Yeah, you confirmed that you're reading a massive strawman into what I'm saying. You're taking it about 10 steps past my observation and assuming my observation assume the endpoint rather than the starting point. To give an example, a couple tweets back.... https://twitter.com/justinquinnn/status/1386383265922232324
2/...you wrote, "...so you test to see how your hypothesis is wrong before you can say it is the best explanation." Of course you do (again, you're preaching to the choir). But I never said anything about the hypothesis being the best explanation, ...
3/...I was only referring to the step prior to that where you formed the hypothesis in the first place. Everything you talked about comes subsequent to that -- which again, I agree with.
4/ Also with your "But is it a statistically significant one?" Yeah, that's important to know. But you don't test for statistical significance of a hypothesis before you have a hypothesis in the first place.
You can follow @LarryCoon.
Tip: mention @twtextapp on a Twitter thread with the keyword “unroll” to get a link to it.

Latest Threads Unrolled: