Many outcomes of interest to management and strategy researchers are in the form of fractions, proportions, or percentages. We review 10 years of research in seven leading strategy and management journals. Instead of implementing best-practice techniques, such as fractional logistic regression, management scholars rely primarily on linear regression, log-odds regression, or the Tobit model. Following up on our review, we present re-estimations of two published papers to show how best-practice methods yield substantially different results than the most commonly used methods. Using simulations, we confirm the results from the reproduced examples in a broader context. Finally, we present a worked example that researchers can lean on when they deal with fractional outcomes.
展开▼
