What an IRB Wants to See in Your Methods Section

2 min read ▶ Listen to this post

An IRB isn't reviewing your statistics for rigor the way a journal reviewer would. It's checking whether the risks to participants are justified by the design — and a vague or confusing methods section reads as a poorly-justified design, whether or not the underlying plan is actually sound. Clarity here is doing double duty: it's both a courtesy to the reviewer and the thing that gets your protocol through in one round instead of three.

What a methods section needs to specify clearly

• The study design, named specifically — RCT, quasi-experimental, observational cohort — and a sentence on why that design fits the question, not just a description of activities.
• Sampling and recruitment, concretely: who, how many, and how selected.
• How the sample size was actually determined. A real power analysis, not a round number picked for convenience — a reviewer wants to see that the sample is neither so small the risk isn't justified by any realistic chance of a meaningful answer, nor so large that more people than necessary are exposed to it.
• Data collection procedures in enough detail that a reviewer can picture exactly what a participant experiences, in what order, and for how long.
• A data security and confidentiality plan matched to the actual sensitivity of what's being collected — de-identification approach, storage, and who has access.
• An analysis plan, even a brief one. It signals the data won't be mined after the fact for whatever comes out significant, which is exactly the kind of thing a careful reviewer is watching for.

The most common mistake

Writing the methods section in vague, aspirational language — "we will analyze the data using appropriate statistical methods" — reads to a reviewer as a sign the plan isn't actually locked in yet, which raises questions about risk instead of answering them. Specificity is the thing that closes the loop.

If you're preparing a methods section for IRB or a related review, this is exactly where our study-design work usually starts — including the power analysis behind the sample-size justification, using our own free power calculators if you want to see the math yourself first.

Related reading

• What Belongs in a Statistical Analysis Plan? — the analysis plan an IRB methods section should point to.

One email whenever we publish something new. No spam, unsubscribe anytime.