The Messenger poll was conducted with 417 registered Georgian voters in the towns of Tbilisi, Kutaisi, Batumi and Zugdidi, and rural areas of Mtskheta-Mtianeti and Kakheti provinces, through a combination of face-to-face and telephone interviews. For telephone interviews, random numbers were dialed by using the appropriate city code as a prefix; when a household was reached, the next birthday rule was used to pick a random respondent from within the household.

Due to an insufficiently formal methodology, a reliable margin of error cannot be calculated. Most of the subsample sizes, such as responses by region, are too small to safely draw inferences from. Nonetheless, broad trends like rural vs. urban support are clearly identifiable.

The questions were asked in the following order:

Are you a registered voter?
Do you think your vote matters?
Did you vote in the previous elections?
If the presidential election was held today, who would you vote for?
[No candidates were named as options.]
There may be a run-off between the top two candidates:
If the run-off includes Mikheil Saakashvili/Levan Gachechiladze against Levan Gachechiladze/Mikheil Saakashvili, who would you vote for?
If the run-off includes Mikheil Saakashvili/Shalva Natelashvili against Shalva Natelashvili/Mikheil Saakashvili, who would you vote for?
If the run-off includes Mikheil Saakashvili/Badri Patarkatsishvili against Badri Patarkatsishvili/Mikheil Saakashvili, who would you vote for?
[For the run-off questions, interviewers were instructed to alternate the order of candidates between respondents.]

Only respondents answering positively to the first two questions were counted.

Click here for an Excel spreadsheet of the aggregate data.