TABLE 3.

Guidelines for analysis by design

Ordinal and interval/Ratio dependent variables
Number of groups in independent variable	Nature of groups in independent variable	Sample size	Distribution of DV		Statistical test
1 group		N ≥ 30	Normally distributed		One-sample z test or t test
		N < 30	Normally distributed		One-sample t test
			Not normal		Sign test
2 groups	Independent	N ≥ 30	Normally distributed		t test
			Not normal		Mann-Whitney U or Wilcoxon rank- sum test
		N < 30	Normally distributed		t test
			Not normal		Mann-Whitney U or Wilcoxon rank- sum test
	Paired	N ≥ 30	Normally distributed		Paired t test
			Not normal		Wilcoxon signed-rank test
		N < 30	Normally distributed		Paired t test
			Not normal		Wilcoxon signed-rank test
3 or more groups	Independent	NA	Normally distributed	1 factor/IV	One-way ANOVA
				≥2 factors/IVs	Two-way or N-way ANOVA
			Not normal		Kruskal-Wallis one-way analysis of variance by ranks
	Dependent	NA	Normally distributed		Repeated-measures ANOVA
			Not normal		Friedman two-way analysis of variance by ranks
Nominal measures
1 group			np and n(1-p) ≥ 5		z-approximation
			np or n(1-p) < 5		Binomial
2 groups	Independent		np < 5		Fisher exact test
			np ≥ 5		Chi-squared test
	Paired/ Case control				McNemar
3 or more groups	Independent		np < 5		Collapse categories for chi-squared test
			np ≥ 5		Chi-squared test
	Dependent				Cochran's Q
Regression analysis
Type of data	Distribution of dependent variables		Number of independent variables		Statistical test
Interval/ratio data	Normally distributed		1		Simple linear regression
			>1		Multiple linear regression
Dichotomous data	NA		1 or more		Binary logistic regression
Categorical (>2 categories) data	NA		1 or more IVs		Multinomial regression