#I loooove having people infodumping in my reblogs and asks so I try to open opportunities for discussion often lol

[statistical-distr-of-polls]

hello! I love seeing people excited about statistics, so this gimmick is very special to me. and it’s extra cool that you’re giving tumblr users a chance to discuss statistics (I’m specifically thinking of the height post.)

one thing: I’d be very cautions when identifying skew. tumblr polls collect categorical data, right? some categorical distributions are numeric or have some inherent order (“ordinal distributions”), while others have no inherent order (“nominal distributions”). the DND ability score poll is a great example of the latter.

from a statistical perspective, the order of these options is arbitrary. you said that this distribution skews slightly right, but if we shuffled the categories the same distribution would appear skewed to the left. think of it graphically: if the categories could be displayed on any order on the x axis, the distribution’s shape is arbitrary.

that’s just my word of caution! anyway, I’m excited for you to get into research/medical statistics (I think you expressed interest in them in the tags?) they’re a lot of fun!

Hi!! Thank you so much for the kind words; im always so happy to know that people enjoy this blog, both for the stats and the polls lol

I’ve had quite a few people caution me on the nominal categorical data and the shape of skew haha! I totally understand, but for the sake of this blog, that is a tenet of stats that I ignore 😔💔 I’ll probably add a disclaimer about it in the intro post, though, since this is like the fourth time someone has mentioned it LOL