Diseño, validación y baremación de un instrumento para evaluar los predictores de la lectura

Abstract

In recent decades, evidence has accumulated that indicates the importance of the development of skills such as phonological awareness in Early Childhood Education. This skills can favor the subsequent learning of reading in all students. We also know that variables such as phonological awareness or naming speed predict the risk of specific learning difficulties in reading (dyslexia) years later and so it is convenient to detect children who perform poorly in these areas. The educational response on the predictors of reading is especially interesting for those children who lag behind in their acquisition. They will show a greater probability of presenting difficulties reading of words accuracy and fluently. This is where our research is inserted, whose general objective is to design, validate and measure an instrument that measures low performance in two widely established predictors of reading (phonological awareness and naming speed). This instrument that we will call EVAPRELEC (Instrument for the Evaluation of Reading Predictors) has tried to fulfill in its design some qualities that make its application funtional. For its development, a series of phases were followed based on what was described by the authors specialized in the construction of tests. First, those variables that are considered predictors of reading were selected and which, as mentioned before, were phonological awareness and naming speed. Subsequently, an item bank was prepared to measure the variables indicated above and the test design was carried out, which finally had seven tasks and 63 items. The test had content validity through the development of a group of experts. In the next phase, a pilot study was developed in which the initial version of the instrument was tested and in which 127 students enrolled in the third year of Early Childhood Education participated. After the analyzes related to the reliability and validity of this first version, the test was reduced to four tasks and 23 items. Subsequently, the study of the final version of the test was carried out with the improvements incorporated after the realization of the pilot study. 149 children enrolled in the third year of Early Childhood Education participated. Evidence of moderate to excellent reliability was found for the different tasks, as well as evidence that is considered sufficient to support the fact that the test shows content validity, internal validity and concurrent validity. In addition, the test scales were established. Finally, the strengths and weaknesses of the test were discussed.