Thammasat University students interested in business, economics, science, sociology, gender studies, and related subjects may find it useful to participate in a free 22 November Zoom webinar on Cassatts in the Attic: Is there a gender gap in the commercialization of science?

The event, on Wednesday, 22 November 2023 at 8am Bangkok time, is organized by the Centre for Innovation and Entrepreneurship and University of Hong Kong (HKU) Business School’s academic area of Management and Strategy.

Its website explains:

Abstract:

We analyze more than 70 million scientific articles to characterize the gender dynamics of commercializing science. The 15-25% gender gap we report is explained neither by the quality of the science nor its ex-ante commercial potential, and is widest among papers with female last authors (i.e., lab heads) when publishing high-quality science. We verify this in a subset of approximately 30,000 “twin” discoveries, for which papers we hand-code the gender of every PI. Results hold whether we define twins via adjacent co-citation or identical biological sequence and structure.

What drives this gap? Prior literature pointed to supply-side factors including reluctance on the part of women to patent or otherwise engage in commercial activity, but we cannot recover evidence for these. Rather, demand-side factors appear to play a key role. First, no gender gap is apparent when women self-commercialize their own discoveries instead of cooperating with existing firms. Second, the gender gap is larger at firms with a higher percentage of male inventors. Third, an increase in accessibility of scientific articles during the Obama administration does not close but rather widens the gender gap. Fourth, consistent with bias we find that discoveries by women that are cooperatively commercialized are of greater value, both using a crude proxy for “marginal” discoveries as well as when instrumenting for the likelihood of cooperative commercialization.

The title of the presentation refers to a once-overlooked, now highly valued American artist, Mary Cassatt.

The speaker will be Professor Matt Marx who teaches management and organizations at the SC Johnson College of Business, Cornell University, New York, the United States of America.

The TU Library collection includes several books about different aspects of the gender gap in science.

Students are invited to participate at this link:

https://library.tu.ac.th/search/laz/result?si=TOTAL&st=KWRD&q=women+science&x=0&y=0

Professor Marx’s research paper coauthored with Dr. Marlène Koffi, Faculty Research Fellow at the University of Toronto, Canada, has been posted on the website of the National Bureau of Economic Research, Cambridge, Massachusetts, USA:

We analyze commercialization among all published articles reported by the Microsoft Academic Graph (MAG) from 1800-2020. Automatically classifying the gender of the authors via forenames, and hand-coding gender for a substantial subsample, we explore the gender dynamics involved in the commercialization of these scientific discoveries. Commercialization is captured via a “patent-paper pair” (Ducor 2000) where (a) the patent cites the paper (b) the patent assignee is a firm (c) where the inventors on the patent overlap with the authors of the paper.

We find a double-digit gender gap in commercialization for papers with a woman in the last-author position on the article. Frequently, the last author is the “lab head” or Principal Investigator on the project. The head of the lab may play a critical role in commercialization; therefore, our analyses focus on the gender of the lab head. Because the gender gap in commercialization is larger among discoveries that are more highly cited and with higher commercial potential, we refer to these uncommercialized discoveries as Cassatts in the Attic after Pittsburgh-born painter and printmaker Mary Cassatt. […]

We conclude that the gender gap in commercialization does not reflect the loss of “unimportant” scientific discoveries. If anything, the gap tends to widen as the scientific quality and commercial relevance of scientific discoveries increase. In other words, there are truly valuable contributions by female researchers in science that are not commercialized, i.e., there are genuinely “Cassatts” in the “attic” of science. […]

As more women participate in science, one would hope that any bias against women would begin to dissipate, particularly with the myriad of examples of women doing high-quality research. Thus, we could expect a reduction in the commercialization gap as time passes.[…]

Conclusion

We provide the first large-scale characterization of the gender dynamics underlying the commercialization of science. Analyzing more than 70 million articles from the Microsoft Academic Graph, we find that scientific teams with women—and especially women as last authors (i.e. lab managers)— suffer a commercialization penalty relative to all-male teams. This result is not explained either by the underlying quality of the science or by its commercial potential, which we establish by using “twin” discoveries. We hand-code the gender for every last author in our twins sample to increase confidence in our estimates. The gender gap is not a mechanical result of commercialized articles being cited more often by patents, nor is it explained by levels of commercialization activity at the authors’ institutions or by the authors’ prior commercialization. Finally, our results are robust not only to twin discoveries assembled via adjacent co-citation but also to identical biological sequences and structure.

What mechanisms explain this gap? We explore both supply-side and demand-side factors. We find only weak support for supply-side factors, including the representation of women in a particular scientific field, and also that women have fewer coauthorship ties to industry. Moreover, we establish that the gap does not exist for entrepreneurial self-commercialization via new ventures but only for cooperative commercialization in collaboration with existing firms. One question is whether firms simply pay less attention to women-led science. We leverage a natural experiment that shifted open access to scientific articles. Although open access seems to spur commercialization generally, easier access to scientific articles only exacerbates the gender gap—suggesting possible bias on the part of

firms. One possibility is that firms pay less attention to women-led science because, as has been shown previously, women are less likely to use self-promoting language in their articles. We find that articles that “boast” are more likely to be commercialized, but this benefit is captured only when the lab head is male. Thus it does not appear to be simply due to word choice. It may be that the percentage of male inventors at potential commercialization partners may help to explain the gap. Papers with women as the last authors are less likely to be commercialized in a (potential) patent-paper pair with a higher percentage of male inventors on the patent, suggesting gender homophily in the construction of commercialization teams.

We lastly explore the implications of our findings. The gender gap is the largest among the highest quality papers and those with the most commercial potential, suggesting negative welfare implications.

In other words, the uncommercialized discoveries are not unimportant but may indeed represent “Cassatts” in the “attic” of science. The gap appears most pronounced in the natural sciences, and although the gap has narrowed in the last decade, it grew substantially from the 1980s through the early 2000s.

To summarize, our findings tend to point toward a bias from the firms’ side, not systematic across all fields, but with some stereotypical and homophilic features. The results of this analysis are keen in informing the public debate about the economic and welfare losses of such discriminatory behavior against scientific publications by women.

(All images courtesy of Wikimedia Commons)