A national poll putting the race between incumbent Republican Gov. Susana Martinez and Democratic Party challenger Gary King at a dead heat is making waves this week in the state's political landscape.
The poll, commissioned by Rasmussen Reports, found King and Martinez both garnering 43 percent approval from likely voters. The poll, conducted over July 21-22, sampled 860 voters, according to Rasmussen. The state Republican Party quickly issued a press release questioning the poll's accuracy, criticizing Rasmussen for oversampling the state's young population, which the party argues doesn't vote as much as older people.
"The Rasmussen sample has the same number of voters aged 18-39 as it does over the age of 65," the press release reads. "This is a provably false sample, as voters aged 65 and over always represent a larger share of the electorate, even in a presidential year."
Still, Rasmussen did not publicly release the poll's crosstab data, which breaks down the age and racial demographics of the people who responded to questions. That information is only available to Rasmussen subscribers.
But the polling company's methodology has been questioned before. Two years ago, Rasmussen, then known as a conservative pollster, predicted that Mitt Romney would win the popular vote in the presidential election. After Obama won the election, Rasmussen changed face and became a Democratic-leaning firm, as Harry Enten explained in a column in the Guardian last year (which the state GOP also cited in its press release).
Enten, who now writes for pollster guru Nate Silver's FiveThirtyEight, criticized Rasmussen for basing much of its polling formula on exit polling data, which he argues disproportionately weighs minorities and young voters.
One way to tell whether a poll is skewed or not is to measure it against other independent polls and see if it's an outlier. Unfortunately, the Rasmussen poll is only the second independent poll in the New Mexico governor's race released this year. The last one, commissioned by Public Policy Polling in March, found King five points behind Martinez. That poll, released more than two months before the gubernatorial primary election, found King leading his Democratic opponents by more than 20 percent. He ended up winning that election by 13 percentage points.
Though Public Policy Polling is constantly criticized by Republicans for its Democratic Party affiliations, the firm is largely respected. Still, the Public Policy Polling poll and the latest Rasmussen poll are simply not enough to give voters an accurate picture of where things stand in the gubernatorial race. Both sides have also released their own polls—Martinez' shows her leading King by 16 points and King's shows her leading him by six points—but relying on internal campaign polling is never a good idea.
So for now, the New Mexico governor's race simply doesn't have enough independent polling yet to show clear trends for how things could shake out in November. In the meantime, nothing is prompting University of Virginia professor of politics Larry Sabato's highly regarded Crystal Ball from changing the race from "Likely Republican."