============================================================================== STATCLINIC STATISTICAL RESULTS REPORTING TEMPLATE Fill-in-the-blank templates for 20 statistical tests used in medical research Aligned with APA 7th edition, CONSORT, STROBE, and PRISMA reporting standards Free tools for every test: https://statclinic.net ============================================================================== HOW TO USE THIS FILE --------------------- Find the test you ran below, copy its template line, and replace every [value] placeholder with your own result. Every template already follows the correct APA notation (italicized statistic symbols, exact p-values, effect size, and 95% CI) expected by medical journals and thesis committees. See the full guide with worked examples at: https://statclinic.net/blog/how-to-write-statistical-results-by-test-medical-research 1. DESCRIPTIVE STATISTICS --------------------------- Normal data: Mean +/- SD [or M = value, SD = value]; range = min-max Non-normal data: Median (IQR = lower-upper) Categorical data: n (%) of the valid (non-missing) sample 2. INDEPENDENT SAMPLES T-TEST -------------------------------- t(df) = [value], p = [value], 95% CI [lower, upper], Cohen's d = [value] 3. PAIRED SAMPLES T-TEST --------------------------- t(df) = [value], p = [value], 95% CI of the difference [lower, upper], Cohen's d = [value] 4. MANN-WHITNEY U TEST ------------------------- U = [value], z = [value], p = [value], r = [value] (report median and IQR per group, not mean and SD) 5. WILCOXON SIGNED-RANK TEST ------------------------------- Z = [value], p = [value], r = [value] (report pre/post medians and IQR) 6. ONE-WAY ANOVA ------------------- F(df_between, df_within) = [value], p = [value], eta-squared = [value] Post-hoc: [Group A] vs [Group B], p = [value] (Tukey / Games-Howell) 7. REPEATED MEASURES ANOVA ----------------------------- Mauchly's test: chi-square(df) = [value], p = [value] If violated -> Greenhouse-Geisser corrected F(df, df) = [value], p = [value], partial eta-squared = [value] If not violated -> Sphericity Assumed F(df, df) = [value], p = [value], partial eta-squared = [value] 8. KRUSKAL-WALLIS TEST ------------------------- H(df) = [value], p = [value], epsilon-squared = [value] Post-hoc (Dunn's, Bonferroni-adjusted): [Group A] vs [Group B], p = [value] 9. FRIEDMAN TEST ------------------- chi-square(df) = [value], p = [value], Kendall's W = [value] 10. CHI-SQUARE TEST ---------------------- chi-square(df, N = [value]) = [value], p = [value], Cramer's V = [value] 11. FISHER'S EXACT TEST -------------------------- Fisher's exact p = [value], Odds Ratio = [value], 95% CI [lower, upper] 12. PEARSON CORRELATION -------------------------- r(df) = [value], p = [value], 95% CI [lower, upper], r-squared = [value] 13. SPEARMAN CORRELATION --------------------------- rho(df) = [value], p = [value], 95% CI [lower, upper] 14. LINEAR REGRESSION ------------------------ Per predictor: B = [value], 95% CI [lower, upper], beta = [value], p = [value] Model: F(df, df) = [value], p = [value], Adjusted R-squared = [value] 15. LOGISTIC REGRESSION -------------------------- Per predictor: Adjusted OR = [value], 95% CI [lower, upper], p = [value] Model: chi-square(df) = [value], p = [value], Nagelkerke R-squared = [value] Calibration: Hosmer-Lemeshow p = [value] (non-significant = good fit) 16. KAPLAN-MEIER SURVIVAL ANALYSIS ------------------------------------- Median survival = [value] (95% CI [lower, upper]) per group Log-rank: chi-square(df) = [value], p = [value] (If median not reached: report survival rate at a fixed time point, e.g. 12-month survival = [value]%) 17. COX PROPORTIONAL HAZARDS REGRESSION ------------------------------------------ Adjusted HR = [value], 95% CI [lower, upper], p = [value] Proportional hazards assumption: [state method, e.g. Schoenfeld residuals, p = value] 18. ROC CURVE ANALYSIS ------------------------- AUC = [value], 95% CI [lower, upper], p = [value] Optimal cutoff (Youden index) = [value]; sensitivity = [value]%, specificity = [value]% 19. INTRACLASS CORRELATION COEFFICIENT (ICC) ------------------------------------------------ ICC = [value], 95% CI [lower, upper] Model: [state model type, e.g. two-way random-effects, absolute agreement] 20. BLAND-ALTMAN ANALYSIS ---------------------------- Mean bias = [value] 95% limits of agreement = [lower] to [upper] 21. COHEN'S KAPPA -------------------- kappa = [value], 95% CI [lower, upper], p = [value] Interpretation: [poor / fair / moderate / substantial / almost perfect] agreement ============================================================================== UNIVERSAL CHECKLIST -- APPLY TO EVERY RESULT BEFORE SUBMISSION ============================================================================== [ ] Named the test and the comparison being made [ ] Reported the test statistic with its degrees of freedom [ ] Reported the exact p-value (three decimals, or "p < 0.001") [ ] Reported an effect size or association measure [ ] Reported the 95% confidence interval where available [ ] Matched the summary statistic to the test (mean+/-SD vs median[IQR]) [ ] Stated the sample size actually analyzed (N) [ ] Followed a significant omnibus test with pairwise/post-hoc results [ ] Applied the correct design-specific guideline (CONSORT / STROBE / PRISMA) [ ] Distinguished statistical significance from clinical significance [ ] Cited the statistical software (name, version, manufacturer) once in Methods [ ] Cross-checked that text, tables, and figures all report the same numbers ============================================================================== Full guide with real, incorrect, and corrected examples for every test above: https://statclinic.net/blog/how-to-write-statistical-results-by-test-medical-research Run any of these tests free at: https://statclinic.net ==============================================================================