Caucasus Barometer 2012, Merged

Data Archive > International Surveys and Data > Multiple Nation Surveys > Summary

Zurabishvili, T. (2020, December 3). Caucasus Barometer 2012, Merged.
The Caucasus Research Resource Centers (CRRC) conducts this annual survey in the South Caucasus (i.e., Armenia, Azerbaijan and Georgia) in order to gauge the social, political and economic issues in the former Soviet Union region. This is the only survey in the region providing reliable comparative data about the opinions, household composition, and economic behavior of the population of the three countries. In addition to the questions asked in previous rounds, CRRC has included new questions assessing social capital and religious views - the latter were asked in collaboration with ARDA. The same survey and methodological approach is applied to all three countries. This merged data file contains the respondents from all three South Caucasus countries and their survey responses. The data file contains 6,715 cases total (2,382 in Armenia, 1,829 in Azerbaijan, and 2,502 in Georgia). Variables were dropped that were not asked across all three countries. So, although Armenian respondents were asked additional questions pertaining to the Armenian Genocide and Georgian respondents were asked additional questions about opinions toward Joseph Stalin, those survey items can only be accessed by downloading the individual country data files (see ARDA's Data Archive for those files). Moreover string/character (non-numeric) variables are not available in the merged data file.
Data File
Cases: 6,715
Variables: 403
Weight Variable: HHWT, INDWT
After the cleaning process was finished, sampling weights were calculated. Sampling weights account for the fact that different members of the population have different probabilities of being selected for interview and thus represent different numbers of people in the overall population, and are necessary when estimating the proportion of the population that would choose a particular response, if interviewed. Sampling weights are then adjusted for non-response. Finally, the respondents were binned into gender and age categories (ages 18-34, 35-54, and 55+), and weights were adjusted so that the weighted age and gender ratios of the sample matched that of the population based on the national census data.
Data Collection
Date Collected: October 26-November 29, 2012
Funded By
Caucasus Research Resource Center (CRRC)
Collection Procedures
This survey uses face-to-face interviews.
Sampling Procedures
This survey uses multi-stage cluster sampling, which included the following steps:

1. The territories of each of the countries (Armenia, Azerbaijan, and Georgia) were divided into four geographical quadrants exclusive of the national capital (i.e., total of five macro-strata). Within each geographic quadrant, voting precincts were classified as being either urban or rural, resulting in nine strata; Capital, North-East urban, North-East rural, North-West urban, North-West rural, South-East urban, South-East rural, South-West urban and South-West rural.

2. Electoral precincts (primary sampling units) were sampled within each stratum with selection probability proportional to the number of registered voters.

3. Random route sampling was used to select households.

4. The Kish table was used to sample respondents (tertiary sampling units) within the sampled households.

These are the response rates across all three countries: 72% in Armenia, 78% in Azerbaijan, and 75% in Georgia.

Results are representative for adult (18+) population of Armenia, Azerbaijan and Georgia, except territories affected by military conflict.

Country-level confidence intervals: +/-1.4 for Armenia; +/-2.1 for Azerbaijan; and +/-1.9 for Georgia.
Principal Investigators
Tinatin Zurabishvili, Research Director, CRRC
Bookmark and Share