Why do we need to go through all this trouble to identify likely voters and create a model of the likely electorate? Surveys are only valuable if they can tell us something about a larger group of people than just the ones we interview. Surveys need to be representative of a group of people we want to know about, in this case, those voters who are likely to cast a ballot in an election.
We know not all voters vote in every election. So, if we are to understand what's going on in an election, we need the opinions of those people who will participate or ultimately cast a ballot. Research tells us that some people who say they are going to vote, don't end up voting. And some people who say they are not going to vote, actually decide to do so in the end.
So we create a model that estimates the probability or likelihood that someone will vote. And in that way, we can get a better sense of the types of people, young versus old, Democrat and Republican, men, women who are going to participate. And we also can find out which candidate they support. This goes a long way to giving us an idea of what's going on in a particular election.