Friday, July 18, 2014

How biased are you, when you check the literature?

As scientists, we tend to think of ourselves as (at least) a little less biased than the average person. However, we still have to rely on mental shortcuts to classify information regarding its importance, relevance and trustworthiness. These shortcuts allow us to survive among the wealth of information we take from our environment, but are also prone to over-simplification and inevitably lead to bias. When I browse my RSS feeds, or the lists of hits in a PubMed or Web of Science query, I have found myself to be unexplainably biased against papers with only Indian, Persian or Chinese author names, but not against papers which only have Thai, Serbian, or Russian names. It is a "gut" reaction, with no input from the "reflective" portion of my brain. All I can do is be aware of it, and make an extra effort  to engage my thinking brain when this happens. This is the reason I have lately become convinced of the importance of enabling double-blind peer review: of course any author would be able to drop enough hints in the text to make their identity obvious, but forcing the referee to start reading the manuscript without any "mental baggage" (either for or against it)  might help researchers from disadvantaged institutions/countries overcome the biases that now play strongly against them

No comments:

Post a Comment