The Misuse of Regression-Based x-Scores as Dependent Variables in Accounting Research
59 Pages Posted: 4 May 2021
Date Written: April 29, 2021
Researchers often use regression-based x-Scores (e.g., C-Score, F-Score, and Z-Score) from a stage 1 model as a dependent variable in stage 2. We argue that this x-Score analysis causes major biases and interpretation problems because (1) x-Score cannot capture new sources of variation, and (2) the estimates are often artifacts of arbitrary technical assumptions. Instead, researchers should just include the test variables and the relevant controls in stage 1, obviating the need for an x Score. We confirm these arguments using simulations. In replication analyses, major published findings change after we remove the bias caused by the misuse of x Score.
Keywords: two-stage estimation; exclusion restrictions; functional forms; identification
JEL Classification: M41, C20, C36, C43
Suggested Citation: Suggested Citation