It is important for policymakers and managers of higher education institutions knowing how well their universities are operating. This article aimed to show that data envelopment analysis (DEA) can be an excellent benchmarking instrument in higher education. First, by using several inputs and outputs at the institutional level, DEA can identify technically efficient institutions that may work as a benchmark in the sector becoming a reliable tool for ranking universities. Second, a bootstrapped–truncated regression allows us to understand the factors affecting technical efficiency of the institutions under evaluation. The case of Spanish public universities is taken as an example to verify the usefulness of the proposed methods. Our empirical strategy was based on a two-stage procedure to evaluate their internal efficiency in the provision of teaching and research. In the first stage, we estimated a technical efficiency score for each university. The average efficiency among Spanish universities was about 92%. In the second stage, we regressed the efficiency scores against a set of covariates to investigate their association with the level of university (in)efficiency. We found that universities with a higher percentage of grantees tend to be less inefficient, and a higher percentage of academics with tenure enhances the productive efficiency of the Spanish higher education sector. Finally, we computed Spearman’s rank correlations between DEA efficiency scores and the classification of Spanish institutions in university rankings such as the SCImago and Shanghai rankings. The results revealed that the ranking positions given by DEA scores to Spanish universities matched their positions in recognized rankings.

中文翻译：

---

基于数据包络分析并以西班牙案例为例的高等教育系统的技术效率绩效

对于高校的决策者和管理者来说，了解其大学的运作状况非常重要。本文旨在说明数据包络分析（DEA）可以成为高等教育中的出色基准工具。首先，通过在机构级别使用几种投入和产出，DEA可以确定技术上高效的机构，这些机构可以作为该行业的基准，成为对大学进行排名的可靠工具。其次，自举式截断回归分析使我们能够了解影响评估机构技术效率的因素。以西班牙公立大学为例，验证了所提方法的有效性。我们的经验策略基于两个阶段的程序，以评估其在提供教学和研究方面的内部效率。在第一阶段，我们估算了每所大学的技术效率得分。西班牙大学的平均效率约为92％。在第二阶段，我们针对一组协变量对效率得分进行回归，以研究它们与大学效率水平之间的关联。我们发现，受助人比例较高的大学往往效率较低，而任职期限较高的学者比例较高，可以提高西班牙高等教育部门的生产效率。最后，我们计算了DEA效率得分与西班牙大学在SCImago和Shanghai排名中的机构分类之间的Spearman等级相关性。