Measuring the efficiency of community colleges using super efficiency approach for the case of non-discretionary factors in data envelopment analysis with sensitivity analysis

Malaysian community colleges play a significant role in achieving Malaysia's vision to be a developed country in 2020. For this reason, their efficiency should appropriately be measured. However, their efficiency measurement using a conventional data envelopment analysis (DEA) model is not appropriate since some of their inputs; e.g., entrant and enrolment students are non-discretionary while a part of output of their graduate employability is discretionary. This paper thus proposes an alternative approach of super efficiency slack-based measure for the case of non-discretionary factors in DEA. The proposed approach was used to evaluate the efficiency of 25 main campuses of Malaysian community colleges from 2012 to 2013. The results support the decision maker of Malaysian community colleges to discriminate and rank efficient and inefficient community colleges in the presence of both super efficiency and non-discretionary factors. The significance of inputs-outputs on efficiency status was tested by sensitivity analysis.