RESEARCH
Statistical research
Good statistical research begins before the analysis: define the question, population, outcome and comparison, then plan how measurements and missing responses will be handled.
Make decisions visible
Explain the sample and measurement process, state the analysis plan, and report effect sizes with uncertainty where appropriate. A small p-value does not show that an effect is large, important or causal. Keep the conclusion within what the study design can support.
WORKED EXAMPLES
Research guides
Plain-language explanations for common analysis decisions.
New worked examples will appear here as they are published.