THE SEARCH FOR ASTERISKS: COMPROMISED STATISTICAL TESTS AND FLAWED THEORIES
被引:106
|
作者:
Bettis, Richard A.
论文数: 0引用数: 0
h-index: 0
机构:
Univ N Carolina, Kenan Flager Business Sch, Strategy & Entrepreneurship Dept, Chapel Hill, NC 27599 USAUniv N Carolina, Kenan Flager Business Sch, Strategy & Entrepreneurship Dept, Chapel Hill, NC 27599 USA
Bettis, Richard A.
[1
]
机构:
[1] Univ N Carolina, Kenan Flager Business Sch, Strategy & Entrepreneurship Dept, Chapel Hill, NC 27599 USA
statistical methodology;
significance levels;
data snooping;
theory testing;
replication;
D O I:
10.1002/smj.975
中图分类号:
F [经济];
学科分类号:
02 ;
摘要:
This paper discusses repeated tests and the resultant reporting of statistical significance when it is actually not present. These errors interact with professional norms such as biases against both replication studies and 'non-results' to undermine the efficacy of our base of empirically tested theory. This raises serious issues for the future of strategic management research. Suggestions are made for dealing with these issues substantively and in terms of professional norms. Copyright (C) 2011 John Wiley & Sons, Ltd.