Study: Male Scientists Publish More, Women Cited More

Why do women scientists publish fewer papers than men? Sociologists of science have been pondering that question since the late 1970s, when they documented the fact that women scientists publish about half as many articles as men do. And a recent study has uncovered a surprising new piece of the "productivity puzzle." Women biochemists publish less than men do, the study acknowledges, but the average paper written by a woman is cited more often than the average paper written by a man, according to J. Scott Long, a sociologist at Indiana University. "What this data is saying is that women may not publish as much [as men], but the work they do publish is utilized by the scientific community more," says Long, author of the study, which was published in the journal Social Forces (71[1]:159-78, September 1992). Long analyzed the publication and citation records of 550 men and 600 women biochemists who got their Ph.D.'s between 1950 and 1963. He was following in the footsteps of sociologists Jonathan Cole and Harriet Zuckerman of Columbia University, who first observed the gender gap in scientific output and christened it the "productivity puzzle," because the reasons for it remained elusive. Long's study was unusual in that he followed his subjects for 17 years, and included 80 percent of those who earned Ph.D.'s in biochemistry during the course of his study years. Thus, his data represent "average" scientists as well as the elite. Like Cole, Zuckerman, and others before him, Long found that the female biochemists published much less than the males, especially in midcareer. During the ninth year after their Ph.D., for example, the men published nearly twice as many papers as the women. By year 17, the gap had moderated somewhat, but men were still publishing nearly 60 percent more. Among women, the average publication rate was lowered by a large contingent who simply didn't publish. During year 10, for example, 40 percent of the women published no articles at all, compared with about 20 percent of men. But even top producers reflected the trend: During year 17, the top 5 percent of women averaged 3.5 or more papers. The top 5 percent of men published five or more papers. Since men had more papers, men also had more overall citations. But women had more citations per paper. By the end of the 17 years, the average paper by a woman was cited 1.5 times more often than the average paper by a man. "This finding surprised me so much I went back and verified the raw data to make sure something wasn't systematically distorting the results," recalls Long. But there was no mistake: In the three years following publication, the average paper by a woman was cited between nine and 13 times. The average paper by a man (as indicated in the accompanying graph) received seven to nine citations in those three years. Long found that men and women had identical patterns of collaboration, so working with others was not skewing the productivity measures. There was one exception: Women were much more likely than men to be coauthors with a spouse. So it seems that women write fewer but higher-impact papers. But why? One possibility is that women are "less strategic" in their approach to publication, say Harvard University researchers Gerhard Sonnert and Gerald Holton, who recently completed a study of 800 male and female scientists who received prestigious postdoctoral fellowships. One portion of their study compared the citation rates of male and female biologists. This study focused on top-ranked researchers, but the results were the same as Long's: Papers by women were cited more often than those by men. "We have the hypothesis that women don't try to squeeze the maximum number of papers out of a given amount of research, and so they might publish fewer but more substantial papers," says sociologist Sonnert. "They try to be very thorough and produce unassailable stuff." When Sonnert and Holton also interviewed their subjects about the characteristics of outstanding research, they found some gender differences. For example, 36 percent of the women said good research was thorough and comprehensive, while only 20 percent of the men mentioned those characteristics. Asked about their approach to publication, individual women scientists offered anecdotes that in some measure support Sonnert's hypothesis. "My own impression is that women tend to be a bit more insecure about how the work will be received," says Susan Gerbi, who is president of the American Society for Cell Biology, a professor at Brown University, and an author of more than 50 publications. "They want an airtight story, a more complete story, before they go public with it, whether in a paper or in a seminar. Then once they do go public with it, it's pretty solid." For example, Harvard chemist Cynthia Friend has written about 70 papers on surface chemistry, and is conscious of the pressure to publish. Nevertheless, when probing the orientation of a molecule on a surface, Friend and her coworkers found that two experiments gave slightly inconsistent results--and she held up the paper to do more experiments. She wanted to nail down the result, although, she says, many other researchers would have simply published anyway. Also at Harvard, anthropologist Maryellen Ruvolo focuses on a specific gene, which she sequences in different primates in order to understand the evolutionary relationships among them. She has a male colleague who does similar work on a different gene and "publishes a paper every time he sequences his gene in one or two new species," Ruvolo says. By contrast, Ruvolo just submitted two papers that included data on six and 11 new gene sequences, respectively. Biochemist Caro-Beth Stewart of the State University of New York, Albany, noted a difference in attitude when she and a male colleague were going over page proofs. He wondered why she spent so much time on the task, since proofs usually took him only an hour or so to scan. "I've never checked page proofs in an hour in my life," says Stewart, who says she felt compelled to double- check and update references. But waiting for more data and double-checking references may not be savvy strategy for climbing the academic ladder. After all, advancement in science depends to some degree on number of publications. If women do adopt a "less careerist" approach, as Holton describes it, they may not be doing themselves any favors. Indeed, women may publish less because they aren't getting good advice--they're not getting the message that numbers count, suggests Friend. She and other scientists say students--both male and female--tend to be idealistic and eager to write the perfect paper. So they need guidance in order to know when to publish. Since several studies have already shown that women are less likely than men to have a mentor to give such guidance, Friend's theory makes some sense, agrees Sonnert. Still, many women scientists--including all of those quoted previously--point out that when it comes to being a perfectionist, there's plenty of male specimens around. For example, it was Stewart's graduate school adviser--a man--who taught her to check proofs carefully. Different subfields may have different standards for publishing, and there are so few women in some fields that generalizations can be "dangerous," says Friend. "Although it does seem to me that many women wait until they're absolutely sure of something, I don't really know that it's because they're female. I know men who do that, too," says endocrinologist Neena Schwartz of Northwestern University, who has about 100 publications to her credit. That's why the work of Sonnert, Holton, and Long is so useful, women scientists say: Their studies move the debate beyond anecdote. But even Long isn't ready to claim that he understands exactly why women lag in publications and lead in citations. A constellation of factors may be responsible, he says. For example, senior scientists may be coauthors on a larger number of "bread and butter" papers that help launch the careers of grad students and postdocs. Women are less likely to be lab directors and so are perhaps less likely to churn out smaller papers. "We can't be sure about all the reasons why," says Long. "But the important thing is that this is occurring. The work of females is cited more per paper than that of men. Their work is not marginal in any way--it's being used by other scientists as much or more than men's work."

Study: Male Scientists Publish More, Women Cited More

Interested in reading more?

Become a Member of