May 3, 2016 | No Comments
Donald Trump has a big double-digit lead in the polls. Or Ted Cruz might be within striking distance of the GOP front-runner.
As Indiana prepares to vote Tuesday, no one really knows.
With only seven public telephone surveys in Indiana in the past month, the state hasn’t seen the same attention as other major primary contests. That’s fewer than the nine public phone polls in the last month before the Wisconsin primary, or the 17 before last month’s New York primary.
The relative dearth of public polling — and the variance of the polls that have been released, including one that showed Cruz leading by 16 points and a few more recent surveys that gave Trump similarly large leads — is the result of a number of factors: a longer primary process than some organizations budgeted for, the relatively short one-week run-up to the primary and unique state laws that make it more difficult to conduct some kinds of phone surveys.
It adds a layer of uncertainty to Tuesday night’s high-stakes primary, in which Trump can strengthen his chances to win a majority of delegates on the first ballot at the convention with a decisive victory in Indiana or Cruz can get a burst of adrenaline.
Struggling to rebound from third-place finishes in five of the six most recent states, Cruz’s campaign has been eager to dismiss the new Indiana surveys showing the Texas senator trailing Trump.
“The polls have been all over the place,” Cruz said Monday on Mike Gallagher’s syndicated conservative-talk-radio show. “The media’s trying to convince you it’s hopeless. I can tell you right now, Indiana is neck and neck.”
Of the seven public polls conducted over the past three weeks, Trump leads six of them. Cruz’s only advantage came from an Indiana University-Purdue University Fort Wayne survey that showed Cruz with a massive 16-point lead.
IPFW pollster Andy Downs told POLITICO his team was surprised when the results came back with Cruz way on top – even wondering if the results for Cruz and Trump were incorrectly reversed.
“At one point, we asked ourselves, ‘Did we code these wrong?’” Downs said.
There are two ways in which the IPFW poll differs from the other surveys, however: First, a long, 15-day field period from April 13-27. Second, IPFW didn’t weight results according to demographic targets.
Demographic weighting allows pollsters to adjust their surveys to make sure they have a representative sample — that they don’t have too many women and too few men, for example, which sometimes results from calling landline phones. Similarly, telephone polls are more likely to capture older voters and could under-represent younger voters without these adjustments.
Downs said he declined to weight to project the electorate because it has been so long since the state held a competitive GOP presidential primary that past targets weren’t available.
“If you want to look at an exciting [Republican] primary” in Indiana, Downs said, “you have to go all the way back to 1976.”
(If the lack of weighting is going to have an impact, Downs said, it will be in the Democratic race. IPFW has Hillary Clinton well ahead of Bernie Sanders, while other polls show a much closer race. Downs said his poll likely under-represents both young voters — who’ve backed Sanders by huge margins in other states — and seniors, who’ve been more likely to support Clinton.)
The other public polls show Trump ahead, but by varying margins. His slimmest lead came in a one-day survey conducted last Wednesday by Ohio-based Clout Research, in which he held a 2-point advantage over Cruz, 37 percent to 35 percent.
Trump’s largest lead, 17 points, was in a poll conducted last Thursday and Friday by Gravis Marketing, a prolific but less-respected firm. (Gravis usually conducts automated surveys, but Indiana is one of two states where those are illegal unless they are preceded by a live caller — one potential explanation for the relative paucity of public polls.)
The most recent reliable poll, an NBC News/Wall Street Journal/Marist survey conducted last Wednesday-Friday and released on Sunday, showed Trump with a commanding 15-point lead over Cruz, 49 percent to 34 percent.
Lee Miringoff, the director of the Marist College Institute for Public Opinion, said the poll came in the wake of two newsmaking maneuvers the Cruz campaign has attempted: the deal with Ohio Gov. John Kasich to allow Cruz a one-on-one shot at Trump in Indiana, and naming one-time rival Carly Fiorina as Cruz’s running mate.
“It looks like, from our numbers, none of that worked out too well for him,” Miringoff said.
Cruz’s camp pushed back, questioning Marist’s track record.
“That particular poll has been way off, over and over and over and over,” Fiorina told former Indiana GOP campaign hand Pete Seat on Seat’s podcast. “In some cases, they’ve been off 22 points a day before an election.”
Asked to respond to Fiorina, Miringoff cited a line from the Mel Brooks western spoof “Blazing Saddles”: “Can’t you see that’s the last act of a desperate man?”
Marist’s biggest miss of the primary campaign was in the Michigan Democratic primary – which Bernie Sanders won by less than 1.5 points after Marist’s final poll showed him 17 points behind Hillary Clinton.
“It’s not accurate” to say they have been “way off,” Miringoff said. “In fact, we’ve had a very good track record in the primaries – which are not easy.”
IPFW’s Downs said Monday he knows the poll is out on a limb showing Cruz ahead, but he stressed that he doesn’t yet know why his numbers differ from other pollsters.
“We really have no idea what the electorate will be,” he said.
Nolan McCaskill contributed to this report.