Predicción del abandono universitario: variables explicativas y medidas de prevención
Abstract
El abandono de los estudios universitarios es un problema cuyos costes son altos tanto para el individuo como para la sociedad. Es por ello que la prevención del mismo es fundamental y cobra especial relevancia en el actual contexto de crisis económica. Diversos autores han desarrollado investigaciones con el objetivo de establecer modelos predictivos de este fenómeno (Castaño, Gallón, Gómez y Vásquez, 2004; Trevizán, Beltrán y Cosolito, 2009; Goldenhersh, Coria y Saino, 2011; Sánchez, 2014). En este artículo se analizan dichos trabajos, identificando las ventajas y desventajas de las metodologías más utilizadas; análisis correlacionales, análisis de regresión logística, análisis de supervivencia y minería de datos. La investigación cuyos resultados aquí se exponen, aplica la primera de las metodologías mencionadas, a fin de comprobar -en lo que respecta al fenómeno del abandono- el valor predictivo de las variables rendimiento académico previo, fecha de matriculación, rendimiento en primer curso de universidad y asistencia a clase. Los resultados confirman la relación de dichas variables con el fenómeno estudiado. Dichos resultados son consistentes con los obtenidos por diversos autores a lo largo del tiempo, y en base a ellos se proponen dos tipos de medidas; por un lado, acciones encaminadas a facilitar el diagnóstico respecto al problema del abandono, y por otro lado, medidas encaminadas a su prevención.
Abstract
University dropout is a problem whose costs are high for both individuals and society. That is the reason why prevention is essential and it is particularly important in the context of the current financial crisis. Several authors have conducted research in order to establish predictive models of this phenomenon (Castaño, Gallon, Gómez and Vásquez, 2004; Trevizán, Beltrán and Cosolito, 2009; Goldenhersh, Coria and Saino, 2011; Sánchez, 2014). This article analyses these works, identifying the advantages and disadvantages of the most used methods; correlational analysis, logistic regression, survival analysis and data mining. The research whose results are here presented applies the first of the aforementioned methodologies, with the aim to test -with regard to the phenomenon of abandonment-, the predictive value of the following variables: prior academic performance, date of enrolment, performance in the first college year and attendance. The results confirm the relationship between these variables and the phenomenon studied. These results are consistent with those obtained by several authors throughout time. To finish, they propose two types of measures based on these results: on one hand, measures to facilitate dropout diagnosis and, on the other hand, measures to prevent it.
Downloads
Downloads
Published
How to Cite
Issue
Section
License
Authors who publish in this journal accept the following conditions:
Authors conserve the copyrights and cede to the journal the right for first publication, with the work registered with the attribution licence of Creative Commons, which allows third parties to use the published work as long as they mention the authorship and the first publication in this journal.
The authors may make other independent and additional contractual agreements for the non-exclusive distribution of the version of the article published in this journal (e.g. include it an institutional repository or publish it in a book) as long as it clearly indicates that the work was first published in this journal.
Authors are allowed and indeed recommended to publish their work on the internet (for example on institutional or personal pages) before and during the revision and publication process, because it can lead to productive exchanges and greater and faster dissemination of the published work (see The Effect of Open Access).