Gender in the Innovation Economy
Friday, Jan. 3, 2020 8:00 AM - 10:00 AM (PST)
- Chair: Julian Kolev, Southern Methodist University
Social Influence among Experts: Field Experimental Evidence from Peer Review
AbstractExpert committees are often considered the gold standard of decision-making, but the quality of their decisions depends crucially on how members influence each other’s opinions. We use a field experiment in scientific peer review to measure experts’ susceptibility to social influence, and identify two novel mechanisms through which heterogeneity in susceptibility can bias outcomes. We exposed 247 faculty members at seven U.S. medical schools reviewing biomedical research proposals to (artificial) scores from other reviews, manipulating both the discipline and direction of those scores. Reviewers updated 47% of the time, but with significant heterogeneity by gender, academic status, and score direction. We find that even in a completely anonymous setting, women scholars updated their scores 13% more than men, and even more so when they worked in male-dominated fields, while very highly cited “superstar” reviewers updated 24% less than others. If evaluators tend to champion “their” candidates, lower updating by high status evaluators advantages their candidates. We also find that lower scores were “sticky” and updated (upward) 38% less than medium and high scores. This asymmetry can favor conservative proposals, as proposals’ demerits loom larger than merits. Our results indicate that expert group deliberation processes that are widespread throughout the economy are subject to biases that require significant attention by scholars and practitioners.
Is Blinded Review Enough? How Gendered Outcomes Arise Even under Anonymous Evaluation
AbstractBlinded review is a direct and increasingly popular approach to reducing the impact of bias, yet its effectiveness is not fully understood. We take advantage of the blinded-review process to document the drivers of gender inclusion in a unique setting: innovative research grant proposals submitted to the Gates Foundation from 2008-2017. Despite blinded review, we find that female applicants receive significantly lower scores, which cannot be explained by ex-ante measures of applicant quality or applicants’ choice of topic. By contrast, we show that the gender score gap is fully mediated after controlling for text-based measures of proposals’ titles and descriptions, with female applicants tending to use narrow, topic-specific words that tend to receive lower reviewer scores. Importantly, the text-based measures that predict higher reviewer scores do not also predict higher ex-post innovative performance. Our results reveal that gender differences in writing and communication are a significant contributor to gender disparities in the evaluation and funding of science and innovation.
Driving Inventor Inclusivity in the Innovation Economy: What Is the Role of Universities and Their Top Inventors as Catalysts for Change?
AbstractDespite growing STEM participation, the contribution of women to patenting is persistently low. We explore the drivers of gender inclusion in the innovation economy by examining U.S. inventors (patentees) at leading universities. If universities, early in women’s careers, successfully promote female inventors they will amplify inclusion in the economy by facilitating that women become productive inventors throughout their careers. Analyzing patents by gender, we build novel measures of female inventor inclusivity for universities and their top inventors. Our analysis shows that universities have greater inventor inclusivity than the US economy. However, there is an inventor gender gap in universities that it is not just a problem of STEM supply: female inventor Inclusion is not rising as fast as women in STEM PhDs. Within universities, top inventors are more inclusive. Importantly, female top inventors are a key catalyst for change in facilitating the inclusion of women in the innovation economy.
Female Inventors and Inventions
AbstractDoes increasing the number of female medical researchers produce greater medical advances for women? In this paper, we investigate if the gender of inventors shapes their types of inventions. Using data on the universe of US biomedical patents, we find that patents with women inventors are significantly more likely to focus on female diseases and conditions. Consistent with the idea of women researchers choosing to innovate for women, we find stronger effects when the lead inventor on the patent is a woman. Women-led research teams are 22 percent more likely to focus on female health outcomes. This link between the gender focus of the scientist and the type of invention, in combination with the rise of women inventors, appears to have shifted the direction of innovation towards female conditions and diseases over the last four decades. Our findings suggest that the demography of inventors matters not just for who invents but also for what is invented.
Texas A&M University
Massachusetts Institute of Technology
Michigan State University
Massachusetts Institute of Technology
- J1 - Demographic Economics
- O3 - Innovation; Research and Development; Technological Change; Intellectual Property Rights