underpowered sample
Too few subjects to detect the effect you're claiming.
What it is
Statistical power is the probability your study would detect a real effect of a given size. With n=20 and a small effect, your power may be 0.25 — meaning even if the effect is real, you'd see it in only 1 of 4 attempts. A 'significant' result from an underpowered study has a higher posterior probability of being a false positive than the headline p-value suggests.
Why a reviewer cares
Reviewers ask: where does the sample size come from? 'We collected 30 mice because that's typical' is not a power analysis. A real power calculation states the smallest effect size you care about, the test you'll use, and the resulting sample size — usually 30+ for medium effects in two-group comparisons.
How to fix it
Run a power calculation BEFORE collecting data. Report the assumed effect size, the source of that assumption (a pilot study, prior literature), and the resulting target N. If the study is already collected, run a sensitivity analysis: what's the smallest effect you could detect with this N and power 0.8? Frame the conclusions accordingly.
This is one of ~15 canonical methodology explainers Paper Review's red-team report links to. To get a full review of your manuscript, start a Paper Review — $5.