Opinion polls and electoral research


Opinion polls and electoral studies: how it works


How political researches are conducted and why are they necessary – Vladymyr Kosenko explains.

Büro.MT has been conducting opinion polls and electoral studies for the development of political campaigns for two decades already. We decided to share experience and asked our researcher and analyst to tell a little about political research projects.

How to understand what opinion polls are true and which of them are just fake data? The nearest upcoming election will happen on March, 31, 2019 – that day the first round of presidential election will take place. Electoral studies on this occasion have already been presented by more than 10 sociological companies. Analyzing them allows us to make two observations:

– No matter how close in time the studies are, the data received by different organizations differs a lot. It doesn’t necessarily mean influencing the results. Some mistakes are defined with the law of large numbers and are initially inbuilt in most of the research projects.

– The structure and quantitative characteristics of the sample can also have a big impact on the numbers received. Sometimes an electoral poll of 1200 respondents has a theoretical bias of about 2.82% (based on the total population of Ukraine in the case of full compliance with demographic quotas), while a poll of 2400 respondents will allow to reduce the size of theoretically acceptable inaccuracy to 2%.



Conducting an opinion poll should start with quality sociological methods, for example with focus group discussions. It can complement them in the result but still is a secondary tool to gain the information.

A poll should be held at the place or respondents’ residence. This allows a real opportunity to create a sample based on demographic characteristics of real electors.

Many political players can be also interested in conducting political research via phone, but two factors should be taken into account here:



All the above means that the real cost of an opinion poll will only get higher as the data received via phone survey will need to be checked through a representative research.

Opinion polls in Ukraine are delivered by many agencies, but we believe you should rely on the companies that have experience in conducting electoral studies, data of which was supported with real results of previous elections.

Let’s reveal several important aspects of quantitative electoral studies delivered through express information sampling:

– Opinion polls should only be conducted at the place of respondents’ residence.

– In order to ensure reliability of the information received no less than 310 people per each 100 000 of population per electoral district should be surveyed.

-During an active electoral campaign such sociological studies should be held no less than once a month.

Büro.MT is always ready to help in conducting opinion polls in Kyiv and whole Ukraine, developing a creative political campaign. We organize electoral studies (both quantitative and qualitative), develop branding of a political party and messages of a campaign (both for a separate candidate at a majoritarian district and for a regional and national political power). Even on the day of the election you are not alone – you can rely on us in exit-polls.

There is always a solution!