Statistical Models in Social and Human Sciences Research: From Classification to the Methodology of Selection and Application

Abstract

Statistical models have become a fundamental pillar of research in the social and human sciences, as they provide scientific tools that contribute to the description, analysis, and interpretation of phenomena, as well as the prediction of their trends. However, the methodological challenge is no longer limited to knowing these models or mastering their techniques; it has extended to the question of how to select the most appropriate statistical model and apply it in a manner consistent with the nature of the research problem, the characteristics of the data, and the objectives of the study. From this perspective, this article aims to provide an analytical examination of statistical models in social and human sciences research, beginning with a conceptual grounding of their meaning and functions, followed by their classification according to their research roles, and culminating in the development of a methodological framework to guide researchers in selecting and appropriately applying suitable statistical models.

The study adopts a descriptive-analytical approach through a review and analysis of contemporary methodological and statistical literature, linking it to the requirements of research in the social and human sciences. The findings indicate that the optimal use of statistical models is not determined by the complexity or abundance of statistical techniques, but rather by the extent to which they are aligned with the research question, the level of measurement of variables, the study design, and the assumptions underlying the model. Such alignment ensures the validity of the analysis, the accuracy of the results, and the strength of the conclusions.

The scientific contribution of this article lies in moving beyond the mere classification of statistical models to proposing a methodological framework that assists researchers in selecting and applying them appropriately. This contributes to improving statistical practices, reducing common methodological errors, and enhancing the quality of quantitative research in the social and human sciences.