Statistics guides

Which test fits your data, what its assumptions actually are, and how to report it. Each guide works through an example on synthetic data, shows the R code, and cites its sources.

• Chi-square or Fisher's exact test: which one should you use?

  When the chi-square approximation holds, where the expected-count-of-5 rule comes from and what it says, and when to use Fisher's exact or McNemar's test.

• How to analyse values below the detection limit

  Why replacing non-detects with half the detection limit biases results, and how Peto-Peto tests and censored regression use them correctly.

• How to choose a statistical test for your data

  Pick a test from four facts about your study: the question, the outcome type, how observations are linked, and how many groups. With a decision table.

• Kaplan-Meier and Cox regression: analysing time-to-event data

  How censoring works, what Kaplan-Meier curves and the log-rank test tell you, how to read a Cox hazard ratio, and when a Weibull AFT model fits better.

• Logistic regression: reading odds ratios and avoiding separation

  How to fit and read a logistic regression for a yes/no outcome: odds ratios with confidence intervals, events per variable, and what to do about separation.

• Multiple comparisons: Bonferroni, Holm, or Benjamini-Hochberg?

  Family-wise error versus false discovery rate, how Bonferroni, Holm and Benjamini-Hochberg differ, and which post hoc test follows ANOVA or Kruskal-Wallis.

• Paired t-test or Wilcoxon signed-rank test?

  Both tests analyse the paired differences, not the raw columns. What each assumes about those differences, how zeros and ties are handled, and what to report.

• T-test or Mann-Whitney U test: which should you use?

  Welch's t-test, Student's t-test and the Mann-Whitney U test answer different questions. What each one tests, and why Welch is a sensible default.

• What makes a statistical analysis reproducible?

  Seeds, pinned package versions, the exact script, session information and a decision record: what reproducibility needs and how it differs from replication.