Registered Voters and Likely Voters

There is an important distinction between polls of registered voters and likely voters.

I have to make this point to people in every election cycle, but it seems particularly relevant this time.

It's easy to be a registered voter. Registration drives are so numerous that, in many places, you can register to vote at your local shopping center or campus. (I recall overhearing a snippet of a conversation between some of my students about voting. One student expressed what appeared to be genuine surprise that she was not automatically registered to vote on her 18th birthday.)

It wasn't that simple when I was a child. Most people know, I guess, of the many barriers that were erected by state and local folks (notoriously in the South but elsewhere, too) to prevent minorities from registering, but it wasn't quite as simple then as it is now.

Even those who did not have to pass literacy tests or jump through any of those other hoops had to go to a specific office during office hours to register. That was what I had to do when I registered to vote for the first time.

Anyway, it took a special effort to get registered to vote in those days. You couldn't just walk up to a card table while you were browsing video games or T–shirts. As a result, I presume that being a registered voter carried a little more weight with pollsters then than it does now.

Gradually, pollsters have been inclined to differentiate between garden–variety registered voters and voters who are truly likely to vote — and the definition of a likely voter may well change as the practice of polling evolves.

But, right now, most polling organizations define likely voters as those who have an established history of participating in elections. Obviously, that de–emphasizes the youngest voters, even if participation is minimally defined (say, for instance, voting in at least the last two election cycles).

Frank Newport, editor in chief of the Gallup Poll, has said that Gallup's designation of likely voter depends on several factors — demographics mostly ("[e]verything else being equal, certain types of people are more likely to vote than others"), although he acknowledges the existence of factors that are "idiosyncratic and can reflect one of a hundred characteristics that come into play in any given election."

(Some organizations define likely voters as those who have some kind of record of participating in midterm elections — so if 2012 is going to be the first election in which you will be old enough to vote, such organizations would not consider you a likely voter, no matter how determined you may be.)

There is some logic in differentiating between those who are merely registered to vote and those who are likely to participate.

As I mentioned earlier, it's pretty easy these days to get your name on the voter registration list — even if it doesn't happen automatically. But actually showing up at the polls and voting is another thing entirely.

Just as it is easy to register to vote, it is also easy to answer a pollster's question about which candidate you support. But it is those who show up who get to make the decisions.

More and more, as Election Day draws closer, pollsters will be emphasizing those voters who are likely to show up — as opposed to those won't make the commitment in time or personal resources to the act of voting.

For Barack Obama, that might not be such a bad thing.

Voters who tend to be regarded as registered but not likely to participate include the young, minorities and low–income people — all groups that propelled Obama to the White House in 2008. Obama's support remains high among blacks, but Molly Ball of The Atlantic has wondered if Obama has problems with young voters (those between 18 and 24).

Obama "remains dramatically better liked by young voters than [Mitt] Romney," Ball writes, but "less than half" support his bid for a second term. Emphasizing only approval and not likelihood of participation could be of benefit to the incumbent — but not if one is familiar with young voters's voting history.

Anyway, most of the polls that you will see at this juncture in the general election campaign are surveys of registered voters. There will be more polls of likely voters the closer we get to the election.

There are a few polls of likely voters out there but not many. Rasmussen released a poll today that was a survey of likely voters — and it shows the race as, for all intents and purposes, a dead heat.

That means little — if anything — about 6½ months before the votes are counted. Besides, most of the polls these days are of registered voters, and those surveys show Obama leading, typically by about four or five points.

Some media outlets have tried to make sense of all the conflicting signals coming from the April surveys, and you can take encouragement from whichever poll supports your position today, but I suggest keeping an eye on the polls of likely voters as we get closer to Election Day.