La corrección del azar en pruebas objetivas: un enfoque basado en una nueva teoría de estados finitos

Abstract

A model of objective test solving behavior is presented and scoring formulae are derived therefrom. The model is a finite state theory considering independent response to each alternative within each item and, therefore, allowing to account for partial knowledge. Probability summation arising from the hypothesized behaviour as well as the characteristics of the format of the test are both taken into account in deriving the scoring formulae. The behaviour of the conventional scoring formulae are analyzed in the light of this model, by means of simulation techniques, and the suitability of the formulae for achieving the goals of testing is also discussed. Finally, some guidelines of work are suggested which are concerned with the extension and empirical validation of the model presented here.