| Article ID | Journal | Published Year | Pages | File Type |
|---|---|---|---|---|
| 311558 | Transportation Research Part A: Policy and Practice | 2012 | 12 Pages |
This paper presents a meta-analysis of variations in seaports’ Mean Technical Efficiency (MTE) scores based on 40 studies published in refereed academic journals. We link the variation in estimated MTE scores to differences in the following factors: the frontier methodology used, which essentially are the Data Envelopment Analysis (DEA) and the Stochastic Frontier Analysis (SFA); regions where seaports are situated; type of data used; number of observations; and the total number of variables used. Furthermore, we compare fixed-effects against a random-effects regression model where the latter assumes that the individual study specific characteristics matter while the former assumes that there is one general tendency across all studies. We present several findings based on the data: (1) the random-effects model outperforms the fixed effects model in explaining the variations in MTEs, (2) recently published studies have lower MTE scores as compared with earlier published studies, (3) studies that used nonparametric DEA models depict higher MTE scores as compared with those that used SFA models, (4) panel data studies have lower TE scores as compared with cross-sectional data, and (5) studies using European seaport data produce lower MTE scores when compared with the rest of the world. Finally, our results contradict some previous meta-analysis studies of TE scores. We encourage the use of random-effects models in meta-analysis studies because they account for individual study specific effects.
► We conduct a meta-analysis of variation in seaports’ Mean Technical Efficiency (MTE) scores. ► Studies that used DEA give higher MTE scores as compared to those that used SFA models. ► Panel data studies have lower MTE scores as compared to those that used cross-sectional data. ► Studies using European data produce lower MTE scores as compared with the rest of the world. ► The use of random-effects model is encouraged since it accounts for study specific effects.
