Academic journal article The Spanish Journal of Psychology

Variables That Predict Academic Achievement in the Spanish Compulsory Secondary Educational System: A Longitudinal, Multi-Level Analysis

Academic journal article The Spanish Journal of Psychology

Variables That Predict Academic Achievement in the Spanish Compulsory Secondary Educational System: A Longitudinal, Multi-Level Analysis

Article excerpt

This article presents a study whose objective was to identify certain personal and institutional variables that are associated with academic achievement among Spanish, secondary school students, and to analyze their influence on the progress of those students over the course of that stage of their education. In order to do this, a longitudinal, multi-level study was conducted in which a total of 965 students and 27 different schools were evaluated in Language, Math and Social Science at three different times (beginning, middle and end of the period). The results show progress in all the schools and in all areas. As for the personal, student variables, the longitudinal, HLM analyses confirmed the importance of sex and sociocultural background and, distinguishing it from other studies, also the predictive capacity of meta-cognitive abilities and learning strategies on success in school. On the institutional level, the school climate and teachers' expectations of their students were the most relevant of the variables studied. The size of the school, the percentage of students who repeat grades, and the leadership of the administration also explained a portion of the variance in some areas.

Keywords: academic achievement; predictor variables; secondary education; longitudinal multilevel analysis; school effectiveness

En el artículo se presenta un estudio cuyo objetivo es identificar determinadas variables personales y de centro asociadas con el rendimiento académico de estudiantes de secundaria españoles y analizar su influencia en el progreso de los alumnos a lo largo de la esta etapa. Para ello, Se realizó un estudio multinivel longitudinal en el que se evaluó a un total de 965 estudiantes de 27 centros distintos en Lengua, Matemáticas y Ciencias Sociales, en tres momentos (inicio, mitad y final de la etapa). Los resultados mostraron progreso en el conjunto de los centros en todas las áreas. Los análisis HLM longitudinales confirmaron en el nivel personal la importancia del sexo y el nivel sociocultural y, a diferencia de otros estudios, también la capacidad predictiva de las habilidades metacognitivas y las estrategias de aprendizaje. En el nivel de escuela, el clima escolar y las expectativas del profesorado hacia los estudiantes fueron las variables más relevantes. El tamaño del centro, el porcentaje de repetidores y el liderazgo del equipo directivo explicaron también una proporción de la varianza en algunas áreas.

Palabras clave: rendimiento escolar, variables predictoras, educación secundaria, análisis mutinivel longitudinal, eficacia de la escuela

Worries about the academic achievement of students have been of constant concern to research on education. Studies about the effectiveness of schools have centered their interest on identifying the factors that are best able to predict achievement in order to intervene early in the process of education to curb academic failure. Fortunately, the procedures for assessing achievement have become increasingly strong, from both a theoretical and methodological perspective. The four factors for evaluation in Stufflebeam and Shinkfield's global model (1985) - context, input, process, and output - are now considered to be essential elements of any evaluation. However, this perspective has been widely expanded upon by studies of the effectiveness of schools that were conducted using other, also valuable, models that have recently begun to include longitudinal analysis (Braun, 2005; Lissitz, 2005). The findings of those studies have been accompanied by methodological developments on multi-level analysis, which allows researchers to view the structural hierarchy of data on educational from a longitudinal perspective (Goldstein, 2003; Hox, 2002; Raudenbush & Bryk, 2002; Sellström & Bremberg, 2006; Zvoch & Stevens, 2006).

Still, only few research studies have used longitudinal, multi-level analysis and among them, even fewer have followed the same students through their education. …

Search by... Author
Show... All Results Primary Sources Peer-reviewed


An unknown error has occurred. Please click the button below to reload the page. If the problem persists, please try again in a little while.