AIC and BIC trends

Getting ready for another semester, thinking about how to communicate to students about statistical philosophy while they are trying to learn about the tool box. The mind wanders, and I was curious about the prevalence of Akaike Information Criterion (AIC) versus Bayesian Information Criterion (BIC) in the literature.

Google Scholar search results: Akaike Information Criterion AIC results in more than 118,000 hits; Bayesian Information Criterion BIC returns more than 79,400 hits.

Here’s an NGRAM from Google’s Ngram Viewer:

But of course, both AIC and BIC are dwarfed by the NHST p-value approach…

… and a Google Scholar search on just “p-value” practically broke the search engine at more than 4 million hits.

Now that THAT’S solved, back to semester planning.

/MD