La competencia lectora en internet en adolescentes argentinos: Evidencias de confiabilidad y validez de la prueba WebLEC

Abstract

La lectura competente en internet requiere habilidades específicas, por lo que resulta relevante contar con herramientas para medirla. Este trabajo evaluó cualidades psicométricas (evidencias de consistencia interna, de validez convergente y de grupos contrastados) de la prueba WebLEC adaptada al contexto rioplatense. WebLEC evalúa la competencia lectora en internet a partir de tareas de búsqueda, integración y evaluación de información en entornos web. Ciento cuarenta y cuatro estudiantes de nivel secundario del área metropolitana de Buenos Aires (Medad = 14.03 años, DE = 0.44) completaron una prueba de comprensión de texto en papel y la prueba WebLEC en una PC o notebook. El WebLEC provee un índice global de desempeño y dos índices de navegación. Los niveles de consistencia interna resultaron aceptables. Se encontraron correlaciones de los tres índices WebLEC con la prueba de comprensión de texto en papel. Asimismo, un modelo de mediación mostró, en sintonía con los antecedentes, que la relación entre el desempeño en comprensión en papel y WebLEC quedó parcialmente mediada por estrategias específicas de la lectura electrónica (eficiencia del patrón de navegación al completar la prueba). Finalmente, el análisis de grupos contrastados mostró que el desempeño fue menor para la muestra de nivel secundario que el del universitario en los tres índices WebLEC. En síntesis, WebLEC Argentina constituye el primer instrumento para medir la competencia lectora en internet en estudiantes en nivel secundario basado en evidencias psicométricas adaptado al contexto local.

Reading is a key competence for the development of individuals in knowledge societies. People are increasingly reading on the internet for a variety of purposes, such as entertainment, information, education, searching for specific data, solving practical problems, among others. In this context, reading literacy is defined as the ability to understand, use, evaluate, reflect, and engage with written texts to achieve goals, develop knowledge, personal potential and participation in society. Proficient reading on the internet requires specific skills within specific environments that complement those of offline reading. Therefore, it is relevant to have a reliable measure of this construct. Argentina does not yet have a validated instrument to assess internet reading literacy in high school students. The WebLEC test assesses reading literacy through 28 items distributed in four scenarios that simulate a web environment: Forum, Wikipedia, Service Portal, and Search Engine. The test is administered on a computer and its application lasts, on average, 45 minutes. Each item proposes a task based on the reading of texts contained in a specific scenario. Based on the PISA framework, literacy is measured through performance in three types of tasks: locate information (access and retrieve information within the text, search and select relevant text), integrate and interpret (represent literal meaning, integrate and generate inferences), and evaluate and reflect (weigh the quality and credibility of information, reflect on content and form of texts, detect and regulate conflicts). The goal of this study was to examine the psychometric qualities of the WebLEC test, in its version adapted to the Rio de la Plata context. This work analyzed the internal consistency of the instrument, its degree of convergence with a paper-based text comprehension test (evidence of convergent validity) and its sensitivity to detect differences between students according to their educational level (evidence of validity of contrasted groups). Additionally, it was examined adequacy into a model that includes navigation efficiency as a mediator between the influence of text comprehension and performance in hypertextual reading tasks. One hundred and forty-four high school students from the Buenos Aires metropolitan area (61 % male, Mage = 14.03 years, SD = .44) voluntarily completed a paper-based text comprehension test (TLC-II) and the WebLEC test on a PC or notebook. WebLEC provides a general index, calculated from response accuracy, and two navigation indices, calculated from the log files. The test proved acceptable levels of internal consistency, α = .731, ω = .732. As for convergent validity, the analysis revealed significant correlations between the three WebLEC indices and the paper-based text comprehension test, .25 < r > .45, p < .002. Likewise, the mediation model showed that text comprehension significantly predicted WebLEC performance (74.8 %), z = 3.57, p < .001, with part of that effect being mediated by navigation efficiency while completing the test (25.2 %), z = 2.39, p = .017. Finally, the results showed that the performance of the high school sample was lower than that observed in a college freshmen students’ sample on all three WebLEC indices, demonstrating that the instrument can identify differences between contrasting groups, p < .001, d > .61. To sum, WebLEC showed a reliable internal structure, moderate to strong associations with a paper-based or offline text comprehension test, and sensitivity to discriminate performance as a function of education level (high school vs. freshmen college students). The results also showed that the linear relationship between WebLEC and the paper-based test was mediated by participants’ navigation efficiency, in accordance with prior literature. In summary, the psychometric evidence indicates that WebLEC Argentina is an adequate tool for measuring internet reading proficiency in secondary school students, being the first standardized instrument of its kind in the local context.