Reflections on Effect Size Measures in Meta-Analyses

Due to its comparability across studies, the most often applied effect size measure in meta-analyses is a correlation coefficient. However, correlation coefficients only express the strength of the statistical relationship but not the substantial impact (leverage effect) of one variable on another. In management as an applied discipline managers and thus researchers are not only interested in the existence of a relationship but want to know something about the leverage effect of a management action/effort on performance (such as sales). Therefore, I discuss what can be concluded from meta-analyses based on correlations and what elasticities based on relative changes of the dependent and independent variable can offer. I present how elasticities and correlations are related together in terms of mathematical formulas. I also show that correlations cannot serve as proxy for impact. The calculation of elasticities requires ratio-scaled variables such as sales, time, or money. The variables found in the majority of management studies are constructs and, as such, interval-scaled. Such scales do not have an absolute point of zero. Therefore, relative changes are arbitrary so I develop pseudo-elasticities as effect size measures that can be used for continuous interval and dummy scales and are comparable across studies.

Sönke Albers is Professor of Marketing and Innovation at Kühne Logistics University and from 2010 to 2016 founding Dean of Research and responsible for ensuring that KLU becomes a research-oriented university that is internationally competitive. He is very active in educating future scientists with  a total of 30 professors originating from his efforts. His research interests are lying in the areas of marketing planning, sales management, diffusion of innovations, and meta-analyses. He is author of about 150 articles in international and German journals such as Marketing Science, Journal of Marketing Research, International Journal of Research in Marketing, Journal of the Academy of Marketing Science and published over 15 books. He was selected to receive the 2011 EMAC Distinguished Marketing Scholar Award that marks the highest honor that a marketing scholar in Europe can receive. Before that he has won together with Marc Fischer the 2009-2010 INFORMS Society for Marketing Science MSI Practice Prize with their work on “Dynamic Marketing Budget Allocation across Countries, Products, and Marketing Activities”. Recently, he has been selected as the 2020 Lifetime Achievement Award recipient of the American Marketing Association Selling & Sales Management Special Interest Group.