Algonquin College trades-program, or the like.

Recently I had the chance to reconnect with friends who recently graduated and the main topic was quite surprising... Many of my colleagues who went through the full four years confessed that they regretted not ‘jumping ship’ to do something more 'hands-on' or 'practical' (rather than studying a theoretical discipline with no aspirations to continue to graduate school or the public sector).

Seems like an easy decision when you think about it though, no? Most university students go to university because they believe (perhaps unwisely?) that it will lead to some higher level of earnings upon graduation. However, if offered your reservation wage with a real wage growth rate similar to your expectations upon graduation, before you finish your studies, why not start your working life earilier?

[Nick Rowe]

Scott: that's interesting. What subjects were the 20-odd "dropouts" taking? Or were they from all over the university? And were they all/mostly guys? Were the opportunities that came up in any way connected to their being at university, or was it just random?

[K]

Nick,

"And some of them wanted to continue arguing after the class."

Awesome!

I suppose there are three categories of students:

1) Those who couldn't care less what you are teaching them but will just slog through the work because they want the credentials;

2) those who couldn't care less what you are teaching them and have no idea why they are there (predominantly boys in this group); and

3) those who are bored regurgitating irrelevant facts but are intelligent and engaged enough to get excited when you say something meaningful and interesting.

Personally I think you can help all three groups by making the homework material really challenging. The first group will learn way more than otherwise; you will help the second group by helping them drop out faster and not waste their time; and you will help the third group by letting them show you, the first group, and most importantly *themselves*, that they are actually really smart and different from their slacker buddies in group 2.

And by all means, ramble on in class. Learning the details requires self study anyways. Class is for synthesis, big picture, and debate. Get them excited, and then set the bar high for the work required. I still remember some lectures from *grade 7* because an inspired English teacher said some stuff I wasn't expecting to hear at school. I really worked hard for that guy.

Scott P Bacon: "every one of the 20-odd 'dropouts' I knew left because they had an awesome opportunity with a local start-up, co-op, fully paid Algonquin College trades-program, or the like."

And their grades were every bit as good as the non-dropouts? Or do Ottawa area startups just have a super high demand for inexperienced college underachievers?

[Neil]

How would you respond to Determinant's comment? If men were more discriminated against in those jobs requiring university credentials, such as jobs in the Public Service, which are a big deal for a university located in Ottawa, then you would expect to see fewer men than women complete university, especially for those degrees which the Public Service normally hires. Which is what we observe.

I don't have the stats, but I'd definitely look beyond entry level positions in assessing this, which seems to be Determinant's primary concerns. When you're hiring for senior positions "qualifications" is a softer set than entry level, where you're mostly looking to hit the check boxes. As far as I can tell, the pro-male discrimination in the workforce, seems to largely come from men having a higher perceived level of qualification, than women with similar resumes. When you look into senior positions in government and business, you'll find some extraordinary men and women, but also plenty of ordinary men. Not so many ordinary women. This same dynamic plays throughout the promotion chain.

Also, even in Ottawa, is there not a significant private sector? I know at the very least there's a significant shadow public service - that contracted out stuff that the Cons love - that isn't bound by the same Employment Equity Act. Government jobs are not the be all and end all on this issue.

Dropping out has an immediate short term cost. And it's certainly possible that a public sector dominated area, that short term cost may be higher for men. Finding an entry level job - man or woman - is hard, and the pay a dropout will receive when they do find something will likely be lower than a graduate. But where I think women have a much more reasonable expectation of being held back in the long term by dropping out, men are generally much more able to talk up their experience and move ahead without a degree.

[hix]

Interesting parallel universe where dropout rates are explained by kids with filthy rich parents at elite Univeristies starting companies. Pretty sure the real world in Canada too, is one where kids with poor parents are completly aware that going to college entails risks which will cause more anxiety and worse performance even when all else is equal. No one is going to prestent University education as desirable to them anyway. Rather they will get told that you can be super happy as a hairdresser and that they should not overaspire.

[Determinant]

Also, even in Ottawa, is there not a significant private sector? I know at the very least there's a significant shadow public service - that contracted out stuff that the Cons love - that isn't bound by the same Employment Equity Act. Government jobs are not the be all and end all on this issue.

Yes, contractors are bound by the Employment Equity Act. The Employment Equity Act applies not only to the Public Service, but to all Federally regulated places of work. That means railways, phone companies, anything nuclear, airlines and shipping; anywhere where the Canada Labour Code runs. A federal office is a federal place of work for everybody.

The Act may not be enforced for these companies, but that is an entirely different story from it doesn't apply at all. The "Shadow Public Service" is not not bound by the Public Service Employment Act, which is another statute entirely and deals with process and the holy definition of "merit". It also makes you a Treasury Board employee, which gets you a government pension, benefits and union membership; all the good stuff.

When you're hiring for senior positions "qualifications" is a softer set than entry level, where you're mostly looking to hit the check boxes.

And where are these mythical entry-level positions? Anything I have seen in most of the private sector has been transformed into a Purple Squirrel: 5 years experience, qualified with applicable Codes, yet only paid entry-level wages.

The Purple Squirrel phenomenon is so widespread I seriously doubt we actually have a skills shortage in this country, rather we have a failure of reason by HR departments and corporate managers generally. Market power run amok.

The government is less prone to purple squirrels, actually. Once you realize the "code" for resumes (get yours professionally done, the format is government-specific) and learn French, you're off to the races. I have repeatedly said this here so that Nick's students might benefit from it if he passes it on.

[Bob Smith]

Somewhat off topic, but I see the Provincial Tories have released their post-secondary education policy document. Their focus on colleges seems to reflect a concern about people attending university but dropping out (or worse, finishing university, only to have to complete a college degree to acquire useful skills).

[Jim Sentance]

http://communications.uwo.ca/western_news/stories/2013/February/study_dropouts_werent_prepared_in_first_place.html

An interesting study. Looks like students drop out after learning from their marks that they aren't cut out for it.

[Determinant]

Demographic snapshot of the Federal Public Service:

[link here NR]

The Employment Equity targets, that is that Public Service profile for the group meets or exceed that of the general labour force has been achieved for Women, Aboriginal Peoples and Visible Minorities. The only group still below target is People with Disabilities.

But I don't see any move to wind down any part Employment Equity programme once the Public Service meets or exceeds that target. I think there is a strong risk of programme drift there. Any remedial programme that does not have a clear "end target" and exit strategy risks becoming dead burden.

[Nick Rowe]

Determinant: do you ever see anyone writing about this? I don't, but maybe I'm just not reading widely enough.

I have heard of "programme drift", but I'm not sure I understand precisely what it means or what causes it. Maybe it's programme drift. Maybe it's sheer inertia. Maybe the true programme always was different from the published. This programme did appear to have a clear "end target", which it has sailed right past.

And this with a conservative government?

[Determinant]

"Programme Drift", as I used it, is a programme that had good intentions and a clear problem at one time, but that problem is now solved, but the solution remains and is now causing its own problems. A well-intended solution has ceased to be the solution and become the problem.

There is a profound disconnect in what employment equity is supposed to achieve in the Public Service in particular (ignoring the Federally-regulated private sector). Policy makers and high-ranking Public Servants may have felt at the beginning that it was to ensure that Canada has a broad and inclusive Public Service that represents all Canadians. That promotes the government's legitimacy in that everyone can feel that their community is represented in the Public Service and that, conversely, they are not excluded. In my opinion, that is both just and laudable. It also has a clear end-target, no more remediation when you aren't short of a given group.

That's the "macro" or "high-level" view, if you will.

The "micro" is that Employment Equity was supposed to give a hand up to disadvantaged groups, therefore give X person a job because they are a member of Y group, which has suffered discrimination historically. That is a profound benefit to Equity Groups even after the macro equity goal is achieved; what beneficiary would want to lose that? Is the programme a remedial, terminating policy or a guise for a rent?

Low-level (clerks and secretarial) jobs are often restricted to Employment Equity groups, it was a problem in Ottawa a year ago when a woman made an application for such a job. A sidenote is that the woman could have benefited from the "Women" column under Employment Equity so it was tempest in a teapot.

The short answer as to why nobody touches this is that arguing against Employment Equity paints you as a bigot. No, I don't see anyone writing about this.

[Frances Woolley]

Nick - "I am pretty sure that if it were the other way around -- if women's retention rates were worse than men's -- we would be hearing a lot about it, and there would be lots of demands that somebody do something."

I very much like the facts and figures part of the post. They speak for themselves. Non-completion is a huge problem - the studies find that people who start university and don't complete tend to have bad economic outcomes - worse IIRC than people who don't go to university at all. People who start university and don't finish even have higher divorce rates than people who complete, though there are some causality issues there. There is a huge amount of ignorance and complacency about such issues on the part of university faculty - and I don't think we can afford to be either.

Yet I don't think it's necessary to say "if women's...". First of all, while this might have been true 20 years ago, I don't think it's true any more - there is a huge amount of concern about boys failing in school. If there isn't concern about retention rates, it's probably just because the data isn't widely available - it's hard to get longitudinal data, and see how people are doing over time. Second, it's not good strategy - it alienates more people than it brings in (committed masculinists being a somewhat small group).

[Nick Rowe]

Frances: maybe. But the data on retention by sex have been available (at least at Carleton) for many years. And Ross Finnie and Theresa Qiu had very good longitudional data, and did many studies on access and retention, but I can only find that short passage in their one study on Atlantic Canada that talks about males and retention. It's unlikely it's only a problem in Atlantic Canada. And anyone can see that there are few males in Arts and Social Sciences, but nobody "sees" it, in the way they do "see" few women where there are few women.

And data collection sometimes follows the agenda, as much as vice versa.

What strategy did feminists use when they wanted to change the agenda? Did they worry about alienating people? It seems to have worked for them. And my interpretation of the data is much much milder than the things feminists have said. I have not jumped to the conclusion that "it's all discrimination". (I very much doubt that it is all "discrimination", though it may be in part.) I'm not even sure that it is a problem.

Yes, there is starting to be a lot of concern about boys failing (being failed?) at school. Just last week I saw a good article by Margaret Wente(?) in the Globe and Mail, that didn't blame it all on the boys. But I don't think it's an accident that it's a woman writing about the issue. Far too often, gender studies = womens studies = feminist studies. And the male exceptions mostly prove the rule.

Committed masculinists are a small group at universities. Those who say they are being even smaller. It may come from outside universities.

I don't have the sort of skills and abilities to collect and analyse the data properly. But others do. All I can do is raise some questions, and hope others will try to answer them.

[Nick Rowe]

From the WSJ, "why men are more likely to drop out"

[Jacques René Giguère]

From VoxEU on effects of mixed-gender classes in high school
http://www.voxeu.org/article/long-run-gains-not-mixing-genders-high-school-classes

[Shining Raven]

I see several problems in constructing a symmetry between women and men here. A lot of this has already been mentioned by other commenters, I agree that the problem is simply not as pressing for men as for women, and this is why there is less initiative from men to do something about it.

For me this is really the key: We still have gender imbalances, and they also hurt men in many regards, since they are pressed into conforming with certain roles, but they are still less inconvenienced by this than women were in the past.

Sure, there are fewer men in the social sciences and in the arts, but this is not seen as a problem, because the pay is crap, its not prestigious, and there are no widely-admired male role models who would inspire men to go into these fields against the disapproval of their environment. Are men short-changed by this? Probably, but then it is really up to them to emancipate themselves, go into "female" fields, and suffer through the teasing of their own gender ("Hahaha, you are studying to be a *nurse*? And you want to become a kindergarden teacher? Haha!" - It would probably be hard for a high school student or young college kid to overcome the pressure of his peers).

It certainly is a problem that there are fewer and fewer male elementary school teachers, for example, but the problem here is certainly not that *women* are having more opportunities than in the past. It is just that women very often moved into the low-paid and less prestigious social jobs, because men had better opportunities. And this is still where it is, in many fields, so men really do not want to go there. Simply raise the pay for elementary school teachers and nurses, and you will see more men there, I am sure... (although these jobs are also a bit of a dead-end, career wise, for most people, so this is also a problem.)

The other cultural problem that I see is the anti-intellectual climate that is often cultivated among boys. This is certainly going to affect academic performance, compared to girls. In my opinion, this is also a problem of sexism in our culture, since there is a lot of pressure on boys to conform to a certain macho picture, and I believe this interferes with their ability and willingness to do academic work. This is clearly an area where sexism hurts men, but of course this is strongly reinforced by the male culture, and I believe that is only ever going to change if men start to perceive it as a problem themselves - i.e. they themselves *want* to do the work, and feel that they have to fight back against the environment that is holding them back. I don't think we are at that point, though, so a male-feminist movement might still be some time in coming. (I am not sure I like the word "masculinist", is this widely used? I never heard it before.)