Robustification of Structural Equation Modelling via Global Sensitivity Analysis

Abstract We propose a method for enhancing the robustness of structural equation modelling, a mul- tivariate statistical analysis technique employed for analyzing causal relationships among different aspects of given phenomena. This enhancement is achieved through the integra- tion of global sensitivity analysis, which assesses how uncertainties in model output can be attributed to various sources of input uncertainty. The robustification process involves several key steps, including bootstrapping evidence, error propagation, and uncertainty quantification. This method introduces a novel approach termed “modelling of the mod- elling process”. To illustrate this approach, we apply it to a previously published test case where structural equation modelling is used to relate the impact of artificial intelligence adoption on employee engagement. By quantifying the uncertainty inherent in the infer- ence from our test case, this procedure generates a more robust and defensible inference. The performed procedure significantly enhances the robustness of the results derived from the test case.