Read the full-text of this article for free

Title: Alpha, Null Hypothesis Statistical Testing and Confidence Intervals: Where Do We Go From Here?

Authors: Tamela D. Ferguson; Wlliam L. Ferguson

Addresses: Author address listing can be found in the "About the Authors" section at the end of the article.

Abstract: The choice of alpha = .05 to determine significance in null hypothesis statistical testing (NHST) has become ingrained in management research, though the choice of .05 level appears to have little scientific basis and is likely simple convenience. More appropriate approaches might be to choose an alpha level based on specific research design or stage of research stream development. Over the years, many have criticized NHST use, with some calling for its ban while advocating the use of confidence intervals instead. To this end, this paper presents a historical review, support for and criticisms of alpha and the related NHST, as well as discussion of issues concerning the use of confidence intervals as an alternative to NHST. The development of the organizational configurations-performance stream in strategic research is used to illustrate the critical importance of researchers making appropriate statistical testing choices.

Keywords: Alpha; null hypothesis statistical testing; confidence intervals; configurations-performance stream.

DOI: 10.1504/JBM.2002.141077

Journal of Business and Management, 2002 Vol.8 No.1, pp.87 - 98

Published online: 05 Sep 2024 *