Hypothesis Test Selector AI Prompt

Help me select the correct statistical test for this analysis.

Data description: {{data_description}}
Research question: {{research_question}}
Sample size: {{n}}
Data types: {{data_types}} (continuous, ordinal, nominal, count)
Number of groups: {{n_groups}}
Design: {{design}} (independent groups, paired/repeated measures, one-sample)

1. Apply the test selection decision tree:

   COMPARING MEANS / CENTRAL TENDENCY:
   - 1 group vs known value, continuous, normal: One-sample t-test
   - 2 independent groups, continuous, normal, equal variance: Independent t-test
   - 2 independent groups, continuous, normal, unequal variance: Welch's t-test (prefer over Student's when in doubt)
   - 2 paired groups, continuous, normal: Paired t-test
   - 3+ independent groups, continuous, normal: One-way ANOVA
   - 3+ groups with 2+ factors: Factorial ANOVA or mixed ANOVA
   - 2 independent groups, non-normal or ordinal: Mann-Whitney U
   - 2 paired groups, non-normal or ordinal: Wilcoxon signed-rank
   - 3+ independent groups, non-normal: Kruskal-Wallis
   - 3+ paired groups, non-normal: Friedman test

   COMPARING PROPORTIONS:
   - 1 proportion vs known value: One-sample z-test or exact binomial
   - 2 independent proportions: Chi-square test or Fisher's exact (use Fisher's if any cell < 5)
   - 2 paired proportions: McNemar's test
   - 3+ independent proportions: Chi-square test of independence

   CORRELATION AND ASSOCIATION:
   - 2 continuous variables, linear: Pearson correlation
   - 2 ordinal or non-normal continuous: Spearman correlation
   - 2 binary or nominal: Phi coefficient, Cramér's V
   - 2 continuous, agreement between raters: Intraclass Correlation (ICC)

   REGRESSION:
   - Continuous outcome, 1+ predictors: Linear regression (check assumptions)
   - Binary outcome: Logistic regression
   - Count outcome: Poisson or negative binomial regression
   - Ordinal outcome: Ordinal logistic regression
   - Time-to-event: Cox proportional hazards

2. Assumption check for selected test:
   - What assumptions must be verified before running this test?
   - How to check each assumption (normality: Shapiro-Wilk if n < 50, Q-Q plot; equal variance: Levene's test; independence: by design)
   - What to do if an assumption is violated (non-parametric alternative, transformation, robust methods)

3. Multiple testing consideration:
   - If you are running more than one test on the same dataset, correction for multiple comparisons is needed
   - Bonferroni: divide alpha by the number of tests (conservative)
   - Benjamini-Hochberg FDR: controls false discovery rate (less conservative, preferred for many tests)

Return: recommended test, assumptions to verify, alternative if assumptions are violated, and multiple comparison strategy.