If professors made $500k/year, would they be Republicans? (updated and corrected)

by John Quiggin on October 9, 2016

Only a tiny minority of American academics are Republicans, a fact that is a continuing source of angst for much of the political right, as well as quite a few centrists. It’s generally assumed that this fact requires some explanation specific to the way in which universities work. The implicit assumption is that the group of those qualified and willing to take up academic jobs is roughly representative of the US population, and therefore contains roughly equal numbers of Democrats and Republicans. To state that submission is to see immediately what’s wrong with it. As a group, academics are obviously not typical of the US population. They have much more education and significantly higher incomes, though not as high as those of highly educated Americans in general. We know that these two characteristics work in opposite directions politically. Other things equal, more income is positively associated with Republican voting while more education is associated with lower support. So, a proper test of the idea that there is something special about academic voting patterns would begin with a multiple regression incorporating income and education as explanatory variables, then see if a dummy variable for academic employment was (statistically and quantitatively) different from zero.

But this is a blog post, so I’m not going to bother with all that hard work. Rather, I’ll point to this New York times article about the voting patterns of doctors. It includes a bivariate regression of voting patterns on income, with specialisations marked as observations It includes a bivariate regression of voting patterns on income for a sample of 30 000 doctors. This graph shows the resulting regression and plots the mean values for different specializations
doctorsvoting

The midpoint, at which specialities are equally divided between Republicans and Democrats is about $325k. The graph stops at $200k, where the predicted proportion of Republicans is 30 per cent. Extrapolating way beyond the range of the data to $75k (the average salary for full-time faculty in US universities and colleges according to Wikipedia), the predicted proportion of Republicans would be around 10 per cent, which is what’s observed in the data.

That assumes that the education variable takes the same value for doctors as for professors, which isn’t quite right. The MD degree takes four years, while a PhD typically takes five. Moreover, much of the MD is taken up with the practical business of learning to do a specific (very important) job, and less with developing skills in argument and reasoning. To the extent that education might have effects on broader approaches to thinking about the world (including politics) I’d expect that grad school would have more of an impact (per year) than med school.

So, if anything, this analysis suggests that, at 10 per cent, Republicans are over-represented. On the other hand, it presents those on the right with a simple solution to the problem. Just raise academic salaries to $500,000 a year and you should see lots of Republican professors.

{ 71 comments }

1

Maz 10.09.16 at 8:14 am

I wonder to what extent the correlation between medical specialty and political views is simply due to sex and race confounds. For example, about 90 % of orthopedic surgeons are white men whereas pediatricians are 57% female and 46% non-white.

2

Tabasco 10.09.16 at 8:57 am

University presidents do make $500,000 a year, if not more. What is known about their political tendencies?

3

Brett 10.09.16 at 10:19 am

That doesn’t look like a really strong correlation, at least from the graph. Instead, they cluster between 40-60% Republican despite a wide disparity in income (relatively wide – double the income isn’t much in terms of income inequality). It also doesn’t seem to correspond with practice set-up – you’ve got doctors who are mostly part of hospital teams in there coupled with independent/clinic practitioners, and it’s a wash.

4

Neil 10.09.16 at 10:34 am

The PhD is only the beginning of a professor’s education. Of course, physicians continue to learn on the job, too, and undertake formal training. But an academic’s further education is exponentially more intense. Much of their day to day work is a continuation of the work they did in graduate school. Assuming years of education have a linear effect on voting, we should expect dramatically lower numbers of republicans (perhaps the actual fugures suggests that the effects of further education drop off beyond a threshold).

5

Ronan(rf) 10.09.16 at 10:49 am

Isn’t the argument re academics that the career generally selects for people of a more liberal political temperament ? I can see this being the case at the extreme end of the jobs on the chart. Pediatrics attracting liberals, orthodpedics attracting conservatives. This is also backed up by the fact (afaik) that political ideology and party support is generally set by upbringing and social identity, which develops early in life.
This is a bigger part of the story, in my guess, how different occupations attract for different value systems, rather than how the occupation itself/and or income shifts political preferences.

6

David Steinsaltz 10.09.16 at 11:23 am

Not only is the correlation weak (as #1 points out) and largely driven by a few outliers — in particular, the highly paid and highly Republican orthopedic surgeons, and the (relatively) low-paid Democratic psychiatrists — but this is an example of “ecological correlation”: Clustering individual features (political leanings and salary here) by groups tends to exaggerate the apparent correlation. I’m not saying the effect isn’t real, but I think Republican need to push for academic salaries around a million dollars, just to be on the safe side.

7

Tabasco 10.09.16 at 11:27 am

@4

Without knowing the facts, I guess that pediatrics is female-dominated and orthopedics is male dominated, hence the salary difference.

8

Zamfir 10.09.16 at 12:38 pm

Whats the difference between pediatrics and orthopaedics?

9

Bob Costas 10.09.16 at 12:40 pm

This doesn’t explain the time series though, only a couple decades back there were far, far more republicans in academia than there are now.

10

P O'Neill 10.09.16 at 1:25 pm

One problem with the comparison is that it’s not just level of income that matters, but how it’s received. Academics may be professionals with advanced degrees, but they’re getting a paycheck each month. Most doctors will be in partnerships or other incorporated structures and look very closely at details of the tax system that don’t apply for standard wage and salary income. Republican tax policy spends an enormous amount of time relative to the number of people it affects on Subchapter S corporations — each time the issue arises, we’re told that it’s for that next Steven Jobs who hasn’t had time yet to incorporate his business, but it’s actually the way that a lot of upper class income is shielded from tax, and the relevant professions (doctors, lawyers) know that. Of course if academics shifted to 500K a year salaries, we might also see the emergence of alternative service provision structures to take advantage of the these loopholes, but then the whole academic contract would change.

11

efcdons 10.09.16 at 1:44 pm

My completely research free, anecdote only experience coming from being the son of an academic physician who works at a large hospital and meeting his friends and colleagues is surgeons are in general more Republican than non-surgeons in the same specialty area.

For example, cardiologists are more likely to be Democratic leaning than cardiac surgeons. It might be explained by money because in my dad’s specialty at least the surgeons do seem to make a lot more money. During (public) high school the neurosurgeon’s kids all drove their own german luxury car while my brothers and I shared a second-hand Japanese car. It was rough.

12

Anarcissie 10.09.16 at 2:11 pm

From my point of view there has not been a lot of practical difference between Republicans and Democrats, especially since the Democratic Party has moved as far right as it has in recent years. So I would think the main qualities apprehended by prospective adherents of either party would be style and traditions of occupation and neighborhood, rather than ideology or policy. The occupation and neighborhood would be related strongly to income and net worth. (Or at least this was the situation before the Republican Party began to fall apart.)

13

Layman 10.09.16 at 2:12 pm

My own experience with corporate executives and members of boards of directors is that high-income people are far more likely to be Republicans. I held my tongue for 15 fucking years listening to their right-wing bullshit. One board member once called me a communist over dinner, after I made the mistake of opining about the social destruction I’d witnessed in countries with big income disparities and a small or non-existent middle class.

14

jake the antisoshul soshulists 10.09.16 at 2:14 pm

It never hurts to follow the money. Though in the case of pediatrics, majority Democratic makes sense. People who specialize in children’s health would tend to favor policies like
S-CHIP and expanded Medicaid. One would think that Orthopedists would champion Medicare, because of the number of joint replacements they do.
I noticed that they did not include cosmetic surgeons. I would suspect they would trend
Republican even more than Othopedists. Unless they were influenced by the “liberal Hollywood friends”. ;-)

15

Maz 10.09.16 at 2:24 pm

My previous comment wasn’t published, so let’s try again. The association between medical speciality and political views is probably mostly due to race and sex confounds. For example, about 90% of orthopedic surgeons are about white males, whereas pediatricians are more than 50% women and almost 50% non-white.

The leftward drift among academics is probably largely explained by the fact that as old white guys retire they are often replaced by women and non-whites.

16

A 10.09.16 at 3:01 pm

is there any discussion of correlation vs causation? Could be that the med students who are more conservative just go for the higher paid specialties.

17

Placeholder 10.09.16 at 5:18 pm

Pursuant to this article breakdowns of ‘liberal academia’ studies show that left politics among academics is correlated to subject – those who actually study society tend to have…left wing views of society (weird huh, how is that?) Engineering and medicine, pace Mark Twain, are vocational degrees with a college education. http://www.washingtonpost.com/wp-dyn/articles/A8427-2005Mar28.html

Incidentally this is also the argument for socialism as Americans understand it, free college education. Journalism majors do not if fact make the sweet ‘we told you there were no WMD’ bucks. The benefits of their work accrue socially not privately and should be provided for as such see also: Man Who Thought Iraq War Swell Idea Thinks Anthropology Useless http://www.flpublicarchaeology.org/blog/blog/2012/12/19/florida-governor-rick-scott-and-anthropology-in-retrospect/
Or in Australia how Steve Keen warned about instability in the financial system for years and was rewarded by his University closing the whole economics department. Silly humanities majors! Don’t you know America can’t afford to be right? #mikepenceheadshake

18

Sebastian H 10.09.16 at 5:38 pm

“The implicit assumption is that the group of those qualified and willing to take up academic jobs is roughly representative of the US population, and therefore contains roughly equal numbers of Democrats and Republicans. To state that submission is to see immediately what’s wrong with it. As a group, academics are obviously not typical of the US population. “

It is fun how this analysis gets to be glossed like that ONLY for academics who don’t want to deal with differential treatment analysis. Shall we try that style when discussing programmers and women? In every other area where we see differential outcomes that stark they are considered strong evidence of direct discrimination.

19

Zamfir 10.09.16 at 5:43 pm

Engineering and medicine, pace Mark Twain, are vocational degrees with a college education.
And vocational degrees are really bad, right?

20

Linnen 10.09.16 at 5:50 pm

Anarcissie @11;
As the saying says; “Assertion made without evidence can be dismissed without evidence.”

Or as currently said; “Citation needed.”

21

jake the antisoshul soshulist 10.09.16 at 6:14 pm

Why do I go into moderation so often?
Is it my nym?

22

Placeholder 10.09.16 at 7:01 pm

@15 I am talking about voting patterns? I assume they vote with their social class status, which ‘professional’ degree holders seem to do. By comparison, ‘academic’ degree holders in humanities subjects become more radical than their social class. I was going to write a much longer post making more of the of the contrast of the Nigel Farages (rich, non-college) and the Ralph Milibands (working class, professor) of this world but I didn’t want to write to derail the thread. John Quiggan’s output on ‘non-college whites’ is available on this blog. http://crookedtimber.org/2012/09/10/the-white-working-class/

The point of course is the above-the-line modest proposal to conservatize academia by raising their pay, sarcastic or not, wouldn’t work. People who derive their social status from being social scientists and derive their status in society by being right aren’t going to change their opinions because they’re paid more.

23

SC 10.09.16 at 7:17 pm

” . . . aren’t going to change their opinions because they’re paid more.”

If you are planning research to test that, I’d like to add my name to the pool of available test subjects. I understand it’s a risky endeavor but I think I can spare, oh, ten years, or even longer, of being paid $500k. For science.

24

cwalken 10.09.16 at 7:26 pm

Neil Gross has already done all the hard work for you in “Why Are Professors Liberal and Why Do Conservatives Care?” and related articles. His answer; professor is a politically typecast profession.

25

PatinIowa 10.09.16 at 7:43 pm

And then there’s this: https://www.ncbi.nlm.nih.gov/pubmed/21670661, which points to declines in empathy in med school and residency. I don’t know if anyone has studied empathy (and other traits linked with political affiliation) in grad students–I can’t see why anyone would care. (I think I can make a case for why they should, but that’s a digression.) It’s a phenomenon that bothers lots of people in medical education.

Part of the Republicanization of surgeons has to do with how they think about their jobs–they (gross overgeneralization warning) believe that they’re the ones who control the operating room and that their decisions and actions accomplish most, if not all, the good that occurs. The more reflective among them will tell you that they have to think that way in order to function while elbow deep in the flesh of another human being, hacking, stitching, sawing and burning, and who knows, maybe they’re right.

My intuition tells me those are the kinds of socializations that lead to conservative beliefs, along with the immense economic rewards for the craft.

26

Donald A. Coffin 10.09.16 at 8:22 pm

Eyeballing the chart, I’d guess the (simple) correlation coefficient is about 0.7, so the R-squared of the regression would be about 0.5. I wouldn’t call that weak.

27

Collin Street 10.09.16 at 8:30 pm

In every other area where we see differential outcomes that stark they are considered strong evidence of direct discrimination.

In every other area we have no reason to suspect an a-priori link between [attribute in question] and [aptitude at Job].

Politics is different, because politics directly arises from your personality and your cognitive approach.

28

bruce wilder 10.09.16 at 8:55 pm

Politics arising from your personality and cognitive style in interaction with other people.

Our personalities and cognitive styles are effectively social strategies actively employed both in motivating and situating ourselves in the (economic) world and in competing and cooperating with others.

A collegial working environment or an independence from direct supervision is going to a very different experience from being a manager or a business executive in a highly structured and hierarchical situation.

With the surgeons, it would be interesting to know if there are different styles of control — some more collegial, some featuring greater domination — and if they correlate differently with political identity and attitudes.

29

bruce wilder 10.09.16 at 8:56 pm

I wonder if business school faculties skew more Republican?

30

Sebastian H 10.09.16 at 9:32 pm

Every job says they are the ones that are different. It would be awfully convenient for academics, especially as they are the ones who study and document discrimination, if ‘academic’ were the only one where that were true.

31

John Quiggin 10.09.16 at 10:05 pm

Sebastian H, the procedure here is bog-standard in labor market studies of discrimination. You control for the observables (age, education and so on) and see if there’s a residual difference captured by the characteristic in question. Gender almost always comes up significant, political affiliation usually not, except in China, where being a CP member is a big plus.

So, there’s nothing surprising about the finding of my rigorous study of the academic labor market. It’s the par result.

32

John Quiggin 10.09.16 at 10:09 pm

The data set for the regression is the entire sample of 34,542. The dots for specialities illustrate how far each specialty is off the regression line. So, the impression that the result is drawn from a small sample and driven by outliers is wrong, reflecting a bit of sloppiness in my original post. I’ve updated to correct this.

33

John Quiggin 10.09.16 at 10:20 pm

@13 It’s not just those who study society. Natural scientists are about as likely to be Republicans as are African Americans (6 per cent, according to Pew)

http://www.people-press.org/2009/07/09/section-4-scientists-politics-and-religion/

34

Donald A. Coffin 10.09.16 at 11:36 pm

Bruce Wilder @ 29:

Yes. Business school faculty are more likely to be Republicans:
“Academic field also makes a tremendous difference, with the humanities averaging a 10:1 D:R ratio and business schools averaging 1.3:1, and with departments ranging from sociology (44:1) to management (1.5:1). Across all departments and institutions, the D:R ratio is 5:1, while in the “soft” liberal‐arts fields, the ratio is higher than 8:1. These findings are generally in line with comparable previous studies.”

Faculty partisan affiliations in all disciplines: A voter‐registration study
Christopher F. Cardiff Department of Economics , San Jose State University , San Jose, CA,
http://dx.doi.org/10.1080/08913810508443639

35

Placeholder 10.09.16 at 11:56 pm

@33
1) Are you replying to me? Layman doesn’t seem to say that.
2) They may vote Republican at the same rates but not Independent. A different page of the same study suggests this is produced by differences among discipline. Democrat v. Independent shows noticeable variations between Chemistry and Physics on the one end and Geosciences and Biology on the other. I suggest that this is because America is one of the few industrialised countries where schools are forbidden from teaching the basics of those disciplines in secondary school. Southern Baptists haven’t had an opinion of Chemistry since Indiana had an opinion on the value of Pi. It seems like those professions are crying out for a chance to vote for a non-Fundamentalist Republican party. Kind of like New England as a whole.

Also, since the author is taking Sebastian’s complaints seriously:
Political affiliation is not an immutable protected class. Indeed, some would say job status in fact determines political affiliation and one can comfortably predict the political persuasions of unionised dock workers and CEOs alike. What other profession is in fact asked to reproduce the opinions of the general public? None. Just academics and teachers.

I re-submit my “professional consequences of reality’s well-known liberal bias” thesis unaltered.

36

John Quiggin 10.10.16 at 1:21 am

@35 Yes, this was a reply to you. Some interesting points there.

37

LFC 10.10.16 at 4:34 am

@Placeholder
Democrat v. Independent shows noticeable variations between Chemistry and Physics on the one end and Geosciences and Biology on the other. I suggest that this is because America is one of the few industrialised countries where schools are forbidden from teaching the basics of those disciplines in secondary school. Southern Baptists haven’t had an opinion of Chemistry since Indiana had an opinion on the value of Pi.

I don’t get offhand the allusion in the last sentence (though it sounds vaguely familiar), but where do you get the impression that US secondary schools are forbidden from teaching the basics of these subjects? There have been periodic battles over the teaching of evolution and over textbooks in certain states, but that hardly amounts to a widespread prohibition on secondary schools teaching the basics of chemistry, physics, and biology.

38

reason 10.10.16 at 10:27 am

Medicine and the US is rather a special case, because of their peculiar medical system. It therefore tends to make voting patterns strongly influenced by personal pecuniary interest (US medical professionals being the best paid in the world at the expense of other Americans).

39

Peter T 10.10.16 at 11:07 am

reason

Everyone’s pay is “at the expense of” everyone else, in the sense that, if they were paid less, others could be paid more. Since production is almost always a cooperative enterprise in which any individual’s contribution is impossible to measure with any exactitude, distribution is fundamentally a political problem. This is not an argument for what doctors earn in the US, but an argument for reducing their political clout. Obamacare is possibly a small step along this road.

40

Cranky Observer 10.10.16 at 11:36 am

= = = placeholder@11:56 PM: Geosciences and Biology on the other. I suggest that this is because America is one of the few industrialised countries where schools are forbidden from teaching the basics of those disciplines in secondary school. = = =

There are a few school districts in the Bible Belt where christianists have been been able to restrict some portions of the biology curriculum, and the trends in the Texas School Textbook Board are disturbing. But the United States is a large place. I have not seen any reports that high schools are “forbidden” from teaching biology or geoscience [only the best-funded would have a separate course in the latter; usually included in 9th grade general science]. Perhaps you could provide some links?

41

faustusnotes 10.10.16 at 11:48 am

Why do people go to such lengths to get good quality data like this (merging files and going through many loops to get the data) and then make all the worst possible mistakes you can make? In particular: Aggregating the data by profession so that it loses all personal information (even though that information is available); conducting a linear regression instead of a logistic regression, thus ensuring your model is wrong; not adjusting for confounders (partly due to the aforementioned aggregation but it is perfectly possible to do the analysis separately by sex, for example); if doing a linear regression, not first logit-transforming the data to get a proper logistic curve for a response that is constrained on the unit interval; and not doing a random effects logistic regression.

Because of this the findings are meaningless. Why, for example, speculate that the effect might be for more women in certain specialties when a proper multilevel logistic regression would just tell you the answer to that question?

Given this JQ’s extrapolation to low incomes is an overestimate of the decline (since a logistic curve would tend to bend away from zero).

Mediocre!!! (someone needs to provide the Mad Max gif for moments like this).

42

DrDick 10.10.16 at 2:13 pm

I think there are other factors which make physicians an inappropriate comparison. Firstly, many people go into medicine for the money, few if any academics do. Likewise, academics in fields which are high paying outside of academics (engineering, computer science, etc.) trend somewhat more conservative than other disciplines.

43

Ebenezer Scrooge 10.10.16 at 2:24 pm

I’m sure that this is on the back of everybody’s mind, but it needs saying: there are Republicans and Republicans. There is the funding wing, and the Taliban-fascist wing. They are very different people. In my experience, the funding wing is fully reality-based, but only cares about the taxes it pays. My guess is that most of the Republican physicians belong to the funding wing of the party. This seems consistent with the income correlation, as well as the salaried/non-salaried distinction.

fwiw, in the 1950’s, physicians were about the most Republican sector around. Since then, a lot more have gone on salary. And are women. And are minorities. (Jews count as minorities for this purpose.)

44

JimV 10.10.16 at 2:43 pm

LFC @ 37: “There have been periodic battles over the teaching of evolution and over textbooks in certain states, but that hardly amounts to a widespread prohibition on secondary schools teaching the basics of chemistry, physics, and biology.”

My anecdotal sense is that there is a widespread de facto albeit not de jure prohibition, at least in rural and suburban America. In my high school bilology course in northern NYS evolution was neither studied nor mentioned. Most of my family went to Houghton College in western NYS. The last I knew, students there must sign a pledge not to engage in non-Christian activities such as mixed-gender dancing or playing games with cards which have graven images on them (such as bridge), and the teachers there believe that the Theory of Evolution is false and the universe was created in seven days. The college turns out mostly grade-school and high-school teachers, and is accredited by NYS (I looked it up) as a source of biology and physics teachers.

I know of three similar colleges (places my nephews and nieces considered) within a 100-mile radius of where I live in Western NYS. When I lived in Mount Vernon in southern Ohio, there was (is) one in town (Nazareth College). When I last visited California, it was to attend a niece’s graduation from Biola (Bible Institute of Los Angeles). This is a biased sampling, but my guess is the portion of such colleges to secular colleges in the USA is similar to the portion of evangelicals in the population, and that they supply a somewhat disproportionate amount of secondary school teachers.

45

JRLRC 10.10.16 at 2:51 pm

I remember David Freedman.

46

LFC 10.10.16 at 3:04 pm

@JimV
I’m sure the picture is somewhat mixed, b/c of the elements you mention. More rural than suburban, I wd guess, but right now we wd just be doing anecdote v anecdote.

My sense, apart from this particular question, is that there has been some good research and writing on U.S. secondary education in recent years, esp. the impact of different approaches to inequality, residential patterns etc. One bk that I’ve read parts of but must read all of is Gerald Grant’s Hope and Despair in the American City (2009), contrasting Raleigh NC w/ Syracuse NY.

47

Sebastian H 10.10.16 at 3:07 pm

“Sebastian H, the procedure here is bog-standard in labor market studies of discrimination. You control for the observables (age, education and so on) and see if there’s a residual difference captured by the characteristic in question.”

There are many things I don’t know about economic analysis, but studies of discrimination in work environments is an area I know. The first question you ask when you want to show that an employer’s expert is misusing a study to excuse the employer’s statistically noticeably pattern of discrimination is: are any of the variables you have used subject to selection effects which are controlled or influenced by the employer.

Here the answer is clearly yes. Your analysis treats ‘education level’ as an independent variable when it is clearly is a variable tightly under the control of the entity which is suspected of discrimination.

This is why models are so important to the discussion. Your model seems to be (so please correct me if I’m misunderstanding) education makes you Democratic leaning and wealth makes you more Republican leaning as independent variables. Therefore we can straight line extrapolate those two variables to see if Republicans are unexpectedly not in academia.

You can’t use merely post-grad education because the differences in political affiliation aren’t nearly strong enough to excuse the employer. You really have to use PhD level education to get the effect you’re looking for. The problem with that is you don’t have much PhD level education that isn’t totally under the control of the institutions which are suspected of discrimination. Then we get to all of the annoying discussions about how ‘self selection’ is heavily influenced by the institution making the disfavored class increasingly uncomfortable at every step of advancement, that ‘how well they fit’ is very often code for discrimination, and that when you introduce lots of highly subjective values about success the candidates being discriminated against always lose out. We clearly don’t want to deal with all that.

So we have to reach OUTSIDE academia. The problem there is that academia controls PhD level education almost exclusively. So you have essentially no comparable group. You have to reach to medical doctors. But now you don’t have any comparison group because the set of doctors making only $70,000 is too small. (You probably should have tried lawyers where the under $70,000 group is almost half). So now you have to straight line extrapolate the ‘trend’ to income levels which are lower than 50% of your last data point, despite the fact that there is no way it avoids being a curve.

It is quite possible that they way you should be looking at it is at that WITHIN AN EDUCATION FIELD your chance of being Republican or Democratic ends up being correlated with income. In that case, physicians show a typical field of highly educated people which show a range where the bottom third of the field in earnings tends toward about 40% Republican and the top third of the field in earnings tends toward about 60%. Which is to say a very noticeable correlation, but not enough to swamp obvious things like “party your parents were a part of” or “party dominant in the neighborhood you grew up in”.

Your story fits better only under the assumption that “education level” is a completely independent variable, not at all under the control of universities.

(Note also that at the level where there are fewer ‘self-selection’ effects there is also much greater diversity in the variable we are studying. This is again exactly what you see in discrimination studies).

48

Theophylact 10.10.16 at 3:16 pm

placeholder @ #35: As a chemist, I’ve observed over the years that engineers tend to be by far the most conservative group in the STEM disciplines, with the order being roughly engineers>chemists>physicists≈mathematicians>biologists.

As I recall, at UC Berkeley during the Free Speech movement in ’64, only the Business, Engineering, and Chemistry faculties continued to hold classes during the strike.

49

exPhD 10.10.16 at 4:56 pm

http://press.princeton.edu/titles/10656.html
Gambetta and Hertog. Henry’s a fan.

We live in a culture of greed, when even scholars are assumed to be greedy. In the past scholars were thought of as better, even as others mocked them for it.

50

Ed Brown 10.10.16 at 6:45 pm

Yeah, doctors are a poor cohort to look at, given their narrow focus. Doctors are notoriously bad financial planners, for example. Many I have spoken with are quite badly informed outside their specialties.

51

Anand Sarwate 10.11.16 at 3:10 am

I think if you did the same chart by discipline in the academy you’d get a similar trend, and not just the 10% for all of academia. In fact, it’s a bit misleading to lump the professoriate into one bucket, since we’re far from cohesive.

I’m in engineering (in the US), and while the majority of my colleagues are leftish, we are significantly better compensated than those at comparable rank in the humanities. Those who teach in professional schools (law/business/medicine) make even more than engineering profs, and from thence come the Volokh Conspirators and their ilk.

In engineering, this is partly a sort of techno-libertarianism, but also where the money comes from: the defense industry funds a lot of academic engineering research, and a more conservative outlook puts you in the right frame of mind to write successful grant proposals.

The higher salaries for engineering and professional schools is a bit of a retention play. Industry salaries being what they are, it takes a vaguely ascetic mindset to stick with a teaching gig when “the market” would push you elsewhere.

As a final note, in grad school the TA Union would come around and try to get engineering students to sign union cards and most students balked because they didn’t see the point — they were funded on grants from their advisors, so what was all this nonsense about working hours? This is because the union was run by people who thought talking about solidarity and CliffNotes Marx would be inspiring, and who had no fucking clue what it is to be a graduate student in a grant-funded lab. What they really should have said was “look, do you want dental insurance? We fought to get you dental insurance so sign the fucking card.” An appeal to the teeth, as it were.

52

Hey Skipper 10.11.16 at 1:55 pm

Likewise, academics in fields which are high paying outside of academics (engineering, computer science, etc.) trend somewhat more conservative than other disciplines.

Which raises the possibility that education, or salary, aren’t particularly important in determining political affiliation.

Most, to the point of all, stars in the entertainment industry are lightly educated and fabulously rich. According to the education-salary model, they should be skew conservative; however, exactly the opposite is the case.

Academics in engineering skew far more conservative than those in, say, anthropology, yet both are equally educated.

Perhaps it is down to something else entirely. In engineering, political leanings don’t make the square root of zero difference: the bridge stands or falls because it was designed and built well, or not. When a hiring committee is looking at candidates, their publications won’t give a hint of ideological leanings, so any committee bias, no matter what it might be, would be far more difficult to implement.

(My mother was a tenured rhetoric and English professor at Florida Atlantic University. She told me that over her 20 years there, the hiring bias became grotesque: no one even slightly conservative was going to get hired, ever. Consider how much easier it is for the hiring committee in an English department to determine political bona fides from publication history than for the Civil Engineering department.)

I’d bet that if you were to plot the objectivity of disciplines — Civil Engineering on one end, and Feminist Studies on the other — there would be a far more obvious skewing of ideological preference. And there appears to be some of that effect within medical specialties — the success criteria for surgeons and anesthesiologists are much different than psychiatrists and family practice doctors.

53

Anarcissie 10.11.16 at 2:08 pm

Hey Skipper 10.11.16 at 1:55 pm @ 49 —
That’s an interesting point. Some have contended that seemingly objective kinds of thinking, for example, the engineering you mention, actually imply an ideology.

54

mdc 10.11.16 at 2:26 pm

I’ve served on academic hiring committees many times, interviewed dozens of candidates, and have never had a clue what their politics were.

55

Hey Skipper 10.11.16 at 3:02 pm

[Anarcissie:] Some have contended that seemingly objective kinds of thinking, for example, the engineering you mention, actually imply an ideology.

I think that is likely putting the cart before the horse. I’ll bet that most people develop preferences for certain activities long before they have any idea what an ideology is, never mind decide upon one over another.

My profession is one where there are a great many ways to get it disastrously wrong, only a few ways to get it right, and the difference has nothing to do with ideology. My profession is as uniformly conservative as an anthropology department is progressive. I also happen to be a gear head, and have, over the years, gotten to know a fair number of car mechanics, which is another very objective profession. In my experience, they are essentially all conservative.

Unlike the vast majority of car mechanics, the practitioners in my profession are all college graduates; many have graduate degrees. Yet both are equally likely to be conservative.

Hazarding a guess, I think this is largely down to right-brain v. left-brain. People whose thinking is left-brain dominant are far more likely to be drawn to objective activities: civil engineer v. architect; car mechanic v. nurse. Since people engaged in objective activities appear to be more conservative, the explanatory factor may well be down to left-brain thinking, not education or income.

There is a pronounced ideology gender gap. It isn’t beyond reason that differences in brain lateralization have something to do with that. If so, then right v. left brain thinking might well be a better predictor of ideological preference than education and income.

[mdc:] I’ve served on academic hiring committees many times, interviewed dozens of candidates, and have never had a clue what their politics were.

In what areas? Had you read their publications? Had those who screened the applicants read their publications?

56

Anarcissie 10.11.16 at 4:59 pm

Hey Skipper 10.11.16 at 3:02 pm @ 52 —
I’m going to give you a semi-anecdotal example which doesn’t necessarily prove anything but may prove interesting.

When I got into the computer-programming craft in 1965, about half of my fellow practitioners were women. In those days no one knew how to create an effective computer programmer and many were needed, so businesses simply vacuumed up likely-seeming candidates almost at random and threw them at the task. Some could do it, other couldn’t. A lot of those who could were women. I suppose since typing (at card punch machines) seemed to be involved in the job, a good many of these people came from female-predominant office jobs: secretaries and typists. who happened to be around when managers were looking for someone to send to computer school (a few weeks of minimal training) and, hey, they could type. A lot of them in fact turned out to be very good at the job. And anyone who was halfway competent found she or he was on the right side of a tight labor market, which was nice.

Things went along this way for a while but eventually it became noticed outside the craft that computer programmers were making a mighty lot of money and that computers were becoming pretty important, raising the social and political status of those who could deal with them. Nevertheless, the field was open: almost anyone who could do the job could get a shot at it. This development greatly excited academic institutions and they began to offer courses and degrees in such fictive disciplines as ‘computer science’ and ‘software engineering’. Business managers are suckers for credentials — it’s an effective mode of CYA — so the credentialed, even if they lacked knowledge and competence, were likely to get hired before the merely competent. As the earlier computer programmers burned out, got interested in other things, were promoted to their level of incompetence, died, etc., the newer credentialed people began to take their place.

Well, we all know what a recently graduated engineer looked like in the 1970s: young, White, male, and of a certain mentality which one might call ‘conservative’. The culture of the work also changed and became much more authoritarian. Even the language and the practices changed.

Not long after this development — the general exclusion of women from the craft, or ‘profession’ as it was now called — began to be considered a Problem and organizations like the ACM had conferences about it. (The racial and social-class components of the Problem were also occasionally noticed.) This Problem has hardly gone away in the 40 years which have ensued. First you do it, then you have a Problem about it, then you have a conference which you get paid to attend! Win-win-win, I guess.

I could go into how the culture and the substance of the work changed, as well as the composition of the workforce, but this is already too long. However, to me it’s an interesting case of a distinct series of political transitions in a seemingly apolitical field. Someone should write it up sometime.

57

mdc 10.11.16 at 5:24 pm

“Had you read their publications?”

Yes.

58

Hey Skipper 10.11.16 at 5:42 pm

[Anarcissie:] When I got into the computer-programming craft in 1965, about half of my fellow practitioners were women.When I got into the computer-programming craft in 1965, about half of my fellow practitioners were women.

(You seem to be about ten years older than I.)

I’m contradicting your personal experience — that’s a fool’s errand — but mine doesn’t mirror it. In the early 1970s, a friend of mine, far smarter than I, had access to the Cal Tech mainframe. The computer areas were exclusively male. My dad was one of the first COBOL programmers; he even wrote a compiler for it. He worked at Burroughs. I don’t recall any women working as programmers there, but I’d have to ask to be sure.

I bought my first computer, an Apple II, in 1981. I’ll bet the demographic for computer purchases then was almost exclusively male. No one I knew who had any apparent interest in learning about the innards of those things was female.

This development greatly excited academic institutions and they began to offer courses and degrees in such fictive disciplines as ‘computer science’ and ‘software engineering’. Business managers are suckers for credentials — it’s an effective mode of CYA — so the credentialed, even if they lacked knowledge and competence, were likely to get hired before the merely competent. This development greatly excited academic institutions and they began to offer courses and degrees in such fictive disciplines as ‘computer science’ and ‘software engineering’. Business managers are suckers for credentials — it’s an effective mode of CYA — so the credentialed, even if they lacked knowledge and competence, were likely to get hired before the merely competent.

In as much as my graduate degree is in computer science, specializing in software engineering, I’ll try not to take that personally.

I think you slam business managers rather too much. Any group will have in it people with neither knowledge nor competence, and it isn’t always easy to discover that ahead of time. Credentials are proxies — among 1000 people with a graduate degree in computer science, is one more, or less, likely to find people competent, knowledgable and trainable in computer areas than a 1,000 people who didn’t graduate high school?

An extreme example, of course, but that highlights the point: credentials matter (after all, not graduating high school is just as much a credential as having an advanced degree.)

Not long after this development — the general exclusion of women from the craft …

Absence is not the same as exclusion. In my profession, women were effectively and legally barred until the late 1970s. Their absence was well and truly down to exclusion.

Forty years later, women comprise about 6%. Exclusion is no longer a very good explanation.

59

Hey Skipper 10.11.16 at 5:44 pm

Oh FFS.

I’m not contradicting your personal experience …

It never ceases to amaze me I spot these things the instant after I hit Submit.

60

Hey Skipper 10.11.16 at 5:45 pm

mdc: In what field?

61

Placeholder 10.11.16 at 6:00 pm

Skipper, how are you using the word objective? Mathematics is the most objective subject because it is a ‘hard science’. Physics is less ‘hard’ than mathematics. Engineering is vocational degree. You seem to be confusing ‘getting paid’ with ‘being subjective’.

Sociologists are more liberal than engineers because they objectively study human society.

62

Ronan(rf) 10.11.16 at 6:08 pm

“. I also happen to be a gear head”

This means something quite different to my ears, and explained so much on my first (mis)reading ; )

63

Ronan(rf) 10.11.16 at 6:12 pm

Skipper, this might be of interest to you ‘re engineering, and supports some of your contentions, in a more qualified, empirical manner.( Afaik , I haven’t read rulebook yet just the articles it’s based on, and reviews )

http://press.princeton.edu/titles/10656.html

64

Cranky Observer 10.11.16 at 6:27 pm

= = = I’m contradicting your personal experience — that’s a fool’s errand — but mine doesn’t mirror it. In the early 1970s, a friend of mine, far smarter than I, had access to the Cal Tech mainframe. The computer areas were exclusively male = = =

Remember that during WWII the computers (in the original sense of the word) and calculating/decryption machine plugboard operators were almost entirely women, and many of them had math degrees ranging from BA to Ph.D. After the war some of those women went on to academic and business careers as programmers and researchers because there were no qualified men to obstruct them. IBM in particular didn’t discriminate and encouraged women to try programming. The bro-heavy computer culture we think of today didn’t really get started until around 1980.

65

Anarcissie 10.11.16 at 6:35 pm

Hey Skipper 10.11.16 at 5:42 pm @ 55 —
Above the front-line level, business management is a very difficult, highly political kind of job, and CYA is usually necessary, especially in large organizations. I am not faulting managers, just noting a circumstance that my main narrative is contingent upon.

As for ‘computer science’ and ‘software engineering’, there probably were such things in actuality in the 1970s, but they didn’t apply to training people to do business applications. My writing about these things is rather impolitely abrupt because the subjects are worthy of much more complex treatment which I don’t have the time and energy to give them. By 1980 I think we’re observing a completion point of the transition to the new, pretty much definitely male culture (in the business world), plus, we have the appearance of entirely new projects like computer-as-model-airplane and computer-as-table-sport which have different cultural roots (also pretty much male).

(The actual process of primitive computer programming is rather like knitting or crocheting, where one slowly and patiently builds up a big thing out of a lot of little things whose shapes don’t seem to have much relation to the shape and purpose of the big thing. It seems sort of feminine — much at odds with the culture of the craft when I last observed it in a office several years ago, where it had become competitive, athletic, and not very productive.)

I really didn’t properly get into what I was trying to express, which was that the substance or content of the work changed during the same period, which almost certainly affected and was affected by the composition and disposition of those who worked with it every day. This is advanced pop sociology and I’ll have to prepare my next rant more carefully and less rudely.

I didn’t know the current participation of women was as low as 6%. That is really striking.

66

Hey Skipper 10.11.16 at 6:49 pm

[Placeholder:] Skipper, how are you using the word objective?

In the sense that the criteria separating success from failure, or good from bad, are clear.

In that regard, Civil Engineering is, compared to Sociology, very objective.

Sociologists are more liberal than engineers because they objectively study human society.

I doubt you can defend the claim that sociologists objectively study human society. Also, you aren’t really addressing the horse-cart problem. Are people sociologists because they are liberal, liberal because they are sociologists, or liberal because academia actively excludes conservatives?

[Anarcissie:] I didn’t know the current participation of women was as low as 6%. That is really striking.

Wow, did I ever confuse the issue. I gave you every reason to expect that 6% refers to the computer field. My bad — my profession, which was for awhile, but is no longer in the computer field, (and is entirely different) is at 6%.

67

Sebastian_H 10.11.16 at 7:11 pm

The computer anecdotes are interesting–especially the suggestion that the exclusion of women became so strong only after it started being pushed through the university system.

68

Placeholder 10.11.16 at 7:21 pm

@Ronan: Less dramatic, but similar conclusion: Engineering degrees have a very reduced effect on religious belief and attendance compared to others similar too…huh…look at that… vocational degrees.
http://www.nber.org/papers/w15182.pdf

In the sense that the criteria separating success from failure, or good from bad, are clear. In that regard, Civil Engineering is, compared to Sociology, very objective.

Objectivity refers to the ability to make measurable and easily verifiable claims. Mathematicians and Philosophers can make claims that necessarily true. Natural Scientists and especially Physicists make claims that are easier to measure and verify than Biologists. Human society, being difficult to measure and providing many unreliable witnesses, is the least objective science. They can still make objectively verifiable claims such as ‘among surgeons voting republican is a dependent variable to which personal income is an independent variable’ (see: the article of this comment thread.) This objectively disproves that surgeons are more ‘objectively minded’ because their views correlate with their self interests and group beliefs.

Engineering is a vocational degree. It is a subset of physics and mathematics used for building things. Like Astrophysics but better paid. Like how Accounting versus Mathematics and Law versus political science or Medicine versus Biology. However, Physicists report liberal beliefs at a higher rate than Engineers despite being a more ‘hard science’! In other words, objectivity does not cause conservatism. Indeed Engineering seems have little predicable effect on political preferences.

Indeed because ‘objective’ insight into human affairs is clouded by human social forms, only those who can see through them than objectively view society. Like Philosophy is the only discipline that knows its own thoughts, Sociology is the only discipline that understand itself- therefore it is the QUEEN OF THE SCIENCES.

69

Anarcissie 10.11.16 at 8:24 pm

Sebastian_H 10.11.16 at 7:11 pm @ 64 —
I think the university system was simply the cutting edge of the social order. Computers actually originated in an academic environment (mostly) and women were allowed to approach them at first because the project was sort of arcane, sort of sealed off from the larger world. It was only when they became commercially valuable and their servants visibly well-paid that women began to be pushed out.

Women and improperly pigmented persons have been subordinated in other areas by other means. It’s not just the universities, conservative as they may be.

70

Sebastian H 10.11.16 at 10:56 pm

I would probably put it differently: the university system was the enforcement edge of the social order. Universities have huge incentives to contribute to a situation where merely demonstrating that you can do the work is not enough. This has happened in all sorts of areas, where demonstrating that you can do the work (often better than the person with the university credential) won’t even get you an interview anymore.

71

Cranky Observer 10.12.16 at 12:26 am

I don’t want to take the thread too far off course, but I would recommend reading CAR Hoare’s Turing Award biography ( http://amturing.acm.org/award_winners/hoare_4622167.cfm ) and Turing Award lecture (linked that that site). Computing science was well on the way to being formalized by 1960 (a good thing too as modern microprocessors, networking systems, etc use dozens of fundamental algorithms discovered by Dijkstra and his peers during that time frame), and software engineering grew out of IBM’s attempt to understand why the System 360 project came so close to failure around 1965.

Comments on this entry are closed.