research inputs, or enlarging research scale. However, there is also around 40–50%
of total 211Us operating in the area of IRS, who might improve their efficiency
through investing more and enlarging research scale.
4.3.2.2 Comparing Research Efficiency Between 985Us
and Non-985Us
In 2006–2010, the DEA-efficient proportion of 985Us is in the trend of steadily
growth, while that of non-985Usfirst increases and then decreases. According to
the data offive-year averages, the DEA-efficient proportion of 985Us is 7.4%
higher than that of non-985Us. However, to either 985Us or non-985Us, the
DEA-efficient proportion is at a relatively low level, and most universities need to
improve their research efficiency.
In 2006–2010, the averages of TE scores for 985Us and non-985Us are around
0.40–0.57 and 0.29–0.37 respectively, so the performance of 985Us is better than
that of non-985Us. According to the results of standard errors of TE scores, there is
a wider variations in 985Us than in non-985Us, which means that the
between-university TE gap is much greater among 985Us.
4.3.2.3 Comparing Research Efficiency Between 211Us by Location
During 2006–2010, the proportion of DEA-efficient 211Us in eastern, central and
western regions are all at a relatively low level, no greater than 9%. Compared with
production frontier of the full sample, the DEA-efficient proportion of central
211Us is higher than that of eastern or western 211Us, and that of western 211Us is
slightly lower than that of eastern 211Us. However, inside the eastern 211Us, it’s
worth to notice that there is no DEA-efficient university in Hu-Su-Zhe region.
During 2006–2010, the averages of TE scores for eastern, central, and western
211Us are around 0.35–0.45, 0.32–0.43, and 0.26–0.38 respectively, and the
technical efficiency status of western 211Us is of the widest variation. Overall
speaking, the TE average scores of 211Us located in most regionsfirst increase,
then decrease, and then increase again, but their values are kept at a relatively low
level. The technical efficiencies of eastern, central and western 211us are in the
pattern of step-wise decaying.
Similar with the results of NEAM disciplines, the standard deviations of TE scores
don’t differ too much between eastern, central and western 211Us. However, the
between-university TE gaps are slightly greater among central and western 211Us.
4.3.2.4 Analysis of Projections on DEA-Inefficient 211Us
In 2006–2010, the redundant proportions of R&D Staff with Senior Title and R&D
Grants of DEA-inefficient universities are pretty high, and theirfive-year averages
162 4 Evaluation on Research Efficiency of 211Us: The DEA Approach