Profiling students at risk of dropout at a university in South Africa

Background: Higher education authorities continue to be concerned about dropout rates among university students. Dropping out affects cost efficiency and tarnishes the reputation of institution. As a result, profiling at risk student is crucial.
 Purpose: To build profile students using administrative data.
 Methods: The researcher employed data mining technique in which predictors were chosen based on their weight evidence (WOE) information value (IV). then used dropout.
 Findings: According findings, if was born years 1931 1967 or 1994 2001; fails more than four modules year with participation average mark 43percent less; has joined second academic year.
 Conclusions: This study concludes that matric scores had no bearing chance dropping for cohorts who attended institution between 2008 2018. However, variables number failed, marks, entry level birth building 2018.