From December 21, 2007-January 29, 2008, IRI fielded 11 polls during the Serbian presidential election cycle. 

The poll field work was conducted by Strategic Marketing Research based in Belgrade, Serbia.  Oversight and analysis was provided by David Williams of Williams & Associates of Salem, Massachusetts and the International Republican Institute.

The first poll had a sample of 1,000 respondents; the 10 tracking polls which followed had a sample size of minimum 500 respondents each.  The total sample size was 6,398 completed interviews
 
The population under study is representative for Serbian population older than 18 years by gender, age, education, settlement size and region:

The research was conducted by face-to-face interviews in respondents’ homes.

Maximum sampling error on whole sample: +/-1.23 percent.

The poll surveyed a nationwide random, three-stage, stratified representative sample. The primary sampling units were polling station territories chosen with probability proportional to size, sampled by cumulative, Lachirie method; secondary sampling units were households chosen with simple random sampling without replacement, sampled with random choice of the starting point and equal steps of choice; tertiary sampling units were respondents sampled by Kish scheme.  First level strata were eight regions and second level strata were two types of settlement, urban and rural.
 
Respondents were disqualified if they were younger than 18 years of age and if they or any member of their close family was employed at a newspaper, television station, radio station or political party.

Survey response rate was 61 percent

Demographic Breakdown