The Efficiency of University Research is Enhanced Primarily by the Level of Political and Civil Liberties

Results from a Data Envelopment Analysis (DEA) reveal that the variable Voice and Accountability, one of the indicators of good governance, significantly correlates with research efficiency. This suggests that factors such as participation in electing government officials, freedom of speech, freedom of association, and freedom of the press—constituting the Voice and Accountability variable—play a crucial role in boosting the efficiency of university research.

The main contribution of Vladimír Holý’s study lies in demonstrating that the right to decide and accountability have a fundamental impact on the efficiency of university research. The Voice and Accountability variable is a parameter of so-called good governance, which ensures a stable state acting in accordance with the law and predictability. Such governance treats all citizens and businesses equally, protects private property, and operates within a system of checks and balances that prevent institutions from abusing power. This environment is considered foundational for economic growth, including the educational sector.

The concept of good governance is based on the Worldwide Governance Indicators, published since 1996 under the auspices of the World Bank. These indicators assess governance globally across nearly all countries based on six criteria:

  • Voice and Accountability,
  • Political Stability and Absence of Violence,
  • Government Effectiveness,
  • Regulatory Quality,
  • Rule of Law, and
  • Control of Corruption.

The Right to Decide and Accountability Drive Research Forward

All models employed in Vladimír Holý’s study consistently found that the Voice and Accountability variable (the right to decide and accountability) has a significantly positive impact on research efficiency. The parameter of accountability explicitly states that “individuals in decision-making positions are answerable to the public and relevant institutions, providing sufficient information to allow understanding and oversight.”

The study further revealed that the Government Effectiveness variable does not significantly correlate with research efficiency. “The results of the government effectiveness index are mixed—while some models suggest it has a significant positive impact, others consider it negligible,” Holý’s study notes.

Analysis of Scientific Publications Issued in the European Union from 2005 to 2020

As a case study, Dr. Holý examined the efficiency of university research in European Union countries, analyzing scientific publications from 2005 to 2020. For inputs, he used gross domestic product (in the research and development sector) and the number of researchers involved, thus accounting for financial and human resources. “It is important to adjust prices for purchasing power parity to ensure meaningful comparisons between countries with different purchasing power levels. This variable reflects financial resources,” Holý explains in his study.

For outputs, he used the number of publications and citations to reflect the quantity and quality of scientific research. “The number of publications and citations represents the number of articles published, reviews, and conference participations,” the study states.

How to Work with an Alternative DEA Model in the Future? Thoughtfully

Given the controversies surrounding the second stage of DEA and its traditional use by researchers, Holý’s study emphasizes the importance of regular robustness checks and avoiding reliance on results from a single model. In this regard, the dynamic ranking model proves useful, as it differs from traditional panel analysis methods and can offer a completely different perspective.

However, like traditional second-stage models, it risks incorrect specification of the data-generating process. Moreover, it can only be used for recurring evaluations, such as annual research efficiency assessments. While dynamic evaluation in the second DEA stage may not be a perfect solution for all situations, it is a valuable addition to the toolkit of operational research.

You can read Dr. Vladimír Holý’s original study at this link.