Abstract

The present study integrates aspects of cognitive psychology and psychometrics (Rupp & Mislevy, 2006) for the analysis of the sources of difficulty in the area of quantitative skills (HC, for its acronym in Spanish) of the Test of Skills and Knowledge Basic (EXHCOBA, for its acronym in Spanish) (Backhoff, Tirado, & Larrazolo, 2001), using the Linear Logistic Test Model (LLTM) developed by Fischer (1973). To achieve this purpose, a group of experts proposed fourteen operations to solve items in the HC area based on the results of retrospective and concurrent verbal reports obtained from a group of students of secondary education. In addition, the cognitive structure determined by experts was analysed, and the dimensionality and the adjustment of the ítems to Rasch model and LLTM was assessed. The results showed that the data fit moderately to the LLTM with a Q-matrix of fourteen attributes and, also that, a refinement of the operations to resolve the items could explain better the sources of difficulty and the performance of examinees in the test.