Reflexiones sobre el uso de la estadística inferencial en investigación didáctica

As a product of educational research activities, a large amount of data is produced, the analysis of which represents a challenge for the researchers. In this context, the large number of interrelated factors that may have an impact on the data makes it extremely difficult to draw reliable conclusions, either when describing or predicting the behavior of a student population, or when evidencing the effect of didactic interventions. Statistical inferential analysis provides tools that allow the systematic and efficient evaluation of a sample of the population to be studied. In this work we present a reflection on the implementation and interpretation of results associated with the application of three inferential instruments (probability distribution, confidence interval and hypothesis test) to a case study derived from our teaching experience. These instruments allowed us to evaluate the influence that the course modality (face-to-face or online) exerts on the academic performance of first-year students of Facultad de Quimica, Universidad de la Republica. In addition, these statistical tools were also applied to the study of the incidence, on the same population, of a didactic intervention in which a set of materials was incorporated into a massive first-year chemistry course (General Chemistry I), with the aim of promoting active learning and self-regulation. In this regard, inferential statistics allowed the systematic and efficient analysis of the information associated with this case study, allowing to draw statistically significant conclusions.