effect size
how big a difference a treatment actually makes, on a standard scale — the number that separates statistically significant from worth taking.
a trial can be positive while the improvement it measures is small. this is the number that says how much.
what it is
Standardized measures like Hedges' g or Cohen's d put treatment effects on one scale: roughly 0.2 small, 0.5 medium, 0.8 large. Psychiatric drug effects mostly live in the small-to-medium range.
why headlines mislead
The landmark network meta-analysis of 21 antidepressants found all of them beat placebo — with between-drug differences smaller than the rankings implied. An AI-chatbot meta-analysis found g=0.61 for depression in young people and a null result for anxiety. Same arithmetic, very different headlines. Statistical significance says an effect is probably real; effect size says whether it is worth the trade-offs.
somewhere to put it
free. anonymous. people who’ve been where you are 🤍
get Resolv Social — it’s freewant the deeper story? read prescribed to fail
questions
what is a good effect size?
Context decides — but 'how big is the effect versus placebo, in numbers?' is the most clarifying question you can ask about any treatment claim.
all 21 antidepressants beat placebo — so they all work?
They all beat placebo in pooled trials; the gaps between them are smaller than the rankings look, and the average effect is modest. Both halves matter.
keep reading
more from the glossary
resolv social is not a clinical product and does not diagnose, treat, or cure any condition. if you’re in crisis, call or text 988 (u.s.), 24/7, free.