Partisan Registration and Contributions of Faculty in Flagship Colleges by Sean Stevens | NAS
D:R Registration Ratio
Of the 12,372 professors sampled, 48.4 percent are registered Democrats and 5.7 percent are registered Republicans, a ratio of 8.5:1. In
2016 Gallup (Jones, 2016) finds that for the general population, 29 percent are Democrats and 26 percent are Republicans, a ratio of 1.1:1. The 8.5:1 ratio is lower than previous findings, such as Langbert, et al. (2016) and Langbert (2018), which found 10:1 to 12:1 because those studies looked at the highest-ranked institutions, in which partisan affiliation is the most skewed. The institutions in this study are the most elite in each state, but they are not in all cases the most elite nationally.
The correlation between the ranks of the 90 nationally ranked institutions in the sample and the natural logs of the D:R ratios is -.4. Since the sample is limited to the four leading institutions in 30 states and the District of Columbia, the relationship is probably weaker than it is in the universe of institutions (see Figure 1):
Logs of Nationally Ranked Institutions’ D:R Registration Ratios by
US News Ranks
D:R Registration Ratios by Discipline
In all nine disciplines, the D:R registration ratio favors the Democratic Party (see Figure 2). Broadly, the natural sciences are the least politically homogenous, with the D:R registration ratio ranging from 4.6:1 (chemistry) to 9.4:1 (Biology). In the social sciences, the D:R registration ratio varies to a much greater extent. Economics has the smallest D:R registration ratio at 3:1 among all nine disciplines, while anthropology has the largest at 42.2:1. With the exception of economics, none of the social sciences have a lower D:R registration ratio than any of the natural sciences. Within the humanities, the D:R registration ratio in English (26.8:1) was more than double the D:R registration ratio in philosophy (11.4:1). In fact, the D:R registration ratio in English is almost identical to the D:R registration ratio in sociology (27:1). This is also the case for philosophy and psychology (11.5:1).
Figure 2. D:R registration ratio by discipline
D:R Registration Ratio by Region
The region is also associated with differences in the D:R registration ratio (see Figure 2), which was highest at colleges located in the Northeast (15.4:1). This is consistent with previous work suggesting that political homogeneity among professors is higher in the Northeast (New England, in particular) compared to the rest of the country (Abrams, 2016). The D:R registration ratio in the Northeast is also roughly three times the size of the D:R registration ratio in the Midwest (4.7:1). The South’s D:R ratio may be elevated by remnants of its traditional association with the Democratic Party.
Figure 2. D:R registration ratio by region
D:R Registration Ratio by Gender
The D:R registration ratio among female faculty (16.4:1) is more than twice the D:R registration ratio among male faculty (6.4:1). Roughly two-thirds of the faculty sampled are male (see Figure 3).
Figure 3. D:R Registration Ratio by Gender
D:R Registration Ratio by Professorial Rank
Differences in the D:R registration ratio by professor rank are also evident, and it was highest among assistant professors (10.5:1) compared to associate (8.7:1) and full professors (8.2:1). It is notable, however, that the percentage of assistant professors not registered to vote (41.4 percent) is roughly double the percentage of full professors not registered to vote (20.3 percent). This discrepancy may be capturing an age or generational effect as well as a national origin effect.
The Census Bureau (2018) finds that 62.4 percent of Americans aged 25 to 44 and 71.4 percent of Americans aged 45 to 64 are registered. Professorial rank is a decent proxy for age, with assistant professors more likely to be newly hired and thus having relocated more recently for employment. As well, assistant professors are more likely to be from outside the United States.
D:R Donor Ratio
The overall D:R donation ratio was 95:1. Among the 12,372 professors sampled, in raw numbers, there were 10,260 nondonors, 2,081 Democratic donors, 22 Republican donors, and 9 donors to both parties across both election cycles. According to Hughes (2017), 12 percent of Americans make political donations, so the 17 percent of professors in this sample is above the national average, but it is below the means of 22 percent for Democrats and 32 percent for highly educated Americans. This can in part be explained in part by the 21.3 percent of professors not born in the United States.
1 The donations are almost exclusively to Democratic candidates and committees.
As can be seen in Table 1 below, the overall ratio of Democratic to Republican donors is more extreme than the registration ratio. The ratio of dollars contributed to Democratic versus Republican candidates and committees is $21: $1. As is evident in Table 1, the most elite institutions such as Berkeley, Brown, Bowdoin, Cal Tech, and Colby have the most extreme registration and contribution ratios. It is also apparent that donations are skewed in favor of the Democrats.
D:R Donor Ratios by Discipline
Across the disciplines sampled, the D:R donor ratio favors the Democrats (see Figure 4). Differences between the natural sciences, the social sciences, and humanities, however, are not as clear. Economics again has the lowest ratio at 17:1. Sociology (31.4:1) and philosophy (79:1) were the two other disciplines for which the D:R donation ratio did not exceed 100:1. Of note, although the D:R registration ratios for chemistry and mathematics did not exceed 6:1, their D:R donor ratios were 113:1 and 118:1 respectively. The percentage of donors within anthropology, English, and philosophy all exceeded 20 percent of the professors sampled from those disciplines respectively.
Figure 4. D:R Donor Ratio by Discipline.
D:R Donor Ratios by Region
In terms of regional differences (see Figure 5), the Northeast (97.3:1) does not have the highest D:R donor ratio, which it does with the D:R registration ratio; rather, the West does at 135:1. Consistent with our findings above for the D:R registration ratio, the Midwest had the lowest D:R donation ratio--at a mere 62.2:1.
Figure 5. D:R Donor Ratio by Region
D:R Donor Ratios by Gender
The D:R donation ratio for female professors (216.0:1) was considerably higher than the D:R donation ratio among male professors (67.5:1).
Figure 6. D:R Donor Ratio by Gender
D:R Donor Ratios by Professorial Rank
With respect to professorial rank, the D:R donation ratio was highest among associate professors at 162.1 and lowest among full professors at 68.7:1. Once again, there is evidence of less involvement with party politics among assistant professors, relative to their higher-ranked peers: only 10 percent of the assistant professors sampled made a political contribution, compared to 16 percent of associate professors and 21 percent of full professors (cf., Putnam, 2000).
Registration rates give a more modest picture of skewness in partisan affiliation among elite academics than donation rates do. Among registered Democratic professors, the donor D:R ratio is 251:1. Among registered Republican professors, the D:R donation ratio also favors Democrats, although at a more modest 4.6:1 rate. It would appear that the professors registered as Republicans often tend to be loosely tied to the Republican Party, so the skewed registration ratios may understate the skewness in academic political affiliation and culture. Among professors registered to minor parties, the D:R donor ratio is 10:0. Among unaffiliated professors, the D:R donor ratio is 50:1. Among professors not registered, the D:R donor ratio is 105:1. These findings triangulate and confirm the overwhelmingly lopsided partisanship in leading colleges.
Summary and Conclusion
The D:R donation ratio and the D:R registration ratio tell a story that is broadly consistent. The D:R donation ratio favors the Democratic Party in all nine disciplines sampled. Compared to the D:R registration ratio, the skewness in the D:R donation ratio for each discipline is more extreme than for registration. For six of the eight or nine disciplines, the D:R donation ratio exceeds 100:1. The D:R donation ratio among female professors is greater than among their male counterparts, and the ratios are lowest in the Midwest. It is also evident that assistant professors are less engaged in party politics as measured by the smaller percentage who register and who make political contributions.
However, the D:R donation rate is highest in the West, not in the Northeast, while with respect to registration it is highest in the Northeast. The donation ratio is also highest among associate professors while the registration ratio is highest for full professors. Whereas with the D:R registration ratios the natural sciences have the least skewness, this was not the case for the D:R donation ratios; economics has the least.
The findings regarding assistant professors are also noteworthy. The D:R registration ratio is highest among this cohort, yet they are also more likely not to be registered to vote. When one also considers that 90 percent of the assistant professors sampled also did not make a political contribution in the 2016 or 2018 federal election cycle, then one plausible interpretation of our findings is that assistant professors are less likely to be involved in traditional party politics, although there is no evidence that is inconsistent with the broad evidence that conservatives and Republicans are absent from leading colleges around the country.
The trend for the D:R registration ratio among the professoriate to increase will likely continue because the D:R registration ratio is highest among assistant professors in our sample. This presumably would also produce an even greater D:R donation ratio.