El presente trabajo investiga el rendimiento académico de los estudiantes universitarios. Se utilizan modelos para variables enteras para analizar el desempeño de la cohorte de ingresantes 2001 de la Facultad de Ciencias Económicas de la UNLP; los resultados obtenidos indican que el modelo Binomial Negativo inflado en cero (ZINB) es el más apropiado. El mismo permite diferenciar entre aquellos factores personales y socioeconómicos que influyen sobre la probabilidad de pertenecer al grupo que no aprueba materias y aquellos factores que influyen sobre el número esperado de la variable materias aprobadas. / This paper examines the academic performance of students at university level. Different count variable models are considered to evaluate academic achievement of the cohort of students who started at Faculty of Economics of La Plata National University in 2001; results suggest that Zero Inflated Negative Binomial model fits better than other. This model allows analyzing in a separate way those personal and socioeconomic factors which have an influence on the probability of belonging to the non-passing group and those factors which have an influence on the expected value of passing a higher number of subjects.
| Identifer | oai:union.ndltd.org:SEDICI/oai:sedici.unlp.edu.ar:10915/3336 |
| Date | January 2007 |
| Creators | Pron, Julieta |
| Contributors | Porto, Alberto |
| Source Sets | Universidad Nacional de La Plata, Sedici |
| Language | Spanish |
| Detected Language | Spanish |
| Type | Tesis, Tesis de maestria |
| Rights | http://creativecommons.org/licenses/by/3.0/, Creative Commons Attribution 3.0 Unported (CC BY 3.0) |
Page generated in 0.0016 seconds