Statistics in Endodontics Part 4: Comparing Means—Basic Statistical Tests in Endodontics

Abstract

Appropriate statistical test selection is essential for valid conclusions in endodontic research and evidence-based clinical practice. This article provides a comprehensive guide to parametric and non-parametric tests for comparing means, emphasizing the critical importance of verifying statistical assumptions before analysis. The review covers fundamental assumptions—continuous dependent variables, independence of observations, normal distribution, homogeneity of variances, and absence of outliers—with practical methods for assessment
using visual inspection and numerical tests. Parametric tests discussed include One-Sample t-test, Independent Samples t-test, Paired Samples t-test, One-Way ANOVA, and One-Way Repeated Measures ANOVA, with detailed guidance on post-hoc testing procedures for
multiple comparisons. When parametric assumptions are violated, non-parametric alternatives are presented with corresponding post-hoc procedures. A decision framework guides test selection based on research design, data characteristics, and assumption verification. Each test includes specific assumptions, hypotheses, and practical examples from endodontic research. Understanding and correctly applying these principles enhances research reliability, reduces statistical errors, and promotes evidence-based clinical decision-making in endodontics.