Effect size

Effect size is a number that describes how large an effect is, not just whether it exists. It tells you whether a result is big enough to matter in real life.

Definition from Shrinktionary, reviewed by Shariq Refai, MD, MBA, FAPA (last reviewed 2026-06-26). Shrinktionary is the network's language layer and carries the canonical short definition; this page adds the research-context detail below. Full definition on Shrinktionary →

Trials measure anxiety in different ways and on different scales. An effect size puts results on a common ruler. A widely used one is Cohen's d. As a rough guide, around 0.2 is a small effect, 0.5 is moderate, and 0.8 or more is large. Another common form is the standardized mean difference, used in meta-analyses.

When this site says CBT shows moderate to large effects for anxiety, that statement is about effect sizes pooled across trials. Effect size is how researchers compare a therapy with a medication, or one drug with another, even when the original studies used different scales.

A meta-analysis reports that exposure therapy for specific phobia has an effect size of about 1.1 compared with waitlist. That's a large effect.

Effect size labels are rough guides, not strict rules. A small effect can still matter at the population level, and a large effect against a weak comparison group can shrink against an active one. Check what the effect was measured against.