Moving beyond multiple regression analysis to algorithms: Calling for adoption of a paradigm shift from symmetric to asymmetric thinking in data analysis and crafting theory
被引:1365
|
作者:
Woodside, Arch G.
论文数: 0引用数: 0
h-index: 0
机构:
Boston Coll, Carroll Sch Management, Dept Mkt, Chestnut Hill, MA 02467 USABoston Coll, Carroll Sch Management, Dept Mkt, Chestnut Hill, MA 02467 USA
Woodside, Arch G.
[1
]
机构:
[1] Boston Coll, Carroll Sch Management, Dept Mkt, Chestnut Hill, MA 02467 USA
Algorithm;
Causal recipe;
Configuration;
Consistency;
Coverage;
Fit validity;
Fuzzy set qualitative comparative analysis;
Multiple regression analysis;
Predictive validity;
D O I:
10.1016/j.jbusres.2012.12.021
中图分类号:
F [经济];
学科分类号:
02 ;
摘要:
This editorial suggests moving beyond relying on the dominant logic of multiple regression analysis (MRA) toward thinking and using algorithms in advancing and testing theory in accounting, consumer research, finance, management, and marketing. The editorial includes an example of testing an MRA model for fit and predictive validity. The same data used for the MRA is used to conduct a fuzzy-set qualitative comparative analysis (fsQCA). The editorial reviews a number of insights by prominent scholars including Gerd Gigerenzer's treatise that "Scientists' tools are not neutral." Tools impact thinking and theory crafting as well theory testing. The discussion may be helpful for early career scholars unfamiliar with David C. McClelland's brilliance in data analysis and in introducing business research scholars to fsQCA as an alternative tool for theory development and data analysis. (C) 2013 Elsevier Inc. All rights reserved.