Bulletin of Computational Applied Mathematics (Bull CompAMa)
A student's academic index using multivariate techniques
Maria Elena López Zamora, Lelly Useche, Carlos Montes Rodríguez
This work faced the construction of an index composed by the factors that influence the student's academic performance. To fulfill this objective, the information from students was collected by means of a questionnaire and later, the exploratory analysis of data was made, where the relevant characteristics of the sample were detected. To study the relations between the factors involved, a multiple correspondence analysis was made and the index was constructed by means of a factorial analysis. The index showed that the student's high performance is mainly associated to data related to academic achievement and personal relationships, owed to their factorial charges of 0.787 y 0.708 respectively. The compound index can be applied in other institutions, serving as basis for decision making and reforms in the educative sector, in order to obtain a lower desertion and a higher academic performance, among other aspects of the educational problems.
Keywords: Index; factor analysis; correspondence analysis; linear combinations; academic performance.