Surveys abound showing that women in academe (and the rest of society) earn less than men. Likewise theories abound for why this is the case, so many years after it ceased to be acceptable for deans (or other bosses) to automatically assume a woman could make do with less.

A scholar at the University of Iowa who has been mining national data presented his latest findings Monday at the annual meeting of the American Educational Research Association. The results in short say that -- even using the most sophisticated possible approach to take into consideration non-sexist reasons for pay differentials -- a pay gap remains, based on gender. And while this can't be definitively tied to sexism, there aren't a lot of likely alternative explanations.

But the study also found that some of the explanations that do exist -- in particular based on disciplines and the types of institutions where women are more likely to find jobs -- suggest that the salary gaps may be here to stay, unless higher education thinks very differently about reward structures.

"Higher education in general is going to have to consider how we reward people, and how our awards differentially affect men and women," said Paul D. Umbach, assistant professor of education at Iowa.

In fact, Umbach's analysis finds a greater share of the salary gap in these general reward policies than in the unexplained category that could be blatant sexism. That's why he said Monday that women face "a double hit" in what they earn -- an average of $3,200 when he has controlled for all factors. Generally, controlling for all factors, he found a gap of 4 percent remained between the salaries of men and women. Controlling only for discipline and institution type, the gap is larger (14 percent) and part of Umbach's concern is that the larger gap may be the one faced by most women.

Generally, Umbach's research has not been quick to proclaim sexism as the cause for salary gaps. He has devoted much of his work to exploring the various explanations for salary gaps that may or may not be directly related to gender and that may or may not be fair. His latest analysis is based on data from 472 four-year colleges and universities, broken down not only by gender but by 87 disciplines. Nearly 8,000 faculty members were studied.

In controlling for all possible factors, Umbach said he wanted to look at the interrelationship of discipline and institution. He said that many salary gap studies are flawed in that they focus on institutions only or disciplines only, when the two are related. In addition, he applied an approach in which faculty members were ranked on productivity and other factors, so that the final comparison (in which only a 4 percent gap remained) compared faculty members on equal ground with respect to those factors and numerous others (seniority, classes taught, etc.).

One of the theories Umbach said he wanted to test was whether -- as departments attract more female faculty members -- their relative pay level goes down. Here, he said that his findings were ambiguous. He did find a clear correlation -- disciplines that have more women have lower average salaries. The complication, he said, is that he found other factors as well. Those disciplines also tend to be teaching-oriented disciplines. Similarly women were disproportionately employed at teaching-oriented institutions, which also pay less. So professors who are women, teach in a field that cares about teaching and work at a college that really cares about teaching face a "triple hit" on salary, he said, "and it adds up to real money."

It's not sexism alone at play, he said, because men who teach in those departments and at those institutions also earn less than men elsewhere (although the teaching-oriented men still earn 4 percent more than the comparable women).

In terms of what to do about this, Umbach acknowledged in an interview that there weren't easy answers. "There are pure market forces we can't ignore," he said. And that explains in part why universities tend to favor professors in fields where they might be recruited outside of academe or have the potential to attract outside support.

But he said that the study suggests that, at the very least, colleges need to continue to study their salary gaps, and not to assume that this is a problem that has been solved.

And if colleges care about the gender gap, it may be time to question assumptions about why people in some disciplines earn more. "Is it fair to reward people who can earn grants from outside over people for whom there aren't grants?" Added Umbach: "At the very least this is something to observe. It perpetuates inequities."