Nor is a response rate of 27% especially unusual or necessarily a cause for concern in the absence of good ex ante reason to believe that the non-responders are very different from the responders (maybe there is some such, but I can’t think of it – obviously it would still be nice if the author had provided some decent data on what the non-responders looked like in other ways). Finally, the aging effects that you refer to are actually more contested in the literature (as best I understand it – I am a consumer rather than a producer) than you might think. The one clear result of the survey literature is that party identification is very stable indeed, and is mostly fixed at a youngish age (there are different models out there of how this happens – Chris Achen’s Bayesian approach is the one that I am most familiar with and I believe is pretty influential). Ideologies do change – but less than you would expect – and any putative aging effects are likely to be relatively slight over the short time period early in one’s working life that we are talking about. There is some work I believe arguing that the openness of people to change ideologies varies dramatically at different points in people’s lives – you are much more open to switching positions early in your life, and relatively unlikely later – but this is quite different (and indeed perhaps supports the proposition that training of type x shapes your political opinions in important and long term ways). There may be cohort effects here too (but again, in the absence of good evidence to the contrary, I suspect they aren’t extraordinarily important – the life experiences of generations a couple of years apart are not usually very different from each other).

Sebastian – I’ve taken a quick look at the most recent posts I’ve done where I got involved in controversy in the comments sections – this http://crookedtimber.org/2007/09/27/man-of-the-year/, this http://crookedtimber.org/2007/09/21/buergerlich/ were the two on the first page where I remember disagreeing with people. I don’t think they support your claims. Very likely my sarcasm – when I use it – exacerbates things – it certainly seems to get under your skin a bit. But, as stated, you do very often come into posts with an interpretation that no-one else seems to have, and that doesn’t seem to have anything to do with the actual claims that I have made. I’d feel a lot more sympathetic if you could point to the place in this post where I hinted, as you claimed, that “the cited fact is somehow important to the question of the validity of traditional economic thinking” or where I lent support to the mistaken idea that “data about which political direction economic training pushes you [is] useful for discussions about the validity of a particular type of economic training.” Those are the claims you made about what I was saying – I don’t see anywhere in the text where I said anything at all to that effect.

Making an assumptions of pervasive rationality and individual pursuit of personal gain in human bahaviour, which are not supported by the evidence. However, economists are looking for a box of analytical tools to enable them to better understand some aspects of social behaviour and – as Joan Robinson of blessed leftist memory once famously wrote – if anyone knows of a better assumption to make of individual behaviour than rational pursuit of personal interest then go for it.

It could be more frutiful here to pick up on the long held methodological position of the Chicago school of economics: science regularly makes simplifying assumptions in constructing theories – what matters is whether the theories yield falsifiable predictions which survive repeated testing.

Here’s my evidence that econ school has no consistent effect on political views. The 231 students were also asked if their views had changed while in school. About as many said they became more liberal as said they became more conservative (most said their views didn’t change). This isn’t perfect, but it’s much stronger than the figures Henry cited.

basically, because the third through eighth letters are the name of a product which is heavily marketed through the medium of comments spam.

Perhaps because your quotation “anyone under 30…” uses the S-word. I believe it has been pointed out before that the filter is suspicious of that term – I’ve no idea why.

Ragout to Walt: I’m willing to explain statistical significance to you, if that’s the issue.

I don’t know about Walt, but I’m always interested in your ideas about statistics and I would certainly welcome an explanation of your claim, which he disparaged. I couldn’t make sense of it.

If it’s a stupid thing to say, I imagine you’ll have no trouble spelling it out. I’m willing to explain statistical significance to you, if that’s the issue.

I think that the evidence I cited is a lot stronger than the evidence you cited. Asking people if their ideology has changed over the last few years has the huge advantage of comparing a fixed group of people over time. Besides aging, the only potentially serious problem I see is that people are often not honest with themselves, so I don’t know if I believe self-reports about changes in ideology. Any other potential problems are pretty subtle.

But the evidence you cited compares one highly selected group to another highly selected group. The problems are tremendous. I mentioned a few, but I could have mentioned many more. I didn’t even get into cohort effects, but as you note, they’re another pretty serious problem for the data you cite (consider the effect of 9/11 or recessions).

And you’re knocking me for not citing an anecdote that supports your view, when you failed to mention the survey evidence that cuts against your claims? Give me a break.

And to the others who are not impressed by my observation that 5th-year students are older than 1st-year students: I didn’t spell it out, because I thought it was too obvious, but it’s widely believed that older people are more conservative. Certainly I can cite opinion polls that suggest that people become more conservative as they age, whether or not they’ve been brainwashed by economists. My speculation is that the years after college see a lot of this shift. Surely you’ve heard the joke that begins, “anyone under 30 who’s not a socialist has no heart…”

Note that this alleged increase in conservatism is based on a survey with a response rate of 27% and a sample size of 231. The increase in conservatives from 10% 1st years to 23% 4+ years is probably of marginal statistical significance, if it’s statistically significant at all.

It’s astounding how much ink gets wasted by people accusing each other of being “left” or “right”… as if all opinion on any matter, economic, social, or political, has to be shoved into this single gray box.

It’s particularly obsolete nowadays, too.

Given the frequent purges of even mildly heterodox economists (e.g., University of Illinois at Urbana-Champaign, Harvard, Columbia, Rutgers, Notre Dame), it’s not surprising that some have compared mainstream economists to Stalinists. See Hodgson quote at: http://robertvienneau.blogspot.com/2007/03/neoclassicalism-as-lysenkoism-in-u-k.html