Finally, we can infer from our evidence that differences in state responses to the West from the beginning of the transition do not appear to be grounded in differences in mass attitudes at least at the point in the transition captured in the surveys discussed here. As with many issues to do with the post-communist transformation, elite responses 29 —both in the East and the West—may have a substantial impact on policies adopted, whether this might be NATO or EU expansion. Over time, these policy choices may also have an effect on mass attitudes (though data available to the authors from subsequent surveys in Russia in 1995, 1996, and 1998 indicate that the shift in public opinion towards the West has not been enormous). But, at least for the period analysed here, there is little reason to believe that mass constraints to building support for the West in Eastern Europe have been powerful ones.
Sampling frames and stratification procedures varied between countries: in some countries census information was considered more reliable; in others electoral records were preferred; in other random route procedures were adopted with a Kish grid being used for final respondent selection. Each of these strategies was considered to be the most effective approach within the countries in which it was adopted.
Response rates were generally high despite a number of non-contacts in some countries. As far as can be told, given the fallibilities of official data, non-response biases are predictably like those in the West. Compared to Census data non-respondents tend to be older and to have lower levels of education. Non-response resulted mainly from non-contacts and refusals. Table 8.A1 summarizes the main characteristics of the surveys.