The New Russian Nationalism Imperialism, Ethnicity and Authoritarianism
backing the ussr 2.0
- The Welch–Satterthwaite method was used to compare means,
assuming the sub- samples did not have equal variance. The results
show that violating the equal variance assumption had no effect on
the estimation of which between- the- sample differences of means
were statistically significant.
- The rotation was 100 per cent randomised. Interviewers were not
set quotas in terms of socio- demographic categories. The difference
in experimental (N = 41) versus control (N = 121) group size is a
by- product of this random selection, given the small proportion of
ethnic non- Russians in the total sample. Comprising just 4.5 per
cent of the total sub- sample in the three polls (N = 1,800) it is con-
sistent with the sampling error margin for each poll.
- Excluding the ‘don’t knows’ and refusals.
- Excluding the ‘don’t knows’ and refusals to answer the question,
with valid N = 166 (non- Russian sub- sample) and N = 2,060 (ethnic
Russian sub- sample).
- Differences between Slavs versus non- Slavs in the ethnic non-
Russian sample were not statistically significant.
- Based on a one- sample t- test.
- One other split- sample experiment with a Putin cue was conducted
in the Krasnodar survey, on support for the political slogan ‘Stop
feeding the Caucasus!’, popular in Russia’s predominantly ethnic
Russian regions at the time of the survey. About half of the ran-
domly selected respondents from the sample were given a prompt
that Putin had condemned the slogan. The prompt had practically
no effect, whether among ethnic Russians or non- Russians. The full
results are not reported here, because the ethnic non- Russian sub-
sample was small – with valid N = 20 for the no- cue question and N
= 17 for the Putin- cue question.
- State Statistics Committee of Ukraine, government website, avail-
able at http://2001.ukrcensus.gov.ua/eng/results/general/language/
Crimea (last accessed 10 August 2014); Federal’naia sluzhba gosu-
darstvennoi statistiki (n.d).
