confidence interval
the range published around a study estimate expressing its statistical precision — a 95% CI is the range that would contain the true value in 95% of identical repeated studies.
every careful number on this site drags a parenthesis behind it — 19.3% (18.8–19.9). the parenthesis is not decoration. it is the study saying how sure it is.
how to read one
Treat the interval, not the point, as the finding. "33.4% (27.5–39.7)" means the study's best estimate is 33.4% and values across that whole range are consistent with the data. A narrow interval means precision (usually a big sample); a wide one means the point estimate is wearing more confidence than it earned. When an interval for a difference includes zero — or a ratio includes 1 — the study did not distinguish the result from no effect.
what it does not mean
A 95% CI is not "a 95% chance the truth is in this range" in the everyday sense — it describes how the method performs across repeated studies. More practically: intervals capture sampling noise only. Bias — a frame that excludes psychiatric hospitals, an 8% response rate — lives entirely outside the parenthesis. A precise interval around a biased estimate is precisely wrong.
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
why do you print the intervals instead of just the number?
Because two studies can share a headline number and differ enormously in how much that number can bear. The interval is the cheapest honest signal of that difference, and dropping it is the first step of most bad science coverage.
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.