Estudio de los juicios metacognitivos en la solución de ítems mediante simulación
Fecha
2024
Autores
Chacón Camacho, Diana Lucía
Título de la revista
ISSN de la revista
Título del volumen
Editor
Resumen
Uno de los procesos metacognitivos que las personas realizan es crear estimaciones sobre su rendimiento en la solución de ítems, estas estimaciones se conocen como juicios metacognitivos y pueden presentar sesgos respecto al desempeño real por diversos factores como la ansiedad, el género entre otros. El modelo de Rasch permite estimar la probabilidad de acertar un ítem en una prueba estandarizada basándose en la habilidad del sujeto resolutor en el constructo y la dificultad del ítem. El presente estudio supone que los humanos realizamos un proceso análogo al descrito por la ecuación de Rasch para establecer los juicios metacognitivos, por lo que propone un modelo computacional de la relación entre el desempeño de ítems en pruebas estandarizadas de matemática y los juicios metacognitivos de la predicción del rendimiento. Dado el modelo computacional basado en la ecuación de Rasch se realizaron dos simulaciones computacionales en la cuales se agregaron errores aditivos a los parámetros de habilidad del sujeto y dificultad del ítem y se determina la estimación metacognitiva del desempeño en los ítems. Finalmente se compararon juicios metacognitivos emitidos por estudiantes en pruebas anteriores con una simulación encontrándose resultados similares a la de los datos reales.
One of the metacognitive processes that people perform is creating estimations about their performance in solving items; these estimations are known as metacognitive judgments and may present biases regarding actual performance due to various factors such as anxiety, gender, among others. The Rasch model allows estimating the probability of correctly answering an item in a standardized test based on the subject's ability in the construct and the item’s difficulty. The present study assumes that humans undergo a process analogous to the one described by the Rasch equation to establish metacognitive judgments, therefore it proposes a computation model of the relationship between item performance on standardized math tests and metacognitive judgments of performance prediction. Given the computational model based on the Rasch equation, two computational simulations were conducted, adding additive errors to the subject's ability parameters and item difficulty, for determining the metacognitive estimation of item performance. Finally, metacognitive judgments issued by students in previous tests were compared with a simulation, finding similar results to those of real data.
One of the metacognitive processes that people perform is creating estimations about their performance in solving items; these estimations are known as metacognitive judgments and may present biases regarding actual performance due to various factors such as anxiety, gender, among others. The Rasch model allows estimating the probability of correctly answering an item in a standardized test based on the subject's ability in the construct and the item’s difficulty. The present study assumes that humans undergo a process analogous to the one described by the Rasch equation to establish metacognitive judgments, therefore it proposes a computation model of the relationship between item performance on standardized math tests and metacognitive judgments of performance prediction. Given the computational model based on the Rasch equation, two computational simulations were conducted, adding additive errors to the subject's ability parameters and item difficulty, for determining the metacognitive estimation of item performance. Finally, metacognitive judgments issued by students in previous tests were compared with a simulation, finding similar results to those of real data.