The different theoretical approaches used in action space research are discussed; then a causal model, based on hypotheses drawn from the relevant literature, is presented. This model is tested in an empirical study on the action spaces of 3000 residents in the region of Hamburg. The results of two different path analytical test strategies of this model are reported, which show the deficiencies of prior, mainly bivariate, analyses. Finally, a cluster analysis of the respondents' activities demonstrates that the ‘sociodemographic groups’ which are traditionally used are not homogeneous in their behaviour.
Stored only when both papers are in the corpus — 22 citations exist upstream.