The identification of individuals by observational gait analysis using closed circuit television footage

This study investigated the ability of individuals with experience in gait analysis to identify people by observing features of gait recorded by closed circuit television cameras (CCTV). Seven experienced analysts each viewed five samples of footage. Each sample showed a “target walker” and five “suspect walkers.” The task of the experienced analysts was to determine which, if any, of the “suspect walkers” was the “target walker.” All of the participant “walkers” wore identical loose fitting clothing to mask anatomical and body contour features, and balaclavas to obscure facial features. The overall results showed that the experienced analysts made a correct decision in 124 of 175 cases (71%), significantly better than would have been expected to have occurred by chance (p < 0.05). A significant variation in correct decisions (p < 0.05) was shown to occur between the various angles from which the footage was recorded, footage recorded in the saggital plane showing the highest number of correct decisions. Significantly more correct decisions (p < 0.05) were also shown to occur when the footage of the “target walker” and that of the “suspect walker” were taken from the same angle. The results suggest that individuals with experience in gait analysis perform well in the comparative identification of suspects from CCTV footage, and therefore do have a role to play as expert witnesses in this field.     

THE TYPOLOGY OF FISCAL DECENTRALISATION SYSTEM: A CLUSTER ANALYSIS APPROACH

The article suggests a novel theoretical framework for empirical measurement of fiscal decentralisation by taking the viewpoint that fiscal decentralisation should be regarded as a system. Then, we review in detail several institutional arrangements regarding fiscally decentralised systems including the following: federal versus unitary state, numbers of tiers of governments, taxing power, borrowing power and independence of local officials. After identifying these institutional arrangements, we can group different countries with similar institutionally fiscal decentralisation systems together in broad categories by using cluster analysis. This analysis reveals a typology of fiscally decentralised systems (FDS): high expenditure/revenue assignment FDS, low expenditure/revenue assignment FDS, revenue transfer FDS, most complete FDS, politically centralised FDS and unitary state FDS. We then employ regression analysis to examine the performance for each type of fiscal decentralisation system. Our empirical findings suggest that most complete FDS is positively associated with selective economic performance as well as various public governance indicators. Copyright © 2011 John Wiley & Sons, Ltd.