When William A. Niskanen, President Reagan's senior economic adviser, earlier this month described the issue of comparable pay for comparable work as a ''medieval concept'' whose ''time has passed,'' it no doubt annoyed some of the women in his audience.
Women's wages average only 59 percent of the average man's wage, the Census Bureau says, the rough level for 60 years. Feminists see ''comparable worth'' legislation along the lines of civil rights laws as a desirable goal. They figure jobs predominantly held by women, such as secretarial work, could be compared scientifically in regard to training, skill, and responsibility to jobs held mostly by men, such as truck driving, which pay more. Then the wage levels of the ''female'' jobs could be raised to a comparable worth.
Presidential candidate Walter Mondale has embraced the concept, but it has been rejected by Republicans.
Some light on the dispute is cast by two new studies:
1. One by the Rand Corporation finds that the gap between men's and women's wages ''is narrowing and is not as immutable as it appears.''
In fact, the report notes, women's pay jumped from 60 percent of men's in 1980 to 64 percent in 1983. This, say economists James P. Smith and Michael P. Ward, is the ''largest and swiftest'' gain of the century. By the year 2000, they predict, that proportion will have risen to at least 74 percent.
2. Another study, by two University of Michigan economists, reckons that a comprehensive comparable-worth policy would cut the disparity between male and female wages by only 4 to 6 percent.
In other words, if the disparity in weekly wages between a traditionally male job, say engineering, and a traditionally female job, say art historian, was $ 100, comparable-worth rules based on nonsex differences in education, experience , race, region of the country, hours worked, industry, marital status, number of children, union membership, and so on would reduce the $100 difference by only $ 4 to $6.
''These findings may disappoint comparable-worth advocates who expect comparable worth to achieve drastic changes in the US relative wage structure,'' Michigan Profs. George Johnson and Gary Solon note. ''They may soothe the fears of comparable-worth opponents who view it as the worst idea since minimum-wage legislation.''
The Rand study, entitled ''Women's Wages and Work in the Twentieth Century,'' says the improvements in women's relative wages are due to solidly improving market skills - particularly those that are influenced by education and work experience - rather than to legislation, affirmative-action programs, or political pressures.
Mr. Smith acknowledges there is a male-female wage gap. If years of education and years of job experience were figured in, it would raise women's wages to around 80 percent of that of men, he reckons.
The remaining gap would be due to discrimination, differences in quality of experience, extra competition for female jobs, and many other factors.
The Rand study does not go into that issue. It mainly examines the progress in women's wages. It says the present gap suggests ''an inflexible labor market that has failed to reward the obviously increasing skill of women as more of them have entered the labor market and more have stayed in it.''
The report says this conclusion is false. The Census statistics do not take account of the constantly changing female labor force. Female job seekers, usually with less education and fewer years of experience than the typical woman already employed, have been flooding into the labor force during the past 60 years.
The researchers note that this trend drags down the average wage of all female workers. Since more than 95 percent of men aged 20 to 64 are usually in the labor force, the character of the male labor force has changed far less.
Their average wages have not been pulled down by an unusual surge of entry-level wages at any time.
In fact, the report finds, ''women's wages are not 59 percent of those of men - they are even lower.''
To reach this conclusion, Mr. Smith and Mr. Ward assume that, like men, virtually all women have been in the labor market for most of the century.
To do that, they assign a wage to women who have not yet entered the work force, taking account of their education and work experience. Then they use those figures together with those of women actually earning wages to establish the ''average wage'' of all women.
This may sound like some hocus-pocus, but it eliminates the distortion caused by the constant shifts in the female labor force.
Using the Census Bureau calculation would show that in 1920 a working woman was making more money relative to men than her 1980 counterpart - 63 percent compared with 60 percent.
But those 1920 working women had a higher education and longer job tenure than the typical working woman in 1980, the authors say.
By this calculation, the ''average wage'' of all women relative to men was 43 percent in 1920 (when only 20 percent of women were in the labor market) to 48 percent in 1950 and 53 percent in 1980.
Across those years, the study finds, women's wages grew 20 percent faster than men's.
The study predicts that, since the education and experience of the female work force has been rising rapidly for many years, the average woman's wage should reach at least 74 percent of the average man's wage by 2000.
Some state governments are experimenting, under court order in at least one case, with comparable-worth provisions for their own employees.
Unless all factors accounting for wage differences are taken into account (such as the cleanliness or boring nature of a job), there will be long lines for some jobs, shortages of workers in other areas.
In the meantime, the market will be correcting the female-male wage gap faster than legislation and regulation can do the job.