Skip to Navigation
University of Pittsburgh
Print This Page Print this pages

May 17, 2007

Gender differential in faculty pay misleading, Beeson says

Taken as group, women faculty at Pitt earn only three-quarters of what their male counterparts earn, according to a report issued last fall by the Association of American University Professors (AAUP). (See Nov. 9 University Times.)

But that statistic can be misleading if not analyzed in its proper context, a Pitt official told a University Senate committee this month.

According to Patricia Beeson, vice provost for graduate studies who tracks gender equity comparisons, extenuating factors include:

• The large number of women hires in recent years.

• The gender make-up of the academic ranks.

• The gender breakdown by school and discipline.

• The gender breakdown by tenure/tenure stream and non-tenure stream.

“When the data came out back in November, when we saw that the overall ratio of the average salary of women compared to men was 74.8 percent and we were low compared to the average in our peer group, we were very concerned, particularly since we’ve tried to address some of these issues,” Beeson said.

Starting with the most recent AAUP data from 2005-06, Beeson offered a comparison of Pittsburgh campus faculty who hold the rank of professor, associate professor or assistant professor against Pitt’s traditional peer group of the other 33 public Association of American Universities (AAU) institutions.

“This comparison does not include lecturers or instructors, but it does include people who are not in the tenure stream who are at the full, associate and assistant ranks,” Beeson told the Senate budget policies committee May 4. “That is not that common at Pitt in the Provost areas, but there are quite a few in the Health Sciences.”

Her comparison also included non-clinical medical faculty, who are instructional faculty as defined by the AAUP, she added.

Beeson said that when she started tracking faculty gender issues for the Provost’s office a couple years ago, she began her comparison to five years earlier, that is, to 1998-99.

“Based on the data that universities report to the AAUP on the numbers of women faculty, in 1998-99, we were around the median of all AAU publics, when 25.3 percent of our faculty were women,” Beeson said.

“Since then, two notable things happened. The first is that all AAU public schools now have more women as a percent of their faculty. The median was about 25 percent, but it has grown to 32-33 percent. And beyond that, Pitt has gone way up to the second highest among the AAU publics, with more than a third of our faculty — more than 37 percent — being women.”

Last fall, the AAUP reported comparative data on the numbers of women who are tenured or in the tenure stream.

“That’s something they hadn’t reported before, so I only have this year’s data,” Beeson noted. Pitt, with women representing 24 percent of its tenured faculty, is a little below the median of the 34 AAU publics, she said. “But there is a fairly flat distribution in this range.”

Pitt ranks 5th among the peer group for number of women in the tenure stream with 42 percent, she pointed out.

“If you look at overall faculty from 1998 until now, we’ve moved from the middle to near the top in hiring more women as a percentage of our faculty. If you look at tenured versus tenure-track, we’re at the median for the tenured, but near the top for tenure-track. That’s something we have to keep in mind when we look at salary data,” Beeson maintained.

Part of her task is to break the academic ranks down to examine the data more carefully, she said.

“The first thing we said was, ‘Part of this salary differential has to do with rank. Since most of the hiring of women has been at the assistant rank more recently, let’s control for the ranks.’”

According to the AAUP report issued last fall, for Pitt’s full professors, women on average earned 87.9 percent of what their male counterparts earn. For associate professors, women earned 92.4 percent of men’s wages; 88.2 percent compared to men for assistant professors, and 74.8 percent compared to men’s salaries for all faculty.

“So a lot of that 25 percent difference in salary overall can be attributed to the distribution of men and women across ranks,” Beeson said.

“The second thing is those data published by the AAUP included data for tenure/tenure-stream and non-tenured faculty. If you’re not in the tenure stream, you have a very different job than those in the tenure stream and you would expect the salaries to be different because the jobs are so different. By taking out the non-tenure stream faculty in these ranks, that also closed the gap a little, from 88 percent ratio to 89.9 for professors, 94.5 percent for associate professors and 91 percent for assistants,” she said.

A third factor to consider, Beeson said, “The AAUP report lumps faculty together by rank regardless of discipline. So the next thing we did was to control as much as we could by school,” she said. “We calculated for each of the schools, and each of the divisions within Arts and Sciences, the ratio of the average woman’s salary to the average man’s salary by rank. For example, full professors who are women in the business school relative to the average salary of full professors who are men in the business school, and so on. Then we added those up, using as weights the percentage of the overall University faculty represented in each school or division.”

That chipped away another 2-2.5 percent of the gender salary gap for the ranks of professor and associate professor, Beeson said. “For full professors, the differential was due to the fact that we have men and women full professors distributed unequally among the schools, and the higher salary schools tend to have more men. For associate professors, once we take into account the tenure and tenure-stream faculty and controlled for the schools, we explained almost all of the differential. Almost all of it is related to those two factors, and the ratio is 97.1 percent.”

Both controlling by tenure/tenure stream and by school still did not explain why women assistant professors made only 94 percent of their male counterparts, Beeson said.

“We can explain a lot of the differential but not all of it. At 94 percent it’s sort of puzzling, because you would think it would be closer for assistant professors, both in the same school, both recent hires. Especially since when we were hiring we wanted to make sure there weren’t any unintended biases. That’s where this issue of hiring so many women in recent years came into play,” she maintained.

By breaking down assistant professors into those who have been at Pitt three or fewer years and those who have been here four-six years, the salary gaps almost disappear, she noted. “When we break it down like that, the percentage of those hired within three years is 97 percent, controlling again for school and tenure status; for those who have been here four-six years, it’s 99.97 percent.”

Both of the assistant professor categories are higher than the overall 94 percent, Beeson noted. “That’s because there’s a different salary average for this group of zero-three years than for that group of four-six years. So the overall is lower because there are more people in the zero-three group, the group that has a lower salary,” she said.

“Overall, things are a lot closer than what we thought they were.”

Beeson also creates a profile of every school and each regional campus that includes listing the gender and salary of every faculty member and where they stand among their peers by rank.

“There are variations across schools and we bring these data to the attention of the deans to show them what they’re averages are and what their distribution is like. In some schools men will tend to be at the high and low extremes [of salary] and the women are in the middle, but in another, you’ll see a clump of women here, a clump there,” she said.

“My job is to encourage the deans to think about the data; there aren’t too many surprises because the deans know what’s going on in their school. Periodically, you need to look at this distribution. Does this distribution make sense to you, given what you know about your individual faculty members: Who has written a book, who’s bringing in grants, and so on? This is a systematic look at that,” Beeson said.

—Peter Hart

Leave a Reply