analysis · Any · free
Check whether this result is statistically significant, then explain like I'm a PM, not a statistician. Data: Test type: [CONVERSION / MEAN / COUNT] Group A: [N + successes OR mean + SD + n] Group B: [N + successes OR mean + SD + n] Alpha: [DEFAULT 0.05] One or two tailed: [SPECIFY] Deliver: The correct test (chi square, two proportion z, Welch's t, etc.) and why that one Test statistic, p value, 95% CI on the difference Verdict: significant / not significant / underpowered Effect size (Cohen's h or d) — is the difference big or trivial? What the p value does NOT mean (kill the wrong intuitions) How much more data I'd need to reach 80% power if underpowered Show the arithmetic, not just the answer.
#statistics #significance #calculator