Entrance and Exit Polls

Suppose that we are trying to gauge voter support for a political candidate. It’s too costly to ask everyone his or her opinion, so we sample the population and find out who they intend to vote for. A political opinion poll of adults is not the best idea, because not all adults can vote. Polling registered voters is better, but many people are registered but yet have not voted in years. A polling methodology that asks the opinion of a statistical sample of likely voters is better yet, but it can be difficult to determine who is likely to vote.

It is hard to figure out likely voters unless we poll people’s opinion as close as possible to election day. To be absolutely certain of capturing a pool of likely voters, we could poll individuals on the way to or from the voting booth. Polling a person right before they cast a ballot is called an entrance poll. Polling them after they have voted is called an exit poll. Since these polls are carried out in the vicinity of the ballot box, they have some advantages over other sorts of political polls.

Similarities and Differences

Entrance and exit polls are very similar; both types of polls are conducted in the vicinity where people vote. Of course there is an obvious difference between these two types of polls. Entrance polls are based upon a future event, and exit polls on a past event. There is always the possibility that a voter will change his or her mind between being asked questions for an entrance poll and then actually voting.

Thus exit polls are considered to be more reliable than entrance polls.

Usefulness of Exit and Entrance Polls

Entrance and exit polls can be used in predicting the outcome of an election that is underway, a perhaps more important use of exit polls is to discern voter intent. Ballots are cast for a number of reasons.

When we see that candidate A won over candidate B, it is clear that more voters preferred A. What is unanswered is why did voters prefer A? Exit polls can help to answer this sort of question.

Another question that exit polls address is who voted. A tally of votes provides no demographic information about who voted for a particular candidate. Exit polling can help provide context on how people of a certain gender, race or religion voted.

Limitations of Entrance and Exit Polls

Although exit polls are informative, there are some limitations. Due to how the polls are conducted there is a greater margin of error for these kinds of polls. The way that the voting population is sampled is a cluster sample. Not every precinct is sampled. Rather certain precincts are selected and then many voters at these precincts are sampled.

Another limitation of exit polls for predictive purposes is that they are typically preformed during the earlier parts of election day. Most have some method for polling early voters, but people who vote later in the day are missed by the poll.

Another limitation is that there is some indication that some groups are more likely to respond to the pollsters questions than other groups.

This can overstate support for a certain candidate. This has been seen in many elections. One of the more famous examples is the 2004 U.S. Presidential Election. Leaked exit polls suggested a win by John Kerry, but when the votes were counted it was found that George W. Bush was the winner of the election.

Restrictions

Since exit polls are conducted while an election is underway, the results from these kinds of polls could influence voters throughout the day. For this reason, many countries have legislation to prevent polling of any kind in the days before an election. Other countries have statutes limiting the release of exit poll data until after all votes have been cast for an election.