The aim of this paper is to analyze the educational gap between 250 local governments, by their location and size, in terms of the efficiency of education. For the analysis, this paper employed the data envelopment analysis(DEA), which can consider input and output factors simultaneously, as a research method. Input factor included student number per teacher, student number per class, student number per staff, and donation fund per student, whereas college entrance rate and employment rate were used as output factor. The data were collected through ‘the school information’ website. Research target was confined to high schools, taking into consideration the variables used for output factor. As a result of analysis,while the high schools in local governments around the capital area showed the high density of students in terms of input factor, college entrance rate and employment rate were higher in the high schools of non-capital area. An efficiency score was also higher in the schools of the non-capital area in both aspects of BCC and CCR models. By the size of local governments, high schools in agricultural-type and small-sized local governments were higher in the efficiency of education, compared to those in populated local governments.