pictures of my rats: @rats
yiddish folktale bot (currently offline): @Yiddish-Folktales
Seattle area
trans 🏳️⚧️ somewhere between (30 - 35)
I'm reading about different kinds of statistical analysis in user studies and
every few paragraphs starts with "However, it is not always straightforward to ..." and then chains through a few things comparing tradeoffs until just going "usually we just aggregate it all together" and then back to "However," and into another layer of measuring and aggregating
I'm afraid the rest of this textbook is just going to be an ever-deepening rabbithole of measuring stats about stats about stats
I'm taking a class on statististic right now, and yep that tends to be it in my experience. Every method has it's tradeoff, every method is good for somethings but not others and every methods make a bunch of assumptions that are important to verify (like the normality of residuals for linear regression, for instance). It is important to ask ourselves what is being sutdies, why x method is applicable to this situation and in what ways the conclusion is limited.
There is a Tumblr post I like that goes: "hmm, today I think I will treat statistic has infallible knowledge downloaded from planet earth's objectively correct stat counter instead of a summary of information someone obtained in some way and is now presenting to you in a certain way" I think it is easy to think like that, but hopefully the more I learn about stat, the more I can see if the method used is relevant or not.
