How many voters are actually undecided?

To the obvious delight of Dave Scharfenberg and myself, over the last three weeks there have been four separate polls conducted on the Rhode Island governor’s race: Rasmussen, WJAR/Quest, WPRI/Fleming and now Brown University. Here’s an updated version of the chart I posted last week showing how the top three candidates fared, as well as the percentage of undecideds, in the surveys:

Except in Democrat Frank Caprio’s case – he consistently wins about a third of voters – these polls are all over the map, with a 10-point spread between the highest and lower results for both independent Lincoln Chafee and John Robitaille, the Republican.

And when it comes to undecided voters, the divergence is even starker – Brown found a whopping 30% of voters haven’t mind up their minds, whereas Rasmussen said just 9% haven’t.

Anything’s possible, but it’s hard to believe the share of undecided voters more than tripled in the week and a half between those two surveys. Digging in a little deeper, one factor I see is that in both Rasmussen’s and our own WPRI/Fleming poll, more effort was made to probe whether self-described undecideds actually had a preferred candidate.

Marion Orr

Marion Orr, who oversees the Brown poll, told me most of the calls for his poll were made by Brown students trained and paid for their work, although sometimes he supplements them with outsiders. Orr also said they are specifically dissuaded from probing voters further.

“They’re trained not to sway the respondents,” he told me. “If they say they’re undecided, that’s what we put them down for.”

Brown’s methodology often leads to a high undecided figure – the Taubman Center had 19% of Rhode Islanders undecided between Obama and McCain two months before the last presidential election [pdf].

The problem there is that winding up with such a large share of undecided voters can limit what the poll tells us. It’s fine if those voters are truly undecided – but as the other poll results showed, it’s possible that further questioning could find a large number of them leaning one way or another, which is what we really want to know. Indeed, Victor Profughi, who did the controversial WJAR/Quest poll, told me last week he regretted not doing more to see whether undecided voters were actually leaning one way or another.

Another question is, who are we polling?

Brown surveyed 565 registered voters, while the other three polls all talked with likely voters, defined different ways. Although Rhode Island has around 700,000 registered voters, only about half of them are expected to show up at the polls next month – and what we really want to know is which way the half that votes is leaning. That’s why Joe Fleming limits our WPRI polls to likely voters as we get close to an election.

For the record, Orr said he did screen his respondents to see which ones were likely to vote, but he wound up deciding to release the results for registered voters instead. “We wanted to include a broader sample,” he said.

“People are still making up their minds,” Orr added. “It’s still a close race, in the sense that there are so many people who are undecided.” About half of those undecided voters uncovered by the Brown poll described themselves as independents, he noted. “I suspect these last few weeks will be decisive.”

3 thoughts on “How many voters are actually undecided?

Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s