How much do Duke professors make?
The answer can vary based on a variety of factors, which the Academic Council heard about in an update from the Faculty Compensation Committee during its last meeting of the academic year.
Amy Herring, Sara and Charles Ayres distinguished professor of statistical science and chair of the FCC, presented the update to the Council. Herring first described the demographic makeup of the regular-rank faculty, which includes both those who are and are not on the tenure track. Non-tenure track faculty positions include professor of the practice and research professor positions, as well as instructor and lecturer positions.
According to Herring, this demographic makeup formed the basis of the “equity analysis” and recommendations about the differences in faculty compensation across the University.
Demographic makeup of faculty members
Across regular-rank faculty, 39% are female. The FCC’s report also found that when broken down by track, 32% of tenure-track faculty members are female, compared to 54% of non-tenure-track faculty.
The FCC also analyzed the racial breakdown of faculty members. According to Herring, 73% of regular-rank faculty are non-Hispanic and white, 13% are Asian and 11% are from “historically underrepresented groups.” Unlike for gender, Herring said that there are not “massive differences” between the distribution of race between tenure-track and non-tenure-track faculty.
“On the tenure track, we have a higher percentage of Asian faculty members, and on the non-tenure track, we have a higher percentage of non-Hispanic white members,” Herring said.
Salary equity analysis
Perhaps the biggest factor affecting salary is the department that the faculty member belongs to, Herring said. She added that department affiliation explains almost half the variability in salary, while additionally controlling for the faculty member's rank can explain 80% of the variation.
“Any kind of equity analysis that we do has to account for these known factors, or it's really meaningless,” she said.
Herring and the FCC built a “robust regression model” to estimate faculty salaries based on race, gender and other factors including rank. This model was then used to generate predicted distributions of faculty salaries based on these factors.
The equity analysis consisted of Herring and the FCC analyzing the predicted distributions of faculty compensation across racial and ethnic groups and gender and comparing them to the predicted distribution of non-Hispanic white men, while separating them by rank.
Among non-tenure-track faculty members, Herring said that women, across racial and ethnic groups, earn less than the estimated median salary for non-Hispanic white men. A notable exception is among non-tenure-track assistant professors, according to Herring.
For faculty members on the tenure track, the overall variance in salaries between non-Hispanic white men and other demographic groups is smaller than for non-tenure-track faculty members.
Herring said that the trend of female professors earning less than male professors still stands for those on the tenure track. But the opposite is true for men from historically underrepresented groups in academia. They tend to earn more in salary compared to non-Hispanic white men.
Across tracks, there is a tendency for salaries of United States citizens to be higher than those of Green Card holders and non-United States citizens who are also not permanent residents.
The FCC also conducted an analysis that built a model for predicting salaries not conditional on race and gender and used it to predict the salaries of faculty members. If a faculty member’s actual salary fell more than one standard deviation below the prediction, their unique identifier was flagged and sent to the provost’s office. Non-tenure track female faculty members are overrepresented in this group, Herring said.
In general, salary increases as rank does, Herring added.
Duke vs. its peers
Faculty members in the department of mathematics and department of economics tend to make more in salary compared to faculty in those departments at peer institutions.
One faculty member asked if the salary comparisons across universities are adjusted for cost of living, to which Herring responded they are not. However, President Vincent Price responded to the question by saying that Durham generally has a lower cost of living compared to the locations of most of Duke’s peer institutions.
Duke’s peer institutions used in the comparison “tend to be in [California,] New England, Chicago,” Herring listed as examples.
In other business
The Council approved technical revisions to the faculty handbook, as well as revisions to the handbook concerning the Office of the Ombuds. It also approved degrees for those set to graduate on May 14.
Faculty members also approved a transition of the Council’s power to the Executive Committee of the Academic Council for the summer months.
Get The Chronicle straight to your inbox
Signup for our weekly newsletter. Cancel at any time.
Adway S. Wadekar is a Trinity junior and former news editor of The Chronicle's 119th volume.