AT THE big annual shindig of American economists, held last week in Philadelphia, one line of research stood out: women in economics. From the low share of female graduates in economics, to the “leaky pipeline” that brings few women all the way to the top ranks in academia, there was no shortage of problems to prompt scientific soul-searching. In that vein, a study by Erin Hengel of the University of Liverpool presented at the conference points to one possible reason why female economists publish fewer papers than male ones: journal editors and reviewers may be holding female authors to a higher standard than male ones.
As Ms Hengel notes, Econometrica, a top journal in the field, instructs authors to write “crisply but clearly” and to take “the extra effort involved in revising and reworking the manuscript until it will be clear to most if not all of our readers”. Clear writing, revising and reworking take time—which means that authors who spend more time revising drafts of their old research have less time for new research.
Ms Hengel analysed the readability of the abstracts of nearly 2,000 papers published between 1950 and 2015 in the top four economics journals. (Other research has found that the writing style in a paper’s abstract is consistent with that in other parts of the paper.) She used the Flesch Reading Ease score, which combines the frequency of easy, shorter words and sentence length to measure how easy a text is to read. The score ranges from zero to 100, with a higher score indicating an easier read.
Papers by women scored, on average, higher than those by men. That was true for both first drafts and final, published versions. But the trajectories of men’s and women’s writing styles diverged.The draft of the very first paper published by a female author was just as readable as the draft of a man’s first paper. Women’s papers, however, became more readable as their careers progressed. No such trends were seen for men.
Women, it seems, had to improve their drafts to get their research published, whereas men did not. Ms Hengel was able to rule out several alternative explanations. It was not because women were writing on easier topics. Nor was the result caused by journal editors becoming more demanding around the same time as more women started publishing papers.
If women are, indeed, held to a higher standard, the review process for their papers would take longer than that for men. Fortunately, one of the journals included in the study happened to have data on this. Analysing these showed that, on average, the review process for papers written only by women took six months longer than that for papers written only by men.