Customer-centric science: Reporting significant research results with rigor, relevance, and practical impact in mind

Herman Aguinis, Steve Werner, JéAnna Lanza Abbott, Cory Angert, Hyung Joon Park, Donna Kohlhausen

Research output: Journal PublicationArticlepeer-review

195 Citations (Scopus)

Abstract

In response to the ongoing concern regarding a science-practice gap, we propose a customer-centric approach to reporting significant research results that involves a sequence of three interdependent steps. The first step involves setting an alpha level (i.e., a priori Type I error rate) that considers the relative seriousness of falsely rejecting a null hypothesis of no effect or relationship (i.e., Type I error) relative to not detecting an existing effect or relationship (i.e., Type II error) and reporting the actual observed p value (i.e., probability that the data would be obtained if the null hypothesis is true). The second step involves reporting estimates of the size of the effect or relationship, which indicate the extent to which an outcome is explained or predicted. The third step includes reporting results of a qualitative study to gather evidence regarding the practical significance of the effect or relationship. Our proposal to report research results with rigor, relevance, and practical impact involves important changes in how we report research results with the goal to bridge the science-practice gap.

Original languageEnglish
Pages (from-to)515-539
Number of pages25
JournalOrganizational Research Methods
Volume13
Issue number3
DOIs
Publication statusPublished - Jul 2010
Externally publishedYes

Keywords

  • hypothesis testing
  • practical significance
  • statistical significance

ASJC Scopus subject areas

  • General Decision Sciences
  • Strategy and Management
  • Management of Technology and Innovation

Fingerprint

Dive into the research topics of 'Customer-centric science: Reporting significant research results with rigor, relevance, and practical impact in mind'. Together they form a unique fingerprint.

Cite this