Premium
Matching method to paradigm in strategy research: limitations of cross‐sectional analysis and some methodological alternatives
Author(s) -
Bowen Harry P.,
Wiersema Margarethe F.
Publication year - 1999
Publication title -
strategic management journal
Language(s) - English
Resource type - Journals
SCImago Journal Rank - 11.035
H-Index - 286
eISSN - 1097-0266
pISSN - 0143-2095
DOI - 10.1002/(sici)1097-0266(199907)20:7<625::aid-smj45>3.0.co;2-v
Subject(s) - matching (statistics) , econometrics , focus (optics) , variation (astronomy) , economics , strategic management , computer science , statistics , mathematics , management , physics , astrophysics , optics
A central focus of empirical research in strategic management has been to understand the relationships associated with the structure–strategy–performance paradigm. To examine these relationships, investigators have relied extensively on cross‐sectional methods that embody the implicit assumption that model parameters are stable across firms and over time. Yet, many of the theoretical constructs used in strategic management have clear firm‐ and time‐specific components. Hence, it might be expected that the parameters of the relationships investigated empirically will vary across firms and over time. Whereas recent research has raised concerns about the use of cross‐sectional analysis when parameters vary over time, little attention has been given to the issue of parameter variability across firms. Given the focus of strategy researchers on firm‐level effects and the predominant reliance on cross‐sectional analysis, accounting for across‐firm variability is a significant methodological issue. Failure to account for such variability can lead to biased parameter estimates and incorrect inferences. This paper argues for the adoption of alternative methods that can overcome the limitations of a cross‐sectional analysis and it offers guidance on how researchers can proceed to use these alternative methods to explicitly incorporate or test for variation in model parameters across firms or over time. Copyright © 1999 John Wiley & Sons, Ltd.