Because the streaming era has made it harder to capture reliable viewership data.

I get that the participants are a particular set of people, which IMO means by definition any data gathered is skewed anyway to the types of people willing to do this manner of study, but I'm certainly not the only person who leaves stuff on in the background just for noise even though I'm not "watching" it. I wonder if they think they can account for that - just because I have something on doesn't mean I'm really paying attention, sometimes. I'm sure that's an issue with any way they try to capture these kinds of metrics. There is a big difference to me with shows I care if they cancel versus shows I have on because I like to hear nature sounds in the background. Or I'll just throw on something I've seen a bunch of times for background noise.