Sample selection and theory development: Implications of firms' varying abilities to appropriately select new ventures arturs kalnins

Abstract

I highlight the need to consider sample selection when developing theory. When a sample is the result of a selection process, the process may be (1) generating empirical relationships consistent with a theoretical explanation that plays no causal role or (2) canceling out an empirical relationship actually generated by a causal process associated with a proposed theory. I argue that firms' varying abilities to appropriately select new ventures and select in or out of samples of such investments can lead to empirical misinterpretation and inappropriate theoretical conclusions.