Analysing Dependent Variables with Multiple Surrogates in Financial Performance Research

Accounting and finance-based researchers often use multiple surrogates to capture the properties of a dependent variable (DV) when studying its predictive relationship with predictors. This often fail to directly connect the study results with the major objective of the research. This paper compares the existing practice with a plausible and less complicated alternative. Using logistic regression, the study converted the a priori expectations of 30 Ph.D research theses in finance and accounting with four dependent surrogates into a probabilistic log values and compared them with the individual surrogate performance on the one hand and the surrogates geometric mean on the other hand. While the geometric mean revealed close connection with the theses’ probabilistic expectations (β = .278, t₍₃₀₎ = .695, R² = .077, p > .10), the individual surrogates results showed singular and combined significant differences with the theses’ a priori expectations (Adj. R² = .0291, F_{(4, 25)} = 22.598, p < .05). The paper recommends unifying multivariate DVs with geometric means for better conclusion in financial performance relational studies.     

Keywords: Logistic regression, Performance, dependent variable, proxy, surrogate