Los equipos de entrenamientoUna muestra para el estudio de los sesgos en la fiabilidad entre parejas de observadores

Abstract

Observational methodology is a method of systematic and objective record of a spontaneous behavior in its natural context and along a continuum. The figure of the observer is essential in the use of observational instruments, because the reliability of the instrument depends, in large part, on its ability and, therefore, the training process. During training of observers should pay attention to various aspects, including contextual factors. One possible source of variation may be in the training of observers. This study presents an analysis of the effect of the sort of training on reliability between observers who train together and separately. The objective is to evaluate the influence of changes in the observers training in the agreement between observers. The participants are three couples who are trained for use the Observational Protocol for Teaching Duties in University. There are two training shifts, morning and afternoon. Two couples will be trained together and a third will be trained in different shifts. The reliability analysis is calculated in two stages, at the end of training and three weeks afterwards. The reliability test is done by Kappa Coefficient of Cohen (1960) and Generalizability Theory (TG) (Cronbach, Gleser, Nanda, and Rajaratnam, 1972).