Discovery of Academic Performance Patterns Using Decision Trees in Generic Skills of Professional Education

The objective of this study aimed to discover academic performance patterns in generic competences of the students in professional programs in the 2011-12 Saber Pro Exam, based on sociodemographic, economic, academic and institutional data stored in the databases of the Instituto Colombiano para la...

Full description

Autores:
Timarán Pereira, Silvio Ricardo; Universidad de Nariño
Hernández Arteaga, Isabel; Universidad Cooperativa de Colombia
Caicedo Zambrano, Segundo Javier; Universidad de Nariño
Hidalgo Troya, Arsenio; Universidad de Nariño
Alvarado Pérez, Juan Carlos; Universidad Cooperativa de Colombia
Tipo de recurso:
Book
Fecha de publicación:
2016
Institución:
Universidad Cooperativa de Colombia
Repositorio:
Repositorio UCC
Idioma:
spa
OAI Identifier:
oai:repository.ucc.edu.co:20.500.12494/44257
Acceso en línea:
https://ediciones.ucc.edu.co/index.php/ucc/catalog/book/36
https://hdl.handle.net/20.500.12494/44257
Palabra clave:
aceleradores psicosociales
catalizadores de la violencia
violencia de género
Violencia de género
Psicología
gender violence
catalysts of violence
Psychosocial accelerators
Gender violence
Psychology
Rights
openAccess
License
http://purl.org/coar/access_right/c_abf2
Description
Summary:The objective of this study aimed to discover academic performance patterns in generic competences of the students in professional programs in the 2011-12 Saber Pro Exam, based on sociodemographic, economic, academic and institutional data stored in the databases of the Instituto Colombiano para la Evaluación de la Educación (icfes), using data mining techniques. The studies carried out until now to analyze the 2011-12 Saber Pro test results are based on information processed using statistical research, in which fundamentally primary variables and relationships are considered, without taking into account the true interrelationships that are generally hidden and can only be discovered using a more complex data treatment, which is possible with the above-mentioned data. Therefore, the crisp-dm methodology was applied and through the use of clean data repositories that were transformed for each competence, patterns associated with good or poor academic performance were discovered, using the classification technique based on decision trees.In the patterns that were found, institutional accreditation and study modality stand out as two important attributes associated with the academic performance of the students in the 2011-12 Saber Pro tests, in the four generic competences. The findings will be incorporated into what already exists and can be integrated into decision-making processes of the icfes and of the governmental and academic institutions that oversee the quality of higher education.