The surveys, which lay the basis for this note, were conducted monthly as part of KONDA Barometer. You can see below the results of our surveys conducted since February 2015, when a referendum for presidential system was first vocalized. The surveys represent the tendencies of electors only within Turkey.
According to the survey, the turnout for the referendum will be around 90 per cent. However, we cannot envisage the impact of a different turnout rate on the results.
The latest survey on the presidential referendum was conducted on April 7-9.(1) The percentage of undecided voters and non-voters is 9 per cent. However, in our multi-dimensional statistical analyses, we have not found any data predicting the political tendencies of undecided voters. Under these circumstances, votes are expected to be as follows: 51.5 percent for “yes” and 48.5 percent for “no.” When this forecast is considered within the survey’s margin of error, a final judgement might be misleading.
KONDA Research and Consultancy Inc.
(1) Within the scope of the survey, 3462 respondents were interviewed face-to-face in their homes in 197 neighborhoods and villages of 124 districts – including central districts – of 30 provinces. It presents the political trends, preferences and profiles of the adult population above the age of 18 in Turkey, as observed on the dates of the field survey. Among the 18 interviews conducted in each neighborhood, quotas on age and gender were enforced. The margin of error of the survey is +/- 1.6 at 95 percent confidence level and +/- 2.2 at 99 percent confidence level.