CASE studies...
...follow an individual example in order to show what mechanisms are in play. It's more or less a short longitudinal study with a sample of 1. They're useless for statistical purposes, even adding a bunch of them together isn't much use as every case study could have used a different procedure.
The quoted study actually has a large sample size with consistent measurements and can, presumably, be used to draw some sort of conclusion with stats to back it up.
You're mixing up the 2 types of study, it's not the article author who's confused about that. You're right that he hasn't cited any of the 'many' case studies, though.
Biting the hand that feeds IT
