Bayesian Estimation of Limited Dependent Variable Spatial Autoregressive Models

A Gibbs sampling (Markov chain Monte Carlo) method for estimating spatial autoregressive limited dependent variable models is presented. The method can accommodate data sets containing spatial outliers and general forms of non‐constant variance. It is argued that there are several advantages to the method proposed here relative to that proposed and illustrated in McMillen (1992) for spatial probit models.     

Fashioned Forest Pasts,Occluded Histories? International Environmental Analysis in West African Locales

This article considers how environmental problematics are produced and interpreted, using case material from West Africa’s humid forest zone. Examing the experiences of several countries over the long term, it is possible to identify a deforestation discourse produced through national and international institutions. This represents forest and social history in particular ways that structure forest conservation but which obscure the experience and knowledge of resource users. Using fine‐grained ethnography to explore how such discourse is experienced and interpreted in a particular locale, the article uncovers problems with ‘discourse’ perspectives which produce analytical dichotomies which confront state and villager, and scientific and ‘local’ knowledges. The authors explore the day‐to‐day encounters between villagers and administrators, and the social and historical experiences which condition these. Instances where the deforestation discourse becomes juxtaposed with villagers’ alternative ideas about landscape history prove relatively few and insignificant, while the powerful material effects of the discourse tend to be interpreted locally within other frames. These findings present departures from the ways relations between citizen sciences and expert institutions have been conceived in recent work on the sociology of science and public policy.     

The Impact of Agglomeration Economies on Estimated Demand Thresholds: An Extension of Wensley and Stabler

Central place theory predicts that geographic markets located in rural areas have lower demand thresholds, and, therefore, a higher frequency of business establishments relative to areas that are more proximate to urban centers, other things equal. Wensley and Stabler (1998) confirm this prediction using data on the location and frequency of business activities in rural Saskatchewan. We demonstrate that this relationship may not always hold true depending on the existence and magnitude of agglomeration economies. If average cost differences associated with being located in an urbanized area are sufficiently large, then the relationship between urban proximity and number of establishments may be reversed. We provide evidence of this reversal using 1996 cross-sectional data on hospital services in Texas.     

Recontextualising Positionality: Geographical Research and Academic Fields of Power

The first article of the first issue of Antipode began with a declaration concerning the values ascribed to publication and the nature of academic work in geography. With that thirty-year-old declaration in mind, this essay reflects on aspects of these values as registered in British academic geography. It does so with particular reference to themes that have been raised in a number of recent debates, namely the questions of commodification/marketisation of and reflexivity in geographical research and scholarship.