|
Most of the empirical analyses of the performance of educational and research
institutes are based on estimation of cost functions focusing on economies of size and scope, or
on analysis of efficiency via Data Envelopment Analysis (DEA) or frontier functions.
Such studies have mainly focused on the UK or the US. The key studies in this area are, Cohn et al. (1989), who have studied the institutes of higher education, De Groot et al. (1991) who have focused on American research universities, Nelson and Hevert (1992), Dunbar
and Lewis (1995), and King (1997). Research on universities in the UK includes Glass et al. (1995a and 1995b) and Athanassopoulos and Shale (1997), while Hashimoto and
Cohn (1997) have investigated the Japanese universities, McMillian and Debasish (1997)
have investigated the Canadian universities, and Abbott and Doucouliagosa (2003) have
studied the Australian universities. The Faculty of Management and Economics is one of the
several faculties of Shahid Bahonar University of Kerman in Iran. It consists of three
departments, viz., accounting, business administration and economics. This paper uses the DEA
method to assess the relative efficiency of these departments for ranking them with regard
to research performance.
DEA has been demonstrated to be a suitable method to measure the relative efficiency of
a set of Decision-Making Units (DMUs) that utilize the same inputs to produce the
same outputs. Basically, DEA is a mathematical programming technique that produces a
single aggregate measure for each DMU in terms of its utilization of inputs to produce
desired outputs (Charnes et al., 1994). The relative efficiency of each DMU is expressed as the
ratio of aggregated outputs to aggregated inputs. Conceptually, each DMU is allowed to
select the weights which are most favorable in calculating its relative efficiency as long as
the same weights will not result in efficiency scores exceeding one for all DMUs. The
efficiency calculations focus on the revealed best-practice production frontier. This enables
the inefficient DMUs to calculate the amount of inputs to be reduced and the amount of
outputs to be increased in order to become efficient. This technique has been widely applied
in measuring the relative efficiency of different industries, especially for
not-for-profit organizations (Seiford, 1996). In order to obtain the highest efficiency score, it is
possible that the DMU will assign a weight of zero to unfavorable factors. This implies that
the associated factors are eliminated from evaluation (Kao and Hung, 2008).
Conventional models in DEA evaluate the efficiency of an observation relative to
a reference set comprising all sample observations, including the observation
under consideration (Banker and Chang, 2006). In contrast, the super-efficiency model in
DEA excludes each observation from its own reference set, so that it is possible to obtain
efficiency scores that exceed one (Banker and
Chang, 2006). Banker and Gifford (1988) suggested
the use of the super-efficiency model to screen out observations with gross data errors,
and obtain more reliable efficiency estimates after removing those identified outliers. Banker et al. (1989) applied this method for outlier identification to analyze cost variances for 117
hospitals. Andersen and Petersen (1993) employed the same
Banker-Gifford model and prescribed the use of the super-efficiency score for ranking efficient units in DEA (Banker
and Chang, 2006). |
