Methodological Proposal for the Design and Validation of Research Instruments Supported by Artificial Intelligence

Abstract

Introduction: the validity of data collection instruments is essential to ensure the quality and replicability of scientific studies; traditional methods require time, resources, and expert participation, making validation difficult. Objective: To develop a procedure for the design and validation of research instruments using Artificial Intelligence as a methodological support tool. Methods: an eight-phase model was designed, ranging from conceptual review and item formulation to linguistic evaluation, simulated rational validation, comprehension verification, internal consistency analysis, and structural optimization.Results: the process demonstrated applicability, technical coherence, and practical utility. ChatGPT 4.5 enabled the automation of analyses and the generation of content aligned with theoretical constructs, optimizing the preliminary validation phases. Conclusions: AI represents a viable alternative in resource-limited settings. While it does not replace classic empirical methods, it complements methodological rigor in key stages. Ethical and technical protocols must be established for its responsible use in scientific research.

Ähnliche Arbeiten