4 Comments

A partial defense of 'trendy topics'... Aren't 'trendy topics' code for areas of research that still have new, important, low hanging fruits (often because of some technological breakthrough). I don't know how to assess whether editors are biased against untrendy topics. Though, I would anticipate that publications on older, less trendy topics more often inadvertently reinvent the wheel or are far more incremental. I'd be interested in seeing a study that can deconvolve this effect.

Expand full comment
author

But then how would you decouple the study authors' motivations in finding that non-trendy ideas are incremental? The infinite regress of paper analysis.

I think there are lots of reasons for trends to happen in science, hoarding around a new breakthrough being one. I'll describe other mechanisms in the next post. I just want to emphasize that Meehl's point is not about whether these trends are defensible, but that you have to take trends into account when assessing scientific literature.

Expand full comment
May 3Liked by Ben Recht

Wonderful read as always. This sentiment is reflected in the recent changes at eLife, in that papers are not accepted/rejected, but merely passed through review. However, that does come with the condition that papers must first make it past the editors, which brings with it, its own biases.

Expand full comment
author

It's a good step along the way! We can't wholly eliminate the biases Meehl raises, but knowing they exist has to inform our community decision-making about norms.

Expand full comment