Probably. After reading the paper I'm pretty sure they messed up their testing.

The problem is that they blow it up HUGE both in the paper and the article without cursory checking. That is as bad as lie. Maybe even worse.

From The paper 7.1.2: "Censored Planet data reveals extremely aggressive DNS blocking of many domains in Norway, with many blocks being consistent in all of our vantage points. During the four month period of increased censorship, 25 ASes observed blocking of more than 10 domains in at least six categories. We observed the most rigorous activity in AS 2116 (CATCHCOM) where more than 50 domains were blocked."

Blocking more than 10 domains in 25 ASes I have no reason to doubt. As I said there is extensive DNS blocking. So they write in a way that makes it sound like all the ASes are blocking hrw.org and match. But it's probably only on CATCHCOM. This is a ISP for mostly huge corporate environments where blocking may be by customer demand. So they take one single source of "anomaly" and blow it up. Pretty much a lie.

It’s such an egregious mistake that you’d have to assume the rest of their data is similarly worthless.

I agree. I mean, were I not a skeptic I'd be walking around dropping the interesting tidbit that "Norway blocks match.com" - it's as bad as the info-memes with nonsense on them people quote from facebook.

Could be. But I also don't recall any of the blocking in Poland they refer to.

I also did not notice it. Will need to check OONI Probe.