[Knowledge Sharing 20] 知識分享 2023-08-19

知識分享 Top 22 common errors and 8 new guidelines for SEM 

Common error (issue) 20: Absence of statistical significance is interpreted as “no effect.”

Ans: Statistically insignificant results simply imply that “no conclusion can be drawn” from the data. To test for “no effect,” use additional analyses (e.g., equivalence tests, Bayes factors).

Reference

Jesper N. Wulff, Gwendolin B. Sajons, Ganna Pogrebna, Sirio Lonati, Nicolas Bastardoz, George C. Banks, John Antonakis, (2023) Common methodological mistakes, The Leadership Quarterly , DOI: 10.1016/j.leaqua.2023.101677