dalejn
commented
1 year ago Hello, we would like to keep this as a space to discuss issues related to code and Binder usage. I take your questions as expressing skeptical opinions about whether these ideas and practices are pointless, misguided, asymmetric, performative, and harmful.
Obviously from having contributed to this work, I disagree. There's a ton of work challenging the concerns you've raised, and I regret I won't be able to do them justice in a GitHub issue. But I may be able to encapsulate a common experience I have of being cited or following a trail of citations in the literature via meme—an experience which disabuses me of the notion that citations are purely scientific, that it is straightforward to objectively identify the "better" citation w.r.t quality, or that people are incentivized to regularly make a concerted effort to do so.
Do you have any scientific evidence to exclude the possibility that some papers are more cited because they are better science? Also hard to imagine anyone considering the gender of author(s) when deciding to cite or not a reference (checking the author to verify their gender).
If you're interested, see the listed peer-reviewed scientific articles in the README. A main finding was that citation patterns differ from what we'd expect from several models of citations where gender doesn't factor in (i.e. random draws) or doesn't factor in explicitly (i.e. draws constructed from data about actual co-citations in scientific social networks). This model also considered conventional metrics of article or journal quality and could not explain away the differences. Lastly, note that given independently collected data and repeated findings on gender differences in citations, the possibility that you raise implies that men do better science than women, which does not make sense to me from the outset.
More broadly, you may be interested in the science of homophily, rich-get-richer effects, implicit association tests, and/or applications of a double-blind approach and its pros/cons. These areas of work probe social reasons for why some work accumulates more citations. Those reasons are quite independent from scientific quality, but rather relate to convenience, name recognition (like a Baader-Meinhof frequency illusion), implicitly learned stereotypes, nepotism, and gender differences in incentives/expectations/standards/rewards.
Also, if my bibliography tells me I have an overrepresentation of female authors (or viceversa), what shall I do to "fix" it? Replace the papers authored by the other gender to meet the "quota"? What if there are no seminal papers written by a specific gender on the issues I research?
Political correctness overriding science?
You are right that fields differ in their demographics and the standards or values of one field do not smoothly generalize to others, and I share your concerns about quotas and tokenism. However some of your skepticism seems rooted in trust of the scientific/publishing process and in belief of meritocracy. With that in mind, you may also be interested in work that critiques meritocracy, the history of science for your fields of interest, or economic research on the validity of quality metrics that become incentives in themselves.
There are many motivations for why someone cites work: relevance, praising, admonishing, thanking, paying homage, displaying literacy/expertise/familiarity of popular/classic/old hits, respecting historically notable firsts, identifying a collaborative niche with like-minded peers, and competition. Notice that many of these reasons are both scientific and social/historical. Science is a social enterprise advanced by people, and citations are one of the major currencies that advance people's careers and status. Politics is about the social relations that exchange power and authority, so yes the overarching practice is political. Ultimately the shared aim is to improve the scope, reach, relevance, and value of science as an activity, as a career, and as governed by institutions. I suspect our disagreement is in how and where we see politics harming scientific progress, and what asymmetries are tolerable.
Given science is a social and political endeavor, we hope making people aware of constrained patterns of reading and citations due to social factors will encourage us to be more deliberate in how we engage with the literature and improve the quality of citations.
Do you have any scientific evidence to exclude the possibility that some papers are more cited because they are better science? Also hard to imagine anyone considering the gender of author(s) when deciding to cite or not a reference (checking the author to verify their gender).
Also, if my bibliography tells me I have an overrepresentation of female authors (or viceversa), what shall I do to "fix" it? Replace the papers authored by the other gender to meet the "quota"? What if there are no seminal papers written by a specific gender on the issues I research?
Political correctness overriding science?