Choosing a Control Chart to Monitor University Student Dropout

Student dropout is an important event in the quality assurance of academic programs in universities. Using traditional control charts to monitor this phenomenon is affected by the excessive variability of the data, which leads to noncompliance with the assumptions required for their application. For this reason, this paper develops a methodology for choosing a control chart for monitoring university student dropout, considering overdispersed data. The methodology consists of four steps: (i) Determination of the variable of interest, (ii) Overdispersion diagnosis, (iii) Development of control charts, and (iv) Analysis of control charts results. The methodology is applied in a real case, and the established choosing criteria facilitated decision-making on the control chart suitable for monitoring.