Back

Reading Outside the Lines: A Systematic Approach for Detecting Bias in Scientific Communication

Osborne, M.; McKenna, T. J.; Crawford, F.; Rueger, T.; Shah, B.; Gott, M. R.; Labadorf, A. T.

2024-06-12 scientific communication and education
10.1101/2024.06.12.598649 bioRxiv
Show abstract

Consciousness of the social impact of science and the potential biases of its authors is critical to understanding, interpreting, and using scientific findings responsibly. This is especially true for sciences concerned with human health and behavior, where societal and unconscious biases may reinforce existing inequities and discriminatory practices. Considering backgrounds and biases, we may notice bias influencing scientists methodological choices and conclusions, even when a work is otherwise scientifically sound. To this end, we created the pedagogical tool Finding inEquity in LIterature and eXperimentation (FELIX). FELIX is a tool that systematizes the detection of bias and subjectivity in scientific communications by using a three-phase progression of (i) Annotation, (ii) Analysis and (iii) Synthesis, where students form a unified argument about the text with a focus on its relationship to social or ethical context. Results from a mixed methods approach indicated the efficacy of our approach in supporting student learning related to reading comprehension, critical thinking skills, and in understanding the social and ethical implications of the research they were reading. We put forward FELIX as a universal method for training students in the reading of scientific communications and as a tool for addressing systemic inequities in science and science education.

Matching journals

The top 3 journals account for 50% of the predicted probability mass.

50% of probability mass above

"Similar papers" are the closest papers from that journal in the model's embedding space. They show what the match is built on, but the ranking comes mostly from a classifier over the whole training set, not from these examples alone.