Gender and salaries: U-M comparison favorable with peers
but more work needed
Read the full gender salary study>
The University compares favorably with other doctoral institutions when evaluating gender differences in faculty salaries, according to an annual report of the American Association of University Professors (AAUP).
A new University analysis shows that since the last study in 2001 the difference between salaries of female faculty and their male counterparts has not significantly changed. Results of the more recent study were presented May 14 to the Senate Advisory Committee on University Affairs.
As in 2001, the study is a detailed analysis of faculty salaries on the Ann Arbor campus, excluding the Medical School (a study of salaries in the Medical School was completed in 2005).
“We take the issue of gender equity in faculty salaries very seriously,” says Lori Pierce, associate provost for academic and faculty affairs. “National data from the AAUP indicates a pattern of salary difference across the country among doctoral universities. We want to be sure we are doing everything we can to address the difference in salaries at Michigan. The new data will help us to do that.” Pierce says.
The current analysis used an approach identical to that of the previous study so that comparisons across years could be made. The analyses compare salary data of tenure and tenure track faculty, excluding instructors, as of November 1999 and 2005. Both studies examined full, associate and assistant professors.
The focus of both studies is compensation, defined by a faculty member’s 9-month salary. For faculty whose salary is determined on a basis other than 9-months, salary is adjusted to its 9-month equivalent. Salaries are expressed in current dollars rather than inflation-adjusted dollars so that the analyses of the 2005 data can be compared directly to the estimates reported in the earlier study.
As in the earlier study, the more recent analysis used two regression models to predict salaries based on several variables known to affect pay level. Along with gender, these included highest degree earned, years since degree, years at the University, departmental/college affiliation, administrative appointments, multiple appointments, market data, rank and years in rank. The overall salary difference after adjusting for all factors indicated that women were paid 2.5 percent less than men.
Pierce notes that the study did not include factors such as scholarly productivity and teaching evaluations. “The study gives us a big picture. It does not take into account the variations in individual achievement,” she says.
The raw data, looking only at gender, showed a 17 percent difference between men’s and women’s salaries. Pierce says this was not unexpected given that women faculty tend to be an earlier stage in their careers. Women faculty included in the study had, on average, been at the University for 10 years and had earned their highest degree 16 years ago. Male faculty members had been at the University an average of 14 years and had earned their highest degree 20 years ago.
Among male faculty in 1999, 19 percent were assistant professors and 22 percent were associate professors. Women were less likely to be full professors than were men: 59 percent of men were at the full professor level compared to 29 percent of the women. The current study shows that more women are full professors compared to the 1999 study: 36 percent presently versus 29 percent in 1999.
The recent study results reveal that gender differences in salary between men and women of comparable rank are smaller than the overall difference. The overall adjusted difference was 2.5 percent, but the difference within rank ranged from 2.3 percent for full professors to 3.1percent for assistant professors. When comparing 1999 to 2005, the difference across all ranks combined is not statistically significant, 1.1 percent to 2.5 percent.
Next steps, according to Pierce, include a follow-up study at the end of 2007-08 academic year.
“We want to explore the extent to which the starting salaries of new faculty members could contribute to the overall difference. Starting salary seems to be an important factor in national studies of gender differences in faculty salaries,” Pierce says.