Artificial Intelligence in the formative assessment of soft skills in upper elementary students

Abstract

The rapid development of technology has transformed teaching, learning, and assessment processes, creating opportunities for the use of Artificial Intelligence tools in educational contexts. The assessment of soft skills, such as critical thinking, problem-solving, effective communication, and collaboration, has become increasingly relevant; however, their measurement remains complex due to the partially subjective nature of these competencies. This limitation is evident at Almirante Alfredo Poveda Burbano School, where assessment practices have traditionally focused mainly on cognitive outcomes. This study aimed to analyze the use of generative Artificial Intelligence in the formative assessment of soft skills in upper basic education students. A quantitative, applied study with a pre-experimental one-group pre-test and post-test design was conducted with a sample of 85 upper basic education students. Formative assessment activities mediated by generative AI were implemented, and the data were analyzed using descriptive statistics, ANOVA, and Pearson’s correlation coefficient. The results showed statistically significant differences between the pre-test and post-test measurements (F = 27.15; p = 0.00199), with improvements in critical thinking from 60% to 75%, problem-solving from 65% to 78%, communication from 55% to 70%, and collaboration from 58% to 72%. Pearson’s correlation coefficient also showed a strong positive association between both measurements (r = 0.98; p = 0.002). It is concluded that the strategic and pedagogically supervised use of generative AI in formative assessment can support the development of soft skills, provided that its implementation avoids technological dependence and promotes reflective, autonomous, and collaborative learning.