According to Fig.4.1, there are four input factors in university research pro-
duction—human resources (or research staff), research grants, research platforms,
and research projects, and two primary outputs—research outcomes and trained
students. There are three channels to connect inputs and outputs. Thefirst channel,
research staff get research funding and then conduct research activities, and then
produce some research outcomes. At the meantime, in those research teams con-
sisted of advisors (or supervisors),^2 some master or doctoral students would join the
projects and conduct research. During the process, advisor would instruct the
postgraduate students how to do research, which makes this process function as
student training. Thus, the master and doctoral graduates are also the by-products to
some extent. The second way, even without research funding, research staff can also
conduct some studies on their own, to publish some articles and train some stu-
dents. This usually happens in pure sciences like mathematics, and humanities like
literature studies. The third way, a high-level research platform gathers money and
researchers together, to conduct high-level research and train young researchers. In
China, a well-recognized fact is that high-level research is often supported by
advanced platforms, such as National Key Laboratories, National Kay Base for
Humanities and Social Sciences Research, National Key Disciplines. These
national-level platforms are supported by much more sufficient funds from the
central government, and are built well in infrastructure, and therefore attract top
researchers to work there and produce cutting-edge knowledge.
4.1.2.3 Literature Concerning Evaluation on University Research
Efficiency
Concerning studies on university research efficiency to date, we can divide them into
two types according to their research methodologies and focuses. Thefirst type
employs theoretical approach to analyze and exemplify some indicator systems of
university research evaluation in detail, and give each indicator a feasible weighting
(Wang and Li 2000 ). However, this type of studies merely merits in theories, but lacks
convincing empirical evidences. Besides, the proposed indicator systems are usually
hard to put into practices, due to data unavailability and problems in statistical
techniques. The second type usually applies quantitative methods, such as DEA, to
empirically evaluate university research efficiency (see Table4.1). This type of
studies partly involves the construction of indicator system, but it usually overem-
phasizes the evaluation methods and quantitative models, without laying solid
background on the construction of indicator system, resulting into the problem that
(^2) In China, in spite of tenured position, such as professor, associate professor, experienced faculty
can also be titled as supervisor of master students, or supervisor of doctor students. A typical
research production model for Chinese university is a professor working with a small group of
postgraduate students.
116 4 Evaluation on Research Efficiency of 211Us: The DEA Approach