The Student Outcomes Survey collects information on the outcomes of two groups of students: those who completed a qualification (graduates) and those who completed only part of a course and then left the VET system (module completers). For many years several students surveyed as module completers have stated that they had gained a qualification and graduated (self-reporting graduates), and these students have been reclassified as graduates for reporting purposes. This biases the survey results as it was found that about two-thirds of them may not have completed the qualification after all. This paper outlines a predictive model which will be used for 2012 Student Outcomes Survey reporting and will ensure that this problem is resolved. As this methodology will change estimates from previous surveys substantially, they will be back cast to 2005 using the new model.
NCVER