Read more about successful election management in associations

Exit polls are surveys which ask people how they voted directly after leaving the polling station. Since maintaining the secrecy of the ballot is a pillar of democracy, conducting exit polls is the only way to gain insights into how and why people voted in a particular way on election day. 

Who conducts exit polls and why?

Exit polls are typically conducted by private polling companies on behalf of major newspapers and television networks. They generally collect three types of data:

  • How people voted
  • Attitudes held by voters which were important in making their decision
  • Demographic data such as age, educational attainment, race and gender

This data is then analysed in order to infer which demographic groups preferred certain candidates and why, as well as providing an early indication of how the entire election might play out.

Exit polling can also be used to guard against election fraud. This is done by checking the data against the final result of the election and noting any major discrepancies.

History of Exit Polls

Warren Mitofsky is credited with conducting the first large-scale exit poll for CBS during the 1967 Kentucky governor’s election. The first US nationwide exit polling operation was undertaken in 1972 and has been standard practice ever since.

In the 1990s, competing news organisations created a single polling service to reduce costs associated with conducting multiple surveys. Having gone through various incarnations, the present day service is known as the National Election Pool, consisting of ABC News, The Associated Press, CBSNews, CNN, Fox News and NBC News.

Data for the 2016 US presidential election was collected exclusively by polling company Edison Research. Surveys were completed by almost 25,000 voters at 350 polling stations across the country, including over 4,000 telephone interviews with absentee voters.


The history of exit polling is, however, marred by numerous controversies. The main cause for concern has come when news organisations have predicted election results before actual polling has closed. This may influence the election result.

In 1980, for example, NBC declared Ronald Reagan had won the US presidential election despite polls remaining open on the west coast. Concerns were raised as to whether this resulted in a lower voter turnout on the west coast. Nowadays, major news networks don’t publish sensitive exit polling data until actual polling has closed in each state.

Germany goes a step further, having made it a criminal offence to publish data before polling stations have closed.

Early publication of exit polls may adversely affect voter turnout, but one surefire way to increase voter turnout is by conducting elections online.

Learn more about boosting voter turnout through online elections here.