Methodological limitations and feasible solutions in the assessment and grade calculation obtained through the rubrics as skills assessment strategies of higher education students and other levels

Abstract

The design and elaboration of rubrics as assessment strategies in higher education and other educational levels, as well as the numerous benefits of their use, are topics widely treated in the scientific literature. However, the aspects related to the calculation of the scores that derive from its use, as well as the reliability and validity of the rubrics have received less attention. The present work, therefore, focuses on the exposition of a set of methodological limitations related to the rubrics and the proposal of feasible solutions to overcome them. The final considerations of the work focus on proposals related to rubrics where the scores are not equivalent to the grading system of the reference country, as well as to its weighting. We also emphasize the biases that derive from the use of correlation and reliability coefficients in the use of intra- and inter-observer reliability and in the validity of content and the desirability to replace with agreement coefficients.