Statistical Testing: A Commentary on Research Integrity and Protocol

16 Pages Posted: 16 Jul 2018

See all articles by Thomas R. Dyckman

Thomas R. Dyckman

Cornell University - Department of Accounting

Stephen A. Zeff

Rice University - Jesse H. Jones Graduate School of Business

Date Written: July 7, 2018

Abstract

A great deal of the research published in accounting in recent years involves statistical analysis, with a preponderance of main articles relying heavily on regression analysis. Our paper addresses improvements to both the quality and effective communication of research results to both practitioners and researchers while retaining the rigor necessary to satisfy our academic peers. We first address the advantages of working with practice to uncover and address important problems. Our focus then is on the research process from selection of a topic to the conclusion. We discuss how attention to each aspect of an investigation can improve the quality and the disclosure of academic research.

Keywords: Reaching practice, topic selection, hypothesis formulation, p-values, effect size, replication and meta-analysis

JEL Classification: M41

Suggested Citation

Dyckman, Thomas R. and Zeff, Stephen A., Statistical Testing: A Commentary on Research Integrity and Protocol (July 7, 2018). Available at SSRN: https://ssrn.com/abstract=3209884 or http://dx.doi.org/10.2139/ssrn.3209884

Thomas R. Dyckman (Contact Author)

Cornell University - Department of Accounting ( email )

Ithaca, NY 14853
United States
607-255-3491 (Phone)
607-254-4590 (Fax)

Stephen A. Zeff

Rice University - Jesse H. Jones Graduate School of Business ( email )

6100 South Main Street
P.O. Box 1892
Houston, TX 77005
United States
713-348-6066 (Phone)
713-348-6296 (Fax)

Register to save articles to
your library

Register

Paper statistics

Downloads
100
rank
260,381
Abstract Views
320
PlumX Metrics