Desenvolvimento e avaliação de um programa de ensino informatizado para compreensão leitora

Abstract

In Applied Behavior Analysis, computerized teaching programs have been developed for the teaching of reading. The objective of this research was to develop a computerized teaching program to evaluate and exercise the reading of texts. We made a theoretical review in the Behavior Analysis literature, and thus we proposed an operational analysis of the concept of reading comprehension, as well as the theoretical foundation to elaborate didactic materials aimed at teaching the comprehension of texts. We developed a computerized program for teaching reading comprehension, called the Fable Reading Program. The Program employs different teaching resources such as questions of literal and inferential comprehension, immediate reinforcement, and increasing and gradual difficulty level. To evaluate how the Program functions, three studies were performed. The first sought to evaluate the effects of the Program on the reading comprehension of the participants. Twenty-four children, with an average age of 10 years, enrolled in the 5th grade of a public school in the city of São Carlos-SP participated in the study. Participants were divided into experimental group (n = 14) and control group (n = 10). They were assessed by a standardized reading comprehension test, PROCOMLE, and by the Program's test books. Results indicated performance improvements of the experimental group. The second study also sought to evaluate the effects of the Program on the reading comprehension of the participants. Sixtysix students from the same school, grade and age group of the previous study participated in this study but composing a single experimental group. We assessed their performance in the Program's test books. In the first block of the Program, we could observe that the average of scores fell in books that introduced inferential questions, indicating the need to include cues that facilitate learning to make inferences in reading; In Blocks B, C and D, the average number of scores remained stable, suggesting that, despite the increasing complexity of books, participants had gains in learning. The third study investigated which were the most difficult questions for students to make adjustments and improvements to the Program based on the data. Data from the 66 participants from the previous study were analyzed. From descriptive statistics, it was found that throughout the Program, in the version tested in this research, there were a total of 24 questions that proved to be the most difficult for participants. Once the target questions were identified, a qualitative analysis was performed describing the stimulus control relationships involved in reading the narratives and resolving the questions. Considering the challenges for teaching reading comprehension, this work was a starting point for future research to continue, as much remains to be improved and tested. More learning resources need to be incorporated into the Program to ensure its effectiveness as a teaching procedure.