Kapil Dev Joshi Logo
Kapil Dev Joshi
Healthcare & Academic Services
Back to Notes
Data AnalysisPinned

Statistical Tests Cheat Sheet

Kapil Dev Joshi-March 1, 2024

Statistical Tests Cheat Sheet


Choose the right statistical test based on your data and research question.


Decision Flowchart


Comparing Groups


Data Type2 Groups3+ Groups
Continuous (normal)Independent t-test / Paired t-testOne-way ANOVA / Repeated measures ANOVA
Continuous (non-normal)Mann-Whitney U / WilcoxonKruskal-Wallis / Friedman
CategoricalChi-square / Fisher's exactChi-square

Relationships


VariablesTest
2 continuous (normal)Pearson correlation
2 continuous (non-normal)Spearman correlation
1 continuous + 1 categoricalPoint-biserial correlation
Multiple continuousMultiple regression
Categorical outcomeLogistic regression

Assumptions to Check


Parametric Tests

  • Normal distribution (Shapiro-Wilk test)
  • Homogeneity of variance (Levene's test)
  • Independence of observations
  • Interval/ratio data

    • Non-Parametric Tests

  • Use when parametric assumptions violated
  • Ordinal data
  • Small sample sizes

    • Sample Size Rules of Thumb


  • T-test: Minimum 30 per group
  • ANOVA: Minimum 20 per group
  • Correlation: Minimum 30 pairs
  • Chi-square: Expected count ≥5 in 80% of cells
  • Regression: 10-20 observations per predictor

    • Effect Size Interpretation


    #statistics#research#data analysis#cheat sheet