… inferential statistics.
I remember the first time I was exposed to null values, p values, and the lot. I found them baffling and bizarre. When well presented, I could follow, but I had a heck of a time making heads or tails of them on my own. I began to wonder if I were dyslexic (I’m not) or if I’d accidentally taken some kind of hallucinogen (I hadn’t). Letters and weird bits of images of words swirled in my head, never seeming to settle. (Which was weird, I’m not going to lie).
Over time, I pieced it together. I think. Nonetheless, the early trauma remains and every time I do any kind of substantial statistics I must check, and recheck them, and then still ask for someone else to check them.
Now it seems that effort was all for not. Even confidence intervals are not safe (that 19 times out of twenty thing that everyone is used to hearing during media discussions of polls and surveys).
Mind, this is a journal mostly outside of my field, and not one I’m likely to which I’m likely to submit an article, so it’s impact on my work is not direct. But still. Who would have thunk it?
Maybe my teenage statistics self was the wiser self after all.