Parametric and non-parametric tests for the comparison of two samples which both include paired and unpaired observations

Derrick, Ben; White, Paul; Toher, Deirdre

Abstract

Samples that include both independent and paired observations cause a dilemma for researchers that covers the full breadth of empirical research. Parametric approaches for the comparison of two samples using all available observations are considered, under normality and non-normality. These approaches are compared to naive and newly proposed non-parametric alternatives.

Citation

Derrick, B., White, P., & Toher, D. (2019). Parametric and non-parametric tests for the comparison of two samples which both include paired and unpaired observations. Journal of modern applied statistical methods : JMASM, 18(1), Article 9. https://doi.org/10.22237/jmasm/1556669520

• Journal Article Type: Article
• Acceptance Date: Jun 29, 2018
• Online Publication Date: Mar 9, 2020
• Publication Date: May 1, 2019
• Deposit Date: Jul 12, 2018
• Publicly Available Date: Jul 12, 2018
• Journal: Journal of Modern Applied Statistical Methods
• Print ISSN: 1538-9472
• Peer Reviewed: Peer Reviewed
• Volume: 18
• Issue: 1
• Article Number: 9
• DOI: https://doi.org/10.22237/jmasm/1556669520
• Keywords: partially overlapping, t-test, non-parametric, paired samples, independent samples
• Public URL: https://uwe-repository.worktribe.com/output/873692
• Publisher URL: http://www.jmasm.com/