Datafolha is the research institute of Grupo Folha and was created in 1983 as a department of Grupo Folha da Manhã, but became independent. Understand how the company conducts electoral research
When carrying out a voting intention survey, Datafolha only listens to a group of people (sample) whose characteristics (age, sex, education, regional distribution, etc.) reflect the set of Brazilian voters (universe)
Datafolha’s national samples have between 2,000 and 2,500 interviews, but there is no minimum or ideal size for an electoral sample. The most important thing is its representativeness, that is, how the interviewees are selected
In national or state surveys, the municipalities that will be part of the survey are first drawn; then, the neighborhoods and points where the interviews will be carried out
To select interviewees, Datafolha uses proportional quotas of sex and age according to data obtained from IBGE and the Superior Electoral Court
The objective of quotas is to guarantee the representation of the universe that is intended to be studied. In the case of electoral surveys, the sample must reflect the distribution of the electorate according to data updated by the TSE
Surveys have a margin of error, which exists because the sample is always smaller than the universe. National Datafolha surveys, in general, have a margin of error of plus or minus two percentage points
The margin of error varies according to the percentage obtained by each candidate. Datafolha discloses the “maximum” margin of error, that is, the margin of error for percentages close to 50%
Datafolha does not carry out custom research for politicians or parties. All surveys are carried out for public dissemination and use by major media outlets
Your subscription helps Folha continue providing independent, quality journalism
See the main news of the day in Brazil and around the world
