the new russian nationalismenough variation on key variables to enable theory- relevant infer-
ences to be drawn.
This analysis uses the polls that Russia’s Romir agency, respected
especially for its business and marketing surveys, carried out from
8 to 27 May 2013. The polls were part of the ‘Nation- building,
nationalism and the new “other” in today’s Russia (NEORUSS)’
project. Four surveys were conducted based on representative
multi- stage probability samples of adult residents of the Russian
Federation (N = 1,000 respondents) and, separately, the cities of
Moscow (N = 600), Krasnodar (N = 600) and Vladivostok (N =
601). In the Russian national sample, respondents were selected
from fifty- eight out of eighty- three provinces in key population
clusters of all eight federal districts of the Russian Federation
(including the North Caucasus district).^8 The sampling error
margin was approximately +/− 3 per cent. City polls followed
identical sampling procedures, resulting in approximately the
same non- response rate and the margin of sampling error of
about +/− 4 per cent. In each poll, all responses were obtained
from face- to- face interviews.^9 The data from these four polls were
merged and parsed into two sub- samples.
The grouping variable for the sub- samples was the respondents’
ethnic self- identification. The interviewers gave respondents the
opportunity to identify with any number of ethnic groups. The
overwhelming majority picked only one. The largest number of
coded self- identifications was three, in the order given by respond-
ents. All respondents answered this question. The first sub- sample
consists only of respondents in the four polls whose first- listed
ethnic self- identification was non- Russian (N = 180): 24 per cent
among these listed Tatar, 22 per cent Ukrainian, 10 per cent
Armenian and 17 per cent belonging to various ethnic groups
from the Caucasus or Central Asia.^10 The second sub- sample (N
= 2,219) consisted solely of respondents who identified them-
selves exclusively as ethnic Russians – minus a random sub- set of
respondents from Krasnodar. The under- sampling in Krasnodar
corrected for the only significant discrepancy in the regional dis-
tribution of respondents between the ethnic Russian and non-
Russian sub- samples – namely, that the proportion of ethnic
Russians interviewed in Krasnodar relative to other locations was