Validity of a rubric to assess graduate research skills; [Validez de una rúbrica para evaluar competencias investigativas en postgrado]

Abstract

Research competencies are the domain of cognitive attitudes put into practice by students. The objective of this study was to analyze the construct validity of an analytical rubric to assess the research competencies of graduate students. Consulting professors assessed 260 theses using this rubric. Construct validity analysis was performed using a cross-validation process between exploratory, which is a statistical technique that accurately explores the underlying dimensions of the latent variables of the observed variables and the test factor analysis, which frames the structural equation models, analyzing the relationships between the indicators of the observed variables and the latent variables. The results of the EFA showed correspondence with the theoretical proposal, where a single factor explained 51% of the variance represented by all items. Evaluation of this model using the EFA revealed a good fit (X2/gl: 2,197; RMSA: 0.085; GFI: 0.902; IFC: 0.932; TLI: 0.912) and evidence of convergent validity (CFE>0.50; CC: 0.902; VME: 0.464). The reliability of the rubric was optimal (α=0.903). It is concluded that the formulated construct is represented within the structure of the instrument. © Licencia de Creative Commons Atribución 4.0 Internacional (CC BY 4.0) https://creativecommons.org/licenses/by/4.0/deed.es