A Great Purge — July Retraction

by Jon Scaccia August 3, 2021

Look at this graph! Wow. There’s a huge purge going on at the Journals of Cellular Biochemistry and Physiology. That’s over a hundred articles between them!

We wrote about this extensively back in May, so check out that post to learn more about paper mills, image generation, and what these journals are doing about it.

Interesting Journal Alert!

While going through the retractions list, I came across a journal that I had never heard of, Accountability in Research. To quote from their aims page,

Accountability in Research: Policies and Quality Assurance is devoted to the examination and critical analysis of systems for maximizing integrity in the conduct of research. It provides an interdisciplinary, international forum for the development of ethics, procedures, standards policies, and concepts to encourage the ethical conduct of research and to enhance the validity of research results. 

Like we did with Implementation Science (but without writing a whole manuscript), I pulled all of the articles that were indexed in PubMed. I then analyzed the content to see the types of things that AIR published on.

AIR was running about 20-25 articles per year until 2020 when a huge number got indexed.

Like in other analyses, I made a network plot of the words in the network to see what the high-level themes might be. We see a lot on some of the general research issues, like IRBs, informed consent, retractions, and misconduct.

We also see, even in this journal, that COVID-19 shows up as a keyword. And this makes sense. With the huge amount of resources and attention being devoted to the biggest public health crisis in the past 100 years (at least in the Western world), there is likely a lot of pressure to get information out there.

And when there is pressure and urgency, there can be lapses in research integrity.

Key Topics in AIR.

I ran an LDA to cluster the topics (we talk a lot of the specific steps and methods in the Implementation Science article. The below figure shows the topics by numbers and at least three keywords that are most predictive of that topic. This raises up some really interesting themes not covered in the network plot above, like financial disclosures (topic 3), plagiarism (topic 12), and clinical trial (topic 15).

Word Vectorization

Something that was cut from the Imp. Science article was our work in Word Vectorization. Basically, any words can be reduced to high dimensional vector, which can help to elucidate relationships between terms. I pulled the top forty terms, then plotting them in a UMAP projection with all terms that occurred more than five times.

This shows a pretty sensible distribution. I took a deeper dive to pull other terms that are important for our work. What we found is that equity does not appear at all. While equity has become a bigger issue in things like CBPR (check out our earlier post on the month in community psychology), it looks like it hasn’t filtered through yet to the broader field of scientific accountability.

There’s a lot to say about WEIRD science and the biases built into research inquiry, but that’s a post for another time.

Download all the retractions

Here’s the full list of Retractions

Don’t forget to check out PubTrawlr!

And have you tried out PubTrawlr yet? Our totally free and open search gathers findings from thousands and thousands of journals to give you up-to-date trends and findings.