Kenneth Macdonald: Annual Report 2001-2002

Kenneth Macdonald (Faculty Fellow) spent some of the year addressing a social policy issue. Government had announced: '150 social services departments in England would be graded with star ratings . . . similar to the system being introduced to distinguish between good, bad and indifferent hospitals. In a decisive break with the past, social services will be ranked on their performance, with more freedom and rewards for the best and more help for the worst.' But detailed multivariate analysis of the proposed indicators indicated that their behaviour was not compatible with their interpretation as measures of competence. Had they located, as claimed, 'top performing councils', then one might reasonably expect that, within one performance area (such as children's services) there would be association across indicators. The data do not support this. (The obvious alternative hypothesis - the indicators measure problem distribution and delivery difficulty, not quality of service - is, at this level of aggregation, not supported. Adding demographic data, and the Oxford-generated DTLGR Indices of Deprivation for districts in England revealed no intelligible patterns. Which suggests that government has, besides misinterpreting its indicators, chosen, on this, the wrong level of aggregation for resource allocation.) In less 'political' vein, he has continued work (with Dr Frazer, New College) on the determinants of political knowledge. And he thinks he has made some progress on his twin obsessions of 'particular' and 'adventitious' goods and their interconnections. It will be interesting to see whether that perception survives the writing-up.

Publication

(with M.S.Grieco) 'Transforming Contours: Information Technology in Cities and Regions in Transition', European Spatial Research and Policy, 9, 2002.