An Analysis of Factors Affecting the Efficiency of University Startup Support in South Korea Using the DEA Double Bootstrap Approach

This study aims to identify the key factors affecting the efficiency of university startup support systems in South Korea. Using panel data on university startup support from 17 metropolitan cities and provinces from 2020 to 2024, this study applied the two-stage Data Envelopment Analysis(DEA) double bootstrap methodology proposed by Simar & Wilson(2007). In the first stage, DEA was used to measure the relative inefficiency scores of regional startup support systems, which were then corrected for statistical bias. In the second stage, a bootstrapped truncated regression was conducted using the bias-corrected inefficiency scores as the dependent variable to analyze the significant determinants of efficiency. The analysis revealed that among numerous university support activities, the number of entrepreneurship courses offered was the key factor that significantly decreased inefficiency(i.e., increased efficiency). Conversely, the ‘number of dedicated staff’ showed a tendency to increase inefficiency(i.e., hinder efficiency). Short-term, event-oriented programs such as startup competitions and camps did not have a statistically significant impact. The findings of this study strongly suggest that when allocating limited resources for university startup support, policy priorities should be set on strengthening the educational infrastructure to enhance student capabilities from a long-term perspective, rather than on one-off events or a simple expansion of administrative personnel.