The Problem With Standard Errors in Effect Size Calculations

If you're calculating an effect size in the co ends D family, generally what you're calculating is a difference between groups divided by a standard deviation. And so if you just mess up and plug a standard error in there as well, or if you plug a standarderror in instead of a standard deviation, you're making the denominator of that of that fraction much, much smaller. That has the effect of massively inflating the effect size. So an effect size that maybe should have been, you know, a D of point four, like a relatively small effect size might be 1.5 when really it's 2 to 3 times bigger.

Transcript

Play full episode

Transcript

Episode notes

The AI-powered Podcast Player

Save insights by tapping your headphones, chat with episodes, discover the best highlights - and more!

Get the app