Stratified Models of Education Production Using Modified DEA and Regression Analysis

Lovell, C. A. Knox; Walters, Lawrence C.; Wood, Lisa

doi:10.1007/978-94-011-0637-5_17

Cited by 87 publications

(79 citation statements)

References 0 publications

Supporting

Mentioning

Contrasting

Unclassified

Order By: Relevance

“…Yet as suggested by Lovell et al (1994), skewness of DEA scores is rarely reported, "and even more rarely put to any good use". We report box plots since they disclose a more scrupulous illustration of the distributions, consequently encoding further information on the features possibly hidden by data, such as multi-modality or the existence of outliers, which may have quite an impact on mean efficiency-along with the aforementioned mass of scores at unity.…”

Section: Resultsmentioning

confidence: 99%

“…Accordingly, a second-stage analysis is called for, as efficiency may be affected not only by inadequate management but also by exogenous factors beyond the control of each local government. As Lovell et al (1994) pointed out, such ideas had been already considered by Lewin and Minton (1986) in their research agenda, calling for studies aimed at illustrating the "feasibility of using DEA (perhaps in combination with other analytical methods) as a mathematic for relating effectiveness outcomes to features of organization design".…”

Section: Efficiency Explanatory Variablesmentioning

confidence: 99%

See 1 more Smart Citation

On the determinants of local government performance: A two-stage nonparametric approach

Balaguer-Coll

Prior

Tortosa‐Ausina

2007

European Economic Review

220

179

View full text Add to dashboard Cite

This article analyzes the efficiency of Comunitat Valenciana (Spain) local governments and their main explanatory variables. The analysis is performed in two stages. Firstly, efficiency is measured via (nonparametric) activity analysis techniques. The measurement techniques also enable the sources of inefficiency to be identified, i.e., whether inefficiencies are primarily overall cost, technical, or allocative. The second stage identifies some critical determinants of efficiency, focusing both on political and fiscal policy variables. In contrast to previous two-stage research studies, our approach performs the latter attempt via nonparametric smoothing techniques, rather that econometric methods -such as OLS or Tobit related. Results show that inefficiencies are largely attributable to allocative factors, resulting in a notable gap between cost and technical efficiency. Inefficiencies are also larger for small municipalities. However, they are not exclusively attributable to poor management, as second-stage analysis reveals both fiscal and policy variables, either within or outside local governments control, to be explicably related to municipalities' performance. KEYWORDS:efficiency, kernel smoothing, local government, nonparametric regression.JEL classification: D24, D60, H71, H72. RESUMENEste trabajo analiza la eficiencia de las corporaciones locales de la Comunidad Valenciana y sus factores determinantes. El análisis se lleva a cabo en dos etapas. En primer lugar, la eficiencia es medida a través de técnicas (no paramétricas) de análisis de la actividad. El empleo de este tipo de técnicas permite también identificar las fuentes de ineficiencia, esto es, si esta es básicamente en costes, técnica o de carácter asignativo. La segunda etapa identifica algunos determinantes clave de la eficiencia, centrándose tanto en variables políticas, así como de política fiscal. Frente a estudios previos en dos etapas, nuestro enfoque lleva a cabo este último objetivo a través de técnicas no paramétricas de suavizado, en lugar de métodos econométricos (tales como MCO o modelos Tobit). Los resultados indican que las ineficiencias son atribuibles en gran parte a factores asignativos, dando lugar a un margen considerable entre eficiencia en costes y eficiencia técnica. La ineficiencia también es mayor para las corporaciones locales más pequeñas. Sin embargo, no es atribuible exclusivamente a una gestión deficiente, pues el análisis de segunda etapa muestra que tanto las variables políticas como las de política fiscal, independientemente de si están o no bajo el control de las corporaciones locales, se hayan relacionadas con su nivel de eficiencia.

show abstract

Section: Resultsmentioning

confidence: 99%

Section: Efficiency Explanatory Variablesmentioning

confidence: 99%

On the determinants of local government performance: A two-stage nonparametric approach

Balaguer-Coll

Prior

Tortosa‐Ausina

2007

European Economic Review

220

179

View full text Add to dashboard Cite

show abstract

“…As illustrated by the application work by Lovell et al (1994): "The primary benefit of this approach is the ability to make finer distinctions between efficient DMUs and to produce a logarithmic MDEA distribution of relative performance scores that are approximately normally distributed" (Charnes et al 1994). Thus, the super-efficiency DEA model has the extraordinary potential to overcome the analytic difficulties arising in the post-DEA analysis posed by the spiked distribution of the DEA efficiency scores.…”

Section: Discussionmentioning

confidence: 99%

“…In practice, further differentiation among efficient DMUs is also desirable and even necessary in many cases. One classical example of the application of the superefficiency DEA method is the work by Lovell et al (1994). Other scholarly efforts attributed to this topic include the work by Charnes et al (1992), Fare et al (1994), Wilson (1995), Rousseau and Semple (1995), Charnes et al (1996), Thrall (1996), Zhu (1996), and Seiford and Zhu (1998 a, b).…”

Section: Introductionmentioning

confidence: 99%

Note: Ranking DMUs with Infeasible Super-Efficiency DEA Models

2002

View full text Add to dashboard Cite

I t has been suggested in the data envelopment analysis (DEA) literature that it is impossible to obtain a full ranking of decision-making units (DMUs) when infeasible subproblems arise in the so-called super-efficiency DEA models under different returns to scale (RTS) assumptions other than constant returns to scale (CRS) and consequently the application of the super-efficiency DEA models under different RTS conditions other than CRS should be restricted. The implications of the infeasibility in super-efficiency DEA models with respect to the efficiency ranking of the DMUs is explored. Based on the analysis, we show that it is still possible to obtain the full ranking of the entire observation set when infeasibilities arise in super-efficiency DEA models.

show abstract

“…The former has the advantage that it is available for all levels of schooling and all years, but its disadvantage is that it is more a measure of input than output. There is, however, a high degree of correlation between student and graduate numbers in the context of schools (since the drop-out rate is relatively low), and hence student numbers (or variables relating to enrolment) have been used to represent teaching outputs in a number of empirical studies (Lovell et al 1994;Ouillette and Vierstraete 2006;Jones et al 2008).…”

Section: Data and Modelsmentioning

confidence: 99%

The efficiency of public schools: the case of Kuwait

Burney

Johnes

Alenezi

et al. 2013

Education Economics

View full text Add to dashboard Cite

This paper investigates the technical, allocative and economic efficiency of public schools in Kuwait over four levels of schooling (kindergartens, primary, intermediate and secondary) and two periods (1999/00 and 2004/05) using data envelopment analysis (DEA). Mean pure technical efficiency varies between 0.695 and 0.852 across all levels of education; the majority of schools at kindergarten, primary and intermediate levels are operating at a point where returns to scale are increasing; and there are considerable cost efficiencies to be gained. In a second stage analysis of the determinants of efficiency, teacher salary and the proportion of teaching staff who are Kuwaiti are highly significant in explaining school efficiency at all levels. The former has a positive effect and the latter a negative effect. Allgirls schools have significantly higher efficiency than all-boys schools. There is limited evidence that geographical location affects efficiency, and this may be a consequence of differences between regions in terms of affluence or density of population.

show abstract

Stratified Models of Education Production Using Modified DEA and Regression Analysis

Cited by 87 publications

References 0 publications

On the determinants of local government performance: A two-stage nonparametric approach

On the determinants of local government performance: A two-stage nonparametric approach

Note: Ranking DMUs with Infeasible Super-Efficiency DEA Models

The efficiency of public schools: the case of Kuwait

Contact Info

Product

Resources

About