Measuring efficiency of public higher education with dea-bcc model in sichuan province, china

Abstract

The study employed a dynamic unbalanced panel data approach to analyze the educational efficiency in public undergraduate universities. Unlike traditional static methods, the study considered the dynamic nature of educational inputs and their time-dependent impact on outputs. By utilizing unbalanced panel data, the study adequately accounted for the variations in the number of observations from each university, which accommodated the unique characteristics of the dataset. The analysis employed were two complementary methods: the DEA-BCC model and the Malmquist productivity index. The DEA-BCC model was used to evaluate the relative efficiency of universities, which considered the lagged effects of educational inputs and their impact on outputs. Additionally, the MPI analysis provided a comprehensive assessment of output changes, which captured both technical efficiency and technological progress. The study was spanned from 2018 to 2022 and covered multiple public undergraduate universities in Sichuan province, China. Time series input and output indicators data of 27 public universities were collected to ensure comprehensive and up-to-date data. Descriptive statistics and Min-Max normalization were performed by using SPSS professional edition software to enhance data comparability and visualization. Considering both static and dynamic indicators, the data were subjected to BCC and Malmquist measurements by using DEAP2.1 software. The study results reveal significant imbalances in educational efficiency and manifested in regional and institutional differences. The research provides a robust efficiency assessment and shed light on the long-term impact of educational inputs on overall university performance. The findings were of paramount importance to policymakers and university administrators, as they uncovered the effectiveness of resource allocation and policy decisions in the education sector. Moreover, the study contributed to the literature on educational efficiency by integrating dynamic considerations and unbalanced panel data and offering valuable insights into the complex long-term relationship between educational inputs and outputs value of .80.