KONSİLCalculator · deterministic
One-way ANOVA sample size calculator
Three or more independent groups: total sample from Cohen's f and the number of groups.
Calculator
No AI, no sign-up — the arithmetic runs in the server-side deterministic solver, validated against G*Power and R pwr.
When to use it
For comparing the means of three or more independent groups in one test: three doses, three devices, four center types. Remember: a significant omnibus ANOVA does not say which groups differ — pre-specify the pairwise comparisons and their multiplicity correction.
Where the inputs come from
- Cohen's f: can be derived from published group means and SDs; if eta-squared is known, f = √(η² / (1 − η²)).
- Number of groups: the real number of arms in the design; a "maybe we'll add one" arm is resolved in the protocol, not here.
Worked example
For three groups and a medium effect (f = 0.25), with α = 0.05 and 80% power the solver requires 53 participants per group (≈ 158 in total). At 20% attrition the total recruitment target is 198. Three well-defined groups beat four thin ones — for power and for interpretation.
How to write it in the protocol
“For the primary analysis comparing [k] groups, the expected effect was f = [value] [with source]. One-way ANOVA with α = 0.05 and 80% power required [n] per group; allowing [x]% attrition, [N] in total were targeted.”
Sources
- Cohen J. Statistical Power Analysis for the Behavioral Sciences, 2nd ed. Lawrence Erlbaum; 1988.
- Faul F, Erdfelder E, Lang A-G, Buchner A. G*Power 3: a flexible statistical power analysis program. Behav Res Methods. 2007;39:175–191.
- Champely S. pwr: Basic Functions for Power Analysis (R package).