It is widely held in the social sciences that behaviour patterns can be predicted from knowledge of peoples' attitudes. Electoral studies in Great Britain largely accept this, but there has been little formal testing of the general hypothesis. In building on work reported in a previous paper, we discuss formal tests of the predictive link between attitudes and voting behaviour, and use voting intentions, strength of party identification, and changes in each as the dependent variables in a series of discriminant analyses. These are reasonably successful, but not overwhelmingly so.
|Number of pages||16|
|Journal||Environment and Planning A: Economy and Space|
|Publication status||Published - Jul 1999|