Abstract
This short essay argues for an expanded conception of publication bias. In addition to consideringthe selective publication of results, I argue that we need to also consider the selective publicationof epistemic by-products—observations and knowledge that scientists accumulate incidentallyin the process of carrying out their work. There are three reasons why we should be concernedabout the exclusion of epistemic by-products from the published literature: first, because theyplay an important role in robust replication attempts; second, because their absence can resultin misplaced scientific certainty; and third, because they contribute to a holistic understandingof natural phenomena. However, identifying and addressing publication bias against epistemicby-products and other undervalued forms of knowledge is more difficult than identifying biasagainst quantitative findings. I argue that scientific pluralism and making data publicly accessibleare two potential remedies for addressing this form of publication bias.
Highlights
Publication bias has long been recognized as a problem in the sciences (Dickersin, 1990; Dickersin, Kay et al, 1987; Rosenthal, 1979)
Expanding the notion of publication bias to include these processes allows for a deeper understanding of how users of the scientific literature might arrive at a false sense of certainty, and offers insight into how gaps form between individuals’ understanding of natural phenomena
This is not to say that the published literature was unimportant—when describing their views on behavior, researchers often quoted data from human behavior genetics studies showing that genetic and environmental factors both mattered in developing alcohol use disorders (Prescott & Kendler, 1999), or mouse studies showing that the genetic makeup of the mice influenced how much they would drink (Rhodes et al, 2007)
Summary
Publication bias ( known as selective publication, or the “file drawer” problem) has long been recognized as a problem in the sciences (Dickersin, 1990; Dickersin, Kay et al, 1987; Rosenthal, 1979). If researchers decide whether to publish based on a property of the result—say, whether the result reaches commonly-used thresholds of statistical significance or supports the researcher’s hypothesis— the subset of study results that appears in the published literature will be biased In this short essay, I argue for an expanded conception of publication bias. Researchers typically choose not to publish these findings not because they fail to reach statistical significance, but because they were never intended to be knowledge in the first place This implicit distinction between the entities that scientists consider to be scientific findings and those that they consider to be anecdata, tacit knowledge, or lab lore acts as an additional filter that prevents some types of knowledge from circulating widely. Expanding the notion of publication bias to include these processes allows for a deeper understanding of how users of the scientific literature might arrive at a false sense of certainty, and offers insight into how gaps form between individuals’ understanding of natural phenomena
Sign-in/Register to view highlights & summary Sign-in/Register to access full text options 
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a callDisclaimer: All third-party content on this website/platform is and will remain the property of their respective owners and is provided on "as is" basis without any warranties, express or implied. Use of third-party content does not indicate any affiliation, sponsorship with or endorsement by them. Any references to third-party content is to identify the corresponding services and shall be considered fair use under The CopyrightLaw.
