Guo-liang Yang

A general framework for describing diversity within systems and similarity between systems with applications in informetrics

Building on the ideas of Stirling (J R Soc Interface, 4(15), 707–719, 2007) and Rafols and Meyer (Scientometrics, 82(2), 263–287, 2010), we borrow models of genetic distance based on gene diversity and propose a general conceptual framework to investigate the diversity within and among systems and the similarity between systems. This framework can be used to reveal the relationship of systems weighted by the similarity of the corresponding categories. Application of the framework to scientometrics is explored to evaluate the balance of national disciplinary structures, and the homogeneity of disciplinary structures between countries.

Cross-efficiency aggregation in DEA models using the evidential-reasoning approach

Cross-efficiency in data envelopment analysis (DEA) models is an effective way to rank decision-making units (DMUs). The common methods to aggregate cross-efficiency do not consider the preference structure of the decision maker (DM). When a DM’s preference structure does not satisfy the “additive independence” condition, a new aggregation method must be proposed. This paper uses the evidential-reasoning (ER) approach to aggregate the cross-efficiencies obtained from cross-evaluation through the transformation of the cross-efficiency matrix to pieces of evidence. This paper provides a new method for cross-efficiency aggregation and a new way for DEA models to reflect a DM’s preference or value judgments. Additionally, this paper presents examples that demonstrate the features of cross-efficiency aggregation using the ER approach, including an empirical example of the evaluation practice of 16 basic research institutes in Chinese Academy of Sciences (CAS) in 2010 that illustrates how the ER approach can be used to aggregate the cross-efficiency matrix produced from DEA models.

A study on directional returns to scale

This paper investigates directional returns to scale (RTS) and illustrates this approach by studying biological institutes of the Chinese Academy of Sciences (CAS). Using the following input–output indicators are proposed: senior professional and technical staffs, middle level and junior professional and technical staffs, research expenditure on personnel salaries and other expenditures, SCI papers, high-quality papers, graduates training and intellectual properties, the paper uses the methods recently proposed by Yang to analyze the directional returns to scale and the effect of directional congestion of biological institutes in Chinese Academy of Sciences. Based on our analysis we come to the following findings: (1) we detect the regime of directional returns to scale (increasing, constant, decreasing) for each biological institute. This information can be used as the basis for decision-making about organizational adjustment; (2) congestion and directional congestion occurs in several biological institutes. In such cases the outputs of these institutes decrease when the inputs increase. Such institutes should analyze the underlying reason for the occurrence of congestion so that S&T resources can be used more efficiently.

Extended utility and DEA models without explicit input

In this paper, we discuss the relationship between multi-attribute utility theory and data envelopment analysis (DEA) models without explicit inputs (DEA-WEI), including dual models and some theoretical analysis of DEA-WEI models. We then propose generic DEA-WEI models with quadratic utility terms. Finally, we provide illustrative examples to show that DEA-WEI with suitable quadratic utility terms are able to reflect some value judgments that the standard DEA models cannot.

Using multi-level frontiers in DEA models to grade countries/territories

Several investigations to and approaches for categorizing academic journals/institutions/countries into different grades have been published in the past. To the best of our knowledge, most existing grading methods use either a weighted sum of quantitative indicators (including the case of one properly defined quantitative indicator) or quantified peer review results. Performance measurement is an important issue of concern for science and technology (S&T) management. In this paper we address this issue, leading to multi-level frontiers resulting from data envelopment analysis (DEA) models to grade selected countries/territories. We use research funding and researchers as input indicators, and take papers, citations and patents as output indicators. Our research results show that using DEA frontiers we can unite countries/territories by different grades. These grades reflect the corresponding countries’ levels of performance with respect to multiple inputs and outputs. Furthermore, we use papers, citations and patents as single output (with research funding and researchers as inputs), respectively, to show country/territory grade changes. In order to increase the insight in this approach, we also incorporate a simple value judgment (that the number of citations is more important than the number of papers) as prior information into the DEA models to study the resulting changes of these Countries/Territories’ performance grades.

Increasing discrimination of DEA evaluation by utilizing distances to anti-efficient frontiers

This paper develops three DEA performance indicators for the purpose of performance ranking by using the distances to both the efficient frontier and the anti-efficient frontier to enhance discrimination power of DEA analysis. The standard DEA models and the Inverted DEA models are used to identify the efficient and anti-efficient frontiers respectively. Important issues like possible intersections of the two frontiers are discussed. Empirical studies show that these indicators indeed have much more discrimination power than that of standard DEA models, and produce consistent ranks. Furthermore, three types of simulation experiments under general conditions are carried out in order to test the performance and characterization of the indicators. The simulation results show that the averages of both the Pearson and Spearman correlation coefficients between true efficiency and indicators are higher than those of true efficiency and efficiency scores estimated by the BCC model when sample size is small. Standard DEA models have not fully taken the advantage of the information implied in the data.We develop another DEA approach based on the idea of utilizing both good and bad frontiers.This paper develops three DEA performance indicators for the purpose of performance ranking.Important issues like possible intersections of the two frontiers are discussed.Empirical studies show that these indicators indeed have much more discrimination power.

Negative data in DEA

One of the important concepts of data envelopment analysis (DEA) is congestion. A decision making unit (DMU) has congestion if an increase (decrease) in one or more input(s) of the DMU leads to a decrease (increase) in one or more its output(s). The drawback of all existing congestion DEA approaches is that they are applicable only to technologies specified by non-negative data, whereas in the real world, it may exist negative data, too. Moreover, specifying the strongly and weakly most congested DMUs is a very important issue for decision makers, however, there is no study on specifying these DMUs in DEA. These two facts are motivations for creating this current study. Hence, in this research, we first introduce a DEA model to determine candidate DMUs for having congestion and then, a DEA approach is presented to detect congestion status of these DMUs. Likewise, we propose two integrated mixed integer programming (MIP)-DEA models to specify the strongly and weakly most congested DMUs. Note that the proposed approach permits the inputs and outputs that can take both negative and non-negative magnitudes. Also, a ranking DEA approach is introduced to rank the specified congested DMUs and identify the least congested DMU. Finally, a numerical example and an empirical application are presented to highlight the purpose of this research. We introduce a DEA model for identifying PConvexefficient DMUs.We recognize strong and weak congestion in the presence of negative data.We present two MIP models for specifying strongly and weakly most congested DMUs.We propose a method to rank the congested DMUs and identify the least congested DMU.Two examples are presented to highlight the application of our proposed approach.

Institutional change and the optimal size of universities

The last years have been characterized by tremendous institutional change in the university sector induced by far-reaching Higher Education Reforms (e.g. Bologna). Building on loose-coupling theory, we hypothesize that smaller universities were better able to adapt to the Higher Education Reforms of the recent years, triggering a decline in the optimal size of universities in the reform period. Using a 12-year panel data set on the inputs and outputs of German universities, we find a tremendous decrease in optimal university size, which is driven by the decline in the optimal scale for the provision of teaching activities. Our results also suggest this drop is also due to fact that the relatively higher administrative overheads of larger universities become an organizational liability in times of rapid institutional change.

Specification of a performance indicator using the evidential-reasoning approach

There are three main problems in DEA concluding overestimation, discrimination, and the handling the peculiar DMUs.The evidential-reasoning approach is used to construct a performance indicator.The constructed performance indicator is employed to combine the information from classic and inverted DEA.The proposed approach can overcome the three mentioned problems.Numerical simulation and empirical studies are conducted to present the findings. There are three primary encountered problems in classic data envelopment analysis (DEA), which they decrease the effectiveness and reliability of decision making based on the obtained information from the classic DEA. These three problems include the following issues: (1) DEA efficiency scores overestimate efficiency and they are biased; (2) In certain cases, the standard DEA models are not as useful as expected in the sense of discriminating the decision making units (DMUs); (3) Specification of the evaluated DMUs as efficient by using DEA are peculiar rather than superiority. Tackling these mentioned problems is the motivation for creating this current study. To overcome these three problems in DEA together and enhance the effectiveness and reliability of the decision-making process, this paper uses the evidential-reasoning (ER) approach to construct a performance indicator for combining the efficiency and anti-efficiency obtained by DEA and inverted DEA models, which they are used to identify the efficient and anti-efficient frontiers, respectively. Numerical simulation tests indicate that our new performance indicator is more suitable for the cases where there are relatively few DMUs to be evaluated with respect to the number of input and output indicators. Furthermore, empirical studies demonstrate that this indicator has considerably more discrimination power than that of the standard DEA models, and also it reduces overestimation and addresses peculiar DMUs, properly.

Measuring the inefficiency of Chinese research universities based on a two-stage network DEA model

Abstract This paper investigates the inefficiency and productivity of 64 Chinese research universities and their evolution over the recent period of 2010–2013, where the production process of each research university is described as a general two-stage network process. We first develop a general two-stage network directional distance framework with carryover variables to gauge the universities’ inefficiencies. Second, to study the evolution of the universities, we develop a Luenberger productivity indicator to measure the productivity changes over time, as well as decompositions. The empirical results show that the Luenberger productivity indicator increased significantly over the examined period. The productivity gains were primarily driven by improvements in efficiency. In other words, the efficiency increased on average over the period of 2010–2013. However, technical changes for many universities were below zero, which led to technology deterioration on average. Finally, based on the estimates, we propose several policy suggestions for improving efficiency and productivity.