 As you've worked through these videos, you've probably realized there's a whole bunch of bullshit out there. Anyway, this is one of those situations where we have to give you the definition of a word, basically, just to tell you what not to do. So this is one of those things. Total count inter-observer agreement. Don't use it. Just don't, alright? All you do is take the total number of instances that person A saw, divide it by the total number of instances that person B saw, smaller divided by the larger. You get the idea. But it's horrifically inaccurate. It's awful. You could have people completely- mention this scenario, alright? You got just two intervals. Person A saw nothing. Person B saw five. Second interval. Person A saw five. Person B saw nothing. So they didn't agree at all. But if you add them both up, they both saw five times. We agree perfectly, Bob. It's horrific. Don't use it. It's bad. Just don't. That's why we have so many videos on un-observer agreement. Because there's a lot of other methods that are way, way better than this one. It's just one of those horrific methods. Don't use it. Please. Stop. Don't use it.