Generalized knowledge comes from cumulating results across studies, a process known as meta-analysis. Efficiently increasing generalized knowledge in a defined area—estimates of price or advertising, for example—is one important goal for research. Because (1) most meta-analyses are based on highly inefficient and unbalanced natural experiments or designs and (2) additional studies are costly, carefully selecting the next study is important. The authors demonstrate that, rather than simply selecting a study that uses currently underrepresented design variables, a procedure that reduces collinearity among design variables will produce far superior improvements in knowledge.
Get full access to this article
View all access options for this article.
References
1.
AssmusGert, FarleyJohn U., and LehmannDonald R. (1984), “How Advertising Affects Sales: Meta-Analysis of Econometric Results,”Journal of Marketing Research, 21(February), 65–74.
2.
BelsleyDavid A. (1980), Conditioning Diagnostics: Collinearity and Weak Date in Regression.New York: John Wiley & Sons.
3.
FarleyJohn U., and LehmannDonald R. (1986), Meta-Analysis in Marketing: Generalization of Response Models.Lexington, MA: Lexington Books.
4.
FarleyJohn U., and SawyerAlan (1995), “Empirical Marketing Generalization Using Meta-Analysis,”Marketing Science, 14(September), G36–46.
5.
LeffNathaniel, and FarleyJohn U. (1980), “Advertising Expenditure in the Developing World,”Journal of International Business Studies, 11(Fall), 65–79.
6.
MelaCarl, and KopallePraveen (1998), “The Impact of Collinearity on Regression: The Asymmetric Effect of Negative and Positive Correlations,” working paper, Business School, Notre Dame University.
7.
RustRoland T., LehmannDonald R., and FarleyJohn U. (1990), “Estimating Publication Bias in Meta-Analysis,”Journal of Marketing Research, 27(May), 220–26.
8.
SilveyS. D. (1969), “Multicollinearity and Imprecise Estimation,”Journal of the Royal Statistical Society, Series B, 31, 539–52.
