Revisiting the Assumptions for Inferential Statistical Analyses: A Conceptual Guide

UNCG Author/Contributor (non-UNCG co-authors, if there are any, appear on document)
Ang Chen, Professor (Creator)
Institution
The University of North Carolina at Greensboro (UNCG )
Web Site: http://library.uncg.edu/

Abstract: Trustworthiness of results from statistical analyses relies on the fulfillment of a set of assumptions made about data. A survey of published pedagogy studies in physical education revealed that examining the assumptions has been overlooked in data analyses and result reporting. The purpose of this article is to provide a conceptual understanding of the assumptions and to summarize available methods to test and address their violations. In the article. we present information to show that it is inappropriate to overlook the assumptions and ignore their violation's impact on interpretation of results. We summarize current findings from theoretical statistical research that start to challenge the conventional be lief about the robustness of traditional inferential statistical analyses, Remedial procedures recommended in the latest statistical theories are presented for researchers to adopt in order to generate valid results from statistical analyses.

Additional Information

Publication
Quest, 2001, 53, 418-439
Language: English
Date: 2001
Keywords
Statistics, Trustworthiness, Results, Analysis