Template

Plan how you will analyse your data | before you collect it.

A clear analysis plan locks down your variables and tests before fieldwork begins, so write-up is faster and reviewers have nothing to query.

Mini template

ObjectiveVariablesAnalysisSoftware
Prevalence of outcomeOutcome (binary)Proportion + 95% CIStata / R
Factors associatedOutcome + predictorsLogistic regressionStata / R
Lived experienceIDI transcriptsThematic analysisNVivo / Dedoose
Frequently asked

Questions researchers ask

When should I write the analysis plan?
Before you collect data. Writing it later forces you to fit analyses to whatever data you happen to have.
What goes in an analysis plan?
For each objective: the variables, the type of analysis, the test or model, the software, and how you will present the results.
Do qualitative studies need an analysis plan?
Yes: describe your coding approach (e.g. thematic, framework), who will code, and how you will check agreement.
What software should I use?
Stata, SPSS, R or Python for quantitative; NVivo, ATLAS.ti, Dedoose or even spreadsheets for qualitative. Pick what your supervisor can actually review.
Will reviewers actually read it?
Yes. A vague analysis plan is a common reason proposals are sent back from ethics or grant review.
Next step

Prepare your ethics submission

Continue