2004 September 13 Monday
David Brooks On Spreadsheet People Versus Paragraph People

David Brooks argues for a widening political split in politics between those who think with words and those who think with numbers.

There are two sorts of people in the information-age elite, spreadsheet people and paragraph people. Spreadsheet people work with numbers, wear loafers and support Republicans. Paragraph people work with prose, don't shine their shoes as often as they should and back Democrats.

Brooks used data from the Center for Responsive Politics (probably from their opensecrets.org web site) to look at which occupations and industries give to Republicans and Democrats. Brooks provides a buch of examples of occupations which are heavily oriented toward the written word that lean heavily toward the Democrats.

Professors, on the other hand, are classic paragraph people and lean Democratic. Eleven academics gave to the Kerry campaign for every 1 who gave to Bush's. Actors like paragraphs, too, albeit short ones. Almost 18 actors gave to Kerry for every 1 who gave to Bush. For self-described authors, the ratio was about 36 to 1. Among journalists, there were 93 Kerry donors for every Bush donor. For librarians, who must like Faulknerian, sprawling paragraphs, the ratio of Kerry to Bush donations was a whopping 223 to 1.

Brooks doesn't provide as many data points for his "spreadsheet Republican" side of the argument. Though he notes that accountants lean heavily Republican and of course analytical numbers-oriented engineers lean Republican as well. But do academic physicists donate to Republicans or Democrats?

Brooks argues that humanities majors in college start to develop resentment toward majors in economics, accounting, engineering, and other "hard subject" majors. This resentment then causes the resenters to join a political party that is opposed to whatever these "hard subject" types favor. Perhaps resentment plays some role in this split. But a more likely explanation is that people who can apply mathematical techniques to what they learn process data about the world differently than those who are limited to verbal reasoning. Hence mathematically skilled people tend to come to conclusions that the verbally oriented people are not even going to understand, let alone agree with.

The opensecrets.org web site has per industry giving to Republicans and Democrats over a period of years. Accountants gave 52% of their donations to Democrats in 1990 but now in 2004 give 66% to Republicans. Why is that? Pharmaeceuticals shifted from 54% Republican in 1990 to 65% Republican in 2004. Part of that shift can be accounted for by the fact that the Democrats controlled the US Senate and House of Representatives in 1990 whereas the Republicans control both houses today. To have influence donors organized by industry tend to donate to incumbents who almost always win reelection. One has to be donating more out of ideological fervor (e.g. college profs or even most of the entertainment industry actors) rather than out desire to bolster an industry lobby in order for one's donations to reflect one's true beliefs. To what extend are donations in each industry made due to firm convictions versus a naked attempt to buy influence?

Check out some of the other industries. Note that computers/internet donations have shifted from a Republican dominance to about a tie. I suspect this is in part due to a heavy media industry presence on the internet. The electrical engineers and computer programmers are probably a shrinking portion of that industry sector.

Update: Mathematical ability is just one component of a larger set of cognitive differences that are likely to be the cause of diffferent political affiliations. People are attracted to political parties because they are attracted to people who share their sensibilities and emotional reactions. (PDF format)

But a recent study by Paul Goren at Arizona State found that voters typically formed their party affiliations before developing specific political values. They become Democrats first and then decide that they, say, oppose capital punishment and support trade unions. But how do they make that initial decision to be a Democrat?


Those M.R.I. scans suggest an explanation. Perhaps we form political affiliations by semiconsciously detecting commonalities with other people, commonalities that ultimately reflect a shared pattern of brain function. In the mid-1960's, the social psychologist Donn Byrne conducted a series of experiments in which the participants were given a description of several hypothetical strangers' attitudes and beliefs. They were then asked which stranger they would most enjoy having as a co-worker. The subjects consistently preferred the company of strangers with attitudes similar to their own. Opposites repel.

Occupations that experience a shift in the average personality types in them will show a change in political affiliations as a result. It is not just the ability of a person to do the tasks in an occupation or their economic interests as a result of being in an occupation that determines their political affiliations. Very instinctive innate tendencies are going to steer them toward the donkeys or the elephants.

Update II: Part of the split between the people who are more mathematical and those who are more verbal is a split between the masculine and the feminine. Women, on average lean further left and men further right. Verbally able men are probably more feminine in their processes and hence do not lean as rightward as men with more mathematical and spatial abilities. But of course a politically correct New York Times columnist is unlikely to bring up average biological differences between groups to explain inter-group differences in political views.

Also, another factor at work with university faculty is that so many of them have tenure and insulation from market forces that they can afford to look down on capitalism and hold more left-wing views. Plus, as Peter Drucker observed many years ago the very ability to become tenured and be secure for life is going to attract more left-leaning people to academic positions in the first place.

Share |      By Randall Parker at 2004 September 13 02:45 PM  Politics American Domestic

Kevin said at September 13, 2004 5:13 PM:

Brooks includes a figure which looks like a knife in the back of his theory: MIT went 94% for Kerry, and there can hardly be a group more thoroughly biased to numbers people than MIT students and faculty.

There might be an out in that the number is described as "employees" of the colleges in question, which presumably includes a broad swath of no-doubt paragraph-people adminstrators and bureaucrats as well as the faculty.

lindenene said at September 13, 2004 5:24 PM:

Yes, but they're academics. Not people out in the free market. I bet Brooks would be better off differing between the rent-seekers and those who are free marketeers and have to deal in pragmatic, common sense, results-oriented work.

John S Bolton said at September 13, 2004 8:10 PM:

An alternative hypothesis for this differential political affiliation effect, would be that the softer subjects are easier to politicize towards the left. The main distinction is not so much the numerical emphasis of one group of subjects, relative to the others, but the fact that one sort of subject is evaluative of human activity, while the other kind is about nature. In the sciences, a theory which predicts falsely is refuted; but in the arts and social sciences the unrealistic quality of a theory will often cause it to be praised as idealistic.

razib said at September 13, 2004 8:33 PM:

brooks is a paragraph person par excellence. he takes a few data points and makes sure they map well to his general squishy thesis, no hint of characterization of variance, skew, mean or modes in these data points. this guys gets the big bucks for browsing around a website??? christ.

gcochran said at September 13, 2004 8:43 PM:

Few people take a serious analytic approach to their job, which necessarily entails using math. Far fewer routinely apply the same analytical methods to a broad range of issues.
I don't think we're talking about much of a voting block.

Brooks doesn't amount to much. The New York Times has one of the worst stables of opinion columnists extant, particularly when you consider that they could easily hire the best if they chose.

razib said at September 13, 2004 9:03 PM:

brooks (and most columnists) are storytellers and not analysts. the former are far more entertaining than the latter, and people will pay more for stories than analysis if they are reading a piece over coffee & breakfast. there is some truth in a lot of the stuff brooks asserts, but his broad dichotomous typologies really are a load of crap.

also, how many real conservatives does this guy know? he's probably a radical right-winger in his bobo social circles. this is exactly the kind of persoon that needs the correction offered by numbers & statistics (which he has access too), instead, he just works off his hunches and observations. my personal experience is that uniform liberal social environments result in conservatives who are prone to use facts & reason to combat the religio-political hysteria they are alway surrounded by. conversely, uniform conservative social environments produce a similar type of liberal as far as public discourse goes. brooks probably lives in the latter situation, so it is easy for him to characterize liberals as factless rhetorical boobs, because his friends are probably that way since he's the most critical conservative they ever run into. but we all know, after the recent iraq fiasco, how empirical some "conservatives" can be, don't we?

btw, a prior i am a bit weirded out by the whole 'spreadsheet people' = right-wing equivalence, after all, isn't it the left that wants to socially 'engineer' everything in a 'scientific' fashion?

razib said at September 13, 2004 9:04 PM:

correction, brooks lives in the former circumstance.

John S Bolton said at September 13, 2004 10:02 PM:

Another source of confounding error might be that the subjects which evaluate human activity have a lot more to say politically, than the remainder. There may be some sorting by political beliefs, at or before the recruitment into one specialization or another, such that the evaluative subjects now tend to repel the more right-wing students. This could lead to a spurious enhancement of correlation.

noone said at September 14, 2004 4:24 AM:

"would be that the softer subjects are easier to politicize towards the left."

The Politics Of Shakespeare(heterosexual patriarchal oppressor) vs. The Art Of Shakespeare.

"after all, isn't it the left that wants to socially 'engineer' everything in a 'scientific' fashion?"

The Left abandoned "scientific socialism" during the 50's,as the reality of Marxist regimes became impossible to deny.If "science" teaches empiriscm,that any theory that does not conform to reality is an invalid theory(marxism fails),then "science" is clearly wrong and must be replaced with something better,in this case "post modernism" which claims to abolish reality itself as a mere construct(reality is merely perspective or opinion).So they can now claim,in the name of philosophy,that any reality that fails to comform to theory is an invalid reality.

It is also my opinion that many of these people vote not for an ideology so much as a statement of social class.

noone said at September 14, 2004 4:29 AM:

oops,that would be "empirism",my keyboard betrayed me.

Luke Lea said at September 14, 2004 6:32 AM:

There are plenty of rent-seeking Republicans, btw. Think Haliburton, or Montana ranchers getting farm support.

Derek Copold said at September 14, 2004 8:20 AM:

Brooks has become a parody of himself. Just when you think he can't possibly come up with yet another lame, half-baked social dichotomy, he does. At first it was kind of funny, but now it's getting pretty lame.

M said at September 14, 2004 9:19 AM:

Actually, Bush once worked in one the more intensely analytical positions I can think of, baseball general manager. I think he was probably in and out of baseball before the serious number crunchers started influencing decisions. Perhaps, Bush backers are paragraph people resentfully working in spreadsheet jobs. Brooks also didn't mention that right wing talk radio show host never bring out a spreadsheet unless they are negotiating a new contract. Perhaps, he is saving that for another column.

gcochran said at September 14, 2004 9:31 AM:

I'm trying to imagine Bush as an ardent disciple of Bill James. Smoke is coming out of my ears.

Randall Parker said at September 14, 2004 11:06 AM:


Bush was the front man. He wasn't involved in back office management according to accounts I read. He was brought in to do P.R. to get voters to vote for a bond to build a new baseball stadium. He managed to get the voters to vote for a bond that made him and his partners rich. There's a lesson there for Bush and America today. He's convincing good Republicans to screw themselves.

BTW, I think it matters not just that one crunches numbers but what numbers one crunches. An academic physicist could crunch numbers all day without ever crunching any financial numbers. Whereas an accountant or economist thinks about money and how money flows change things and how businesses and economies work. The people who crunch financial numbers are going to know more about how human societies really function. Hence their politics will be different.

razib said at September 14, 2004 2:05 PM:

well, also, baseball is a cartel. i don't know if one wants to argue for it as an exemplar of the free market....

Leonidas said at September 15, 2004 10:20 AM:

About 10 years ago I remember an Accounting prof telling the class about how a group of radicals in the 60's had attempted to get the Accounting Department involved in some of the protests of the day. The prof told the radicals that Accounting didn't really lend itself to radicalism the way that Anthropology or English Lit ever could. Accounting's as unradical as it gets. The only thing the prof could think of was getting rid of payroll deductions. He told the radicals that there would be a groundswell of protest against taxes and government waste if that were to happen (Imagine everyone writing the government a check for several thousand dollars every April). No takers among the "radicals"; just goes to show you how vested in the "system" they really were.

Post a comment
Name (not anon or anonymous):
Email Address:
Remember info?

Web parapundit.com
Go Read More Posts On ParaPundit
Site Traffic Info
The contents of this site are copyright