Tutoría entre iguales y su relación con el rendimiento matemático de estudiantes de la licenciatura en Actuaria

Zusammenfassung

The subject of Differential and Integral Calculus has a highly significant weight in the curriculum, as it is the subject with the highest number of weekly hours; greater number of credits and possess the seriation that is related to more subjects in the curricular map. However, it is also the subject with the highest index of suspensions. In order to reduce the rate of non-approval of the subject of Differential and Integral Calculus I in the actuary course of the Facultad de Estudios Superiores Acatlán, of the Universidad Nacional Autónoma de México, a strategy of tutoring between equals denominated Program Preventive in Matters with High Reprobation Rates. In order to determine the effect of this Peer Tutoring Program on students' academic performance, a pre-experimental cohort design was used, with a prospective type with pre and post measures, in which the independent variable was defined as Didactic methodology with levels traditional/tutoring between peers and the dependent variable Academic performance, expressed with the final grades of the subject. The final grades of the subject of Differential and Integral Calculus I were collected through the consultation of the Integral System of Institutional Statistics database, which were exported in flat format to be processed in the computer program R, in Which was carried out the corresponding statistical analyzes. With the results of the analyzes, we were able to verify that the students who were submitted to the tutorial intervention of peers, had a greater performance in the matter of Differential Calculus and Integral I, than the students who studied it with a traditional methodology. By analyzing categorical data, specifically with the non-parametric Mann-Whitney U test, this yield difference was shown to be statistically significant, with a confidence level of 95%. Subsequently, the test of independence � for two populations was applied to the data organized in a contingency table. The result indicated the existence of a statistically significant relationship between the variable Didactic methodology and the variable Academic performance. Consequently, it was determined the level of association between variables with the calculation of the coefficients �, � of contingencies and � of Cramér, as well as their graphical representations through association diagrams and mosaic diagrams, were quantified through nonparametric methods. In general, a positive association level of 20%, already rounded value, was obtained. Once this was done, a generalized linear model, particularly a binary logistic regression model, was fitted for predictive purposes. This model showed a probability of passing the subject of 66% when the subject is taken under peer tutoring and 57% when taken in the traditional way. The research also reports the lack of studies on the effect of peer tutoring on academic performance and poses the possibility, with the aim of overcoming the limitations of the present study, of the design of a quasi experimental or experimental research when the conditions allow it. Finally, it is proposed to extend research to aspects such as the effect of peer tutoring on dropout, terminal efficiency and titration rates.