Artificial intelligence versus traditional software in PLS-SEM: data analysis with smart PLS and ChatGPT

Abstract

This study examines the performance capabilities of Artificial Intelligence (AI) tools in relation to conventional statistical software tools including ChatGPT for executing Partial Least Squares Structural Equation Modeling (PLS-SEM) analysis. This research uses artificial intelligence technologies for PLS-SEM analyses and compares them to the Smart PLS 4 software. The research examines a new technology but applies data analysis to electric vehicle usage factors rather than artificial intelligence. The study collected data from 304 participants for analysis through ChatGPT 5.0 Pro and SmartPLS 4 software platforms. The statistical results produced by AI tools equal those of specialized software but show minor discrepancies in key elements including factor loadings reliability measures and path coefficients. Research indicates that AI tools function as a practical statistical computation platform which increases researcher access to advanced analytical methods. This research shows that while ChatGPT can effectively replicate PLS-SEM results comparable to specialized software, its opaque computational process and minor methodological inconsistencies highlight the need for careful validation of AI-driven analyses.

Ähnliche Arbeiten