to the different combination of input-output indicators. In an earlier study, Beasley
( 1990 ) applied DEA method to evaluate the teaching and research efficiency in 52
departments (physics and chemistry) in UK. More recently, a study from Abramo
et al. ( 2008 ) applied a combined approach of DEA and bibliometrics, to measure
the research efficiency of different disciplines in Italian universities, and found
significant research efficiency differences among varied disciplines. Johnes and Li
( 2008 ) used DEA approach to evaluate the research efficiency of 109 universities in
2003 and 2004 separately, and found that the average research efficiency score of
Chinese universities was above 0.9; the research efficiency ranking in 2003 and
2004 are highly correlated with statistical significance by DEA results; the average
research efficiency score of comprehensive universities was higher than that of
specialized universities; the average research efficiency score of universities in
eastern China was higher than that in western China.
Compared with those static evaluations on research efficiency mentioned above,
the dynamic evaluations on research efficiency are relatively minority. It’s well
known that Index can be used to measure TFP changes over time. Johnes ( 2008 )
used distance-function and DEA method to calculate the Malmquist Index of 112
British universities from 1996/97 to 2004/05. The results showed that the efficiency
score of British universities grew by 1% every year. Furthermore, this study found
that the growth was mainly attributed to the increase of TC (or technical change),
about 6% every year, while the TE (or technical efficiency) decreased by 5% every
year. The rapid reform in higher education sector had positive effect on production
techniques in universities, but the cost was the decline of technical efficiency.
2.4.2 Empirical Studies on Chinese Universities.............
Accompanying the entrance and fast dissemination of efficiency evaluation methods
like DEA to China, many Chinese scholars began to use these methods to evaluate
the research efficiency and productivity in Chinese higher education. Some
influential studies are summarized in Table2.2.
Lu et al. ( 2006 ) applied DEA method to compare the research efficiency of
universities administered by China Ministry of Education (MOE) located in dif-
ferent regions. They found that university research efficiency score decreases across
China from eastern, central, to western, and their research efficiency and scale
efficiency isfluctuating in each region with their own special trend. Enlarging
research scale was the primary channel for Chinese universities to improve their
research efficiency. Tian and Miao ( 2006 ) applied DEA method to calculate
research efficiency scores of 510 Chinese universities. Theirfinding is similar with
that of Lu’s, namely, the average score of technical efficiency decreases across
China from eastern, central, to western. Some studies (Xu 2009 ; Li and Ren 2009 )
also did the analysis at provincial level, and used DEA method to examine the
research efficiency of universities between different provinces. They found that
research efficiency of universities had certain positive association with the
2.4 Empirical Studies in Evaluating University... 19