Science Studies Colloquium

Every day our life is made easier by efficient algorithms that search and rank scientific information. Yet, these algorithms have an issue: they are trained on citation data that is ingrained with human biases. Therefore the output is inherently biased too, creating inequalities, raising concerns of discrimination, and harming economic growth. In this talk, I focus on recent quantitative efforts in science studies to (1) uncover bias mechanisms in science, (2) use this knowledge about biases to uncover inequalities in the scientific enterprise, and (3) create fair metrics and algorithms.

The Open Access movement promised to increase knowledge sharing within and beyond academia, sidestepping issues of structural inequality. In spite of this initial hope, emerging evidence shows that the model of author facing charges is erecting a new barrier, threatening equity in who can contribute to the scientific record. This talk will investigate the role of institutional resources in shaping publication outcomes across fields and countries, and discuss the wider implications of current threats to an equitable implementation of Open Research.

The scientific community calls on senior women to help mentoring the next generation of scientists. Yet research indicates that particularly senior women often get undervalued in academia, from hitting glass ceilings to receiving less recognition for accomplishments than male peers. We examine whether these gender differences in evaluations of senior scientists, who are poised to serve as mentors, also affect the evaluation of protégés and their work. Identifying mentors of 4,556 scientists with competitive early career funding from the U.S. National Institutes of Health (NIH), we document a citation discount of 10% on the average paper published by women- relative to men-mentored protégés. Using data on both the publications of the mentored protégés as well as the citing articles, we distinguish supply-side (i.e., scientists’ offerings) from demand-side (i.e., actions by the scientific community) explanations. Supply-side factors appear to account for about 40% of the citation discount. The remainder is explained by men citing protégés of women less often than protégés of men. This gender-biased treatment on the demand-side particularly afflicts work in the most impactful research areas that draw the most citations. Although both men and women mentors spur protégés to producing their very best work, the science community gives women-mentored protégés less recognition for it. These findings raise concerns about an unbiased discourse on the best scientific contributions and about systemic limitations to women serving as mentors as a means to closing gender gaps in science.

We examined male-female collaboration practices of all internationally visible (25,000) Polish university professors based on their 160,000 Scopus-indexed publications. We merged a national registry of 100,000 scientists (with full administrative and biographical data) with the Scopus publication database. We examined the propensity to conduct same-sex collaboration across male-dominated, female-dominated, and gender-balanced disciplines. Across all age-groups and all academic positions, the majority of male scientists collaborate solely with men. The majority of female scientists, in contrast, do not collaborate with women at all. The gender homophily principle (i.e. collaborating predominantly with scientists of the same sex) works powerfully for male scientists – but does not seem to work for female scientists. We examined the propensity to engage in same-sex collaboration across several new dimensions. This research goes beyond traditional bibliometric studies of gender-based homophily in research collaboration by combining the data routinely inaccessible to large-scale studies (such as the biological age of all scientists, and the stages of their academic careers) and the data routinely accessible in bibliometric studies, such as journal prestige, academic disciplines, and institutional type. We discuss practical implications of our research for academic careers.

The quantification and oversimplification of academic success is harming scholarly communities in all disciplines. Scholarly metrics, such as the h-index or impact factor, are widely applied in academic tenure and funding decisions, but often inappropriately. This has created publication pressure, leading to a range of adverse effects and even scientific misconduct. To improve the understanding and appropriate use of scholarly metrics, the concept of metrics literacies has been proposed as an integrated set of competencies, dispositions, and knowledge that empower individuals to recognize, interpret, critically assess, and effectively and ethically use scholarly metrics. Attempts to address the lack of metrics literacies in the wider academic community, such as the Leiden Manifesto and The San Francisco Declaration on Research Assessment (DORA) or books such as Becoming Metric-Wise: A Bibliometric Guide for Researchers and Measuring Research are published as textual documents. As scholars are already overwhelmed by too much to read, the question arises why not to provide more efficient and effective multimedia open educational resources (OERs) instead of text? The Metrics Literacies project aims to aims to educate researchers and research administrators by developing, testing and disseminating educational videos to reduce the misuse of metrics in academia.

Reputation and recognition gained through publications has been a crucial merit for career advancement in science since the birth of the research university in the late 18th century. The ability to publish research is instrumental both for gaining recognition within a specific field of research, and for the possibility of getting a permanent position in academia. Through a content analysis of assessment reports for professorial candidates in biomedicine, economics and history this presentation discusses how publication oeuvres are compared and evaluated. Specifically the analysis focus on how referees use bibliometric measures and other types of 'judgment devices' to assess and rank candidates. Moreover, specific attention is given to how evaluators mobilize the idea of 'career trajectories' to weave together disparate bits of evidence extracted from the bylines of researchers' CVs.

The presentation will discuss two issues of hijacked journals. The first shows how cost-reduction strategies of hijacked journals expansion allow to detect these fraudulent publishers. The second issue is connected with the belief in the literature that naïve authors submit papers to hijacked journals. The presentation will demonstrate that there is a group of dishonest authors who are attracted by the fast publication with no peer review process that is offered by hijacked journals.