Monday, October 22, 2007

Academic Bias

Neil Gross (of Harvard University) and Solon Simmons (of George Mason University), writing in "The Social and Political Views of American Professors," do two things:
  • assess many previous studies of academicians' politics, and
  • report their own findings on the matter, based on a survey of 1,417 professors.
The bottom line is that Gross and Simmons try hard -- but fail -- to minimize the Left's domination of academia.

This is from the concluding section of the paper:
Although we would not contest the claim that professors are one of the most liberal occupational groups in American society, or that the professoriate is a Democratic stronghold, we have shown that there is a sizable, and often ignored, center/center-left contingent within the faculty.... (page 72)
But "center/center left" is a subjective and misleading description. Going back to pages 35 and 36, we find this:
To get a better handle on the relationship between political orientation and party affiliation, we constructed a new variable by performing a factor analysis on three items from our survey: the political orientation variable, allowed to remain on a seven point scale; the party affiliation variable, also kept in its original seven point scale; and a question we wrote that asked respondents to locate themselves on a continuum ranging from “extremely left” to “extremely right.” The analysis extracted one common underlying dimension, accounting for nearly 85 percent of the variance on the three items, and in our view representing a more robust measure of overall political orientation than has typically been employed in faculty surveys. Figure 1 [p. 36] shows the distribution of this new politics variable. The further to the left a professor is, the lower her or his score. A score of four indicates the middle of the distribution, which may be interpreted as a moderate political identity.... [T]he figure indicates not simply that most respondents are located on the left hand side of the distribution, but also that significant numbers of them are located near the center left, a fact too often ignored in discussions that treat the university as a site of uniform liberalism.
As if the "center left" were not of the Left. The "center left" is simply a less overtly menacing animal than the "hard left," just as Nikita Kruschchev was a less overtly menacing figure than Josef Stalin. The crucial fact is that both Kruschchev and Stalin were enemies of liberty, as is the American Left -- however its adherents choose to describe themselves.

Gross and Simmons earlier (page 27) reveal the slipperiness of their political taxonomy as they explain how they manipulated respondents' self-classifications:
In order to assess whether there were differences between the slightlys [those classifying themselves as "slightly liberal" and "slightly conservative"] and their colleagues further at the extremes, we averaged scores on all twelve of the Pew [Values survey] items. In this exercise, a score of 1 would indicate the most liberal response possible on all of the items, a score of 3 would indicate an intermediary position, and a score of 5 would indicate the most conservative response possible on all items. The score of those who stated their political orientation as extremely liberal or liberal was 1.4, while the score of those who identified themselves as conservative or extremely conservative was 3.7. The scores of those respondents closer to the center of the distribution in terms of political orientation were different: the slightly liberal scored at 1.7, middle of the roaders at 2.2, and the slightly conservative 2.8. Although the differences here between the slightly conservative and their more conservative colleagues are greater than the differences between the slightly liberal and their more liberal colleagues, that there are differences at all provides further reason to think that the slightlys should not be treated as belonging to the extremes.

Collapsing the data accordingly to a three point scale, we find that 44.1 percent of respondents [9.4 percent "very liberal" plus 34.7 percent "liberal"] can be classified as liberals, 46.6 percent [18.1 percent "slightly liberal" plus 18.0 percent "moderate" plus 10.5 percent "slightly conservative"] as moderates, and 9.2 percent [8.0 percent "conservative" plus 1.2 percent "very conservative"] as conservatives. Such a recoding thus reveals a moderate bloc that – while consisting of more liberal- than conservative-leaning moderates – is nevertheless equal in size to the liberal bloc.
If a "score" of 3 indicates an "intermediary position" (on a scale of 1 to 5), everyone who self-identifies in the range from "very liberal" (1) to "slightly conservative" (2.8) is left-of-center. Even the average "conservative" and "very conservative" respondent is barely right-of-center. Granting, for the sake of argument, that "slightly conservative" is "moderate" rather than "liberal," here is the correct breakdown:
80.2 percent "liberal" (Left)
10.5 percent "moderate"
9.2 percent "conservative"
That breakdown is entirely consistent with the most revealing data of all: the voting preferences of the respondents. From page 36:
In Table 10, we show the distribution of Democratic, Republican, and other votes in the 2004 Presidential elections across broad disciplinary fields. Averaging the figures for the social sciences and humanities generates a ratio of Democratic to Republican voters of 8.1 to 1. It is in business and health-science fields that Bush fared better, though even in business Kerry did better than Bush by a margin of more than 2:1.

Table 10


Kerry

Bush

Nader

Other

Phys/bio sciences

77.4

20.8

0.9

0.9

Social sciences

87.6

6.2

1.8

4.4

Humanities

83.7

15.0

0.0

1.3

Comp sci/engineering

61.9

33.3

0.0

4.8

Health sciences

48.1

51.9

0.0

0.0

Business

65.4

32.1

2.6

0.0

Other

81.6

17.5

0.3

0.6

Total

77.6

20.4

0.5

1.5


Q.E.D.

The Gross-Simmons paper is worth reading for its rich detail. But do not be seduced by the authors' attempt to minimize the academy's strong Leftward bias. It is real, and the authors' own data confirm its reality.

P.S. Gross and Simmons -- like the typical product of post-World War II "education," the media, and other fish in water -- seem unaware that what now passes for "moderation" is far to the left of the pre-Depression, pre-War, pro-Constitution mainstream.

In the election of 1972, George McGovern polled only 38 percent of the popular vote. In the elections of 2000 and 2004, both Al Gore and John Kerry (nothing, if not McGovernites) polled 48 percent of the popular vote. Throw in the Naderites, and the Left's share hovers around 50 percent.

Throw in big-government "conservativism," and you have...well, just what we've got: something much closer to socialism than to laissez-faire capitalism. Why? Mises explains, in "Middle-of-the-Road Policy Leads to Socialism":
The conflict of the two principles [capitalism and socialism] is irreconcilable and does not allow for any compromise. Control is indivisible. Either the consumers' demand as manifested on the market decides for what purposes and how the factors of production should be employed, or the government takes care of these matters. There is nothing that could mitigate the opposition between these two contradictory principles. They preclude each other.
That is to say, the good intentions (and ill intentions) of those who would intervene willy-nilly in private affairs for the sake of "the public good," "the children," and cheap "compassion" can lead to nothing but ruinous state socialism. One cannot be "slightly liberal," "moderate," or even "slightly conservative" in the defense of liberty.

I refuse to be nonjudgmental in such matters. You are either for liberty or you are against it.

Related posts:
What Is the Point of Academic Freedom?
How to Deal with Left-Wing Academic Blather
Lefty Profs
Apropos Academic Freedom and Western Values
Why So Few Free-Market Economists?
The Shoe Is on the Other Foot
Affirmative Action for Conservatives and Libertarians?