Have Pollsters Forgotten That Nevada Is An Early State?

Welcome to Pollapalooza, our weekly polling roundup.

Poll(s) of the week

There are few constants in the presidential primary, but the order in which early states vote is roughly set in stone: first Iowa, then New Hampshire and then either Nevada or South Carolina, depending on the party.¹ This is critical to how candidates operate in a presidential primary because what happens in one state can influence what happens in the next state to vote. Knowing this, campaigns often tailor their strategies to how they can best contend in each individual state, emphasizing the early-voting states as they try to advance deeper into the primary.

This is why it is important to know how candidates are faring in each state, rather than just how they are performing nationally. We’re about six months away from the first nomination contest in Iowa, but a look at the number of surveys conducted in each state gives us a window into which states are — and aren’t — being polled enough based on their position on the primary calendar. With the addition of an Iowa survey by Monmouth University released on Thursday, Iowa now has more Democratic presidential primary polls than any other state thus far in the 2020 cycle — 22 surveys. But while Iowa leads the way, another early-voting state, Nevada, has been surveyed just three times! Granted, Iowa does hold the first caucus, but even states like South Carolina, California and Texas, which go after Nevada, have been polled far more. And yes, each of these states could host a crucial contest, but so could Nevada!

However, the dearth of polls in Nevada isn’t anything new. The Silver State only became an early state in the 2008 nomination process, and it’s struggled since then to attract pollsters’ attention. In 2008, at the same point that we are at now — about six months before Iowa — just 3 percent of the state-level Democratic and GOP primary polls had come from Nevada. During the 2012 Republican primary, about 5 percent of state-level polls were from Nevada six months before Iowa. And by that point in 2016, only 3 percent of Democratic polls and 2 percent of GOP polls came from there. For comparison, the three polls conducted up to this point in the 2020 cycle equal just 2 percent of all state-level surveys.²

One reason for the lack of polls is that it’s hard to poll Nevada. The caucuses aren’t as embedded in the state’s political fabric as contests in other early states — Iowa’s caucuses have led since 1972, New Hampshire has held the first primary since 1920 and South Carolina’s “First in the South” primary dates back to 1980 on the Republican side. And considering caucuses have lower turnout than primaries, it’s challenging for pollsters to survey participants, too, especially given Nevada’s fairly transient population and the unusual hours many residents work because of the state’s hotel and gaming industries. All of this makes it difficult — and costly — for pollsters to survey Nevada caucus-goers.

Nevertheless, veteran Nevada political journalist Jon Ralston isn’t totally buying the claims that Nevada is too tough to survey more regularly. “I think ultimately it’s a cop-out by pollsters and by others,” he told me. But Ralston also senses that the campaigns and media aren’t giving Nevada its due either. “I don’t think people take us as seriously as an early state as they take Iowa and New Hampshire, who have been there forever,” he said. Nevada may be the frontier of polling, but a state with this much say early in the nomination process should be surveyed more.

Other polling bites

• Morning Consult has released its monthly numbers on President Trump’s approval rating in each of the 50 states, finding that he had a net negative approval rating (approval rating minus disapproval rating) in 30 states in July. Trump was underwater in all three states that made the difference in the 2016 election: Michigan (-11), Pennsylvania (-8) and Wisconsin (-14).
• A new report from the Pew Research Center found that Americans remain supportive of separation of powers in government, but perhaps less solidly than before. Pew’s polling showed that 66 percent felt it was “too risky” to give the president more power to tackle many of the country’s problems while 29 percent said that the country could more effectively handle problems if presidents “didn’t have to worry so much about Congress or the courts.” Back in 2018, 76 percent felt giving the president more power was too risky and 21 percent said presidents could operate more effectively if they had less concern about Congress and the courts. Most of the change comes from Republicans: 70 percent said in 2018 that they opposed giving more power to the president, but only 51 percent did in the new poll. By comparison, Democrats saw virtually no change: 82 percent said they felt it was too risky versus 83 percent in 2018.
• North Carolina could be a key battleground state in the 2020 general election, and SurveyUSA has some early head-to-head numbers for Trump against different Democratic contenders. Trump performed worst against former Vice President Joe Biden, earning 41 percent support to Biden’s 49 percent. Trump also trailed Sen. Bernie Sanders by 4 points and Sen. Elizabeth Warren by 1 point, but he led Sen. Kamala Harris by 1 point and Pete Buttigieg, mayor of South Bend, Indiana, by 2 points.
• In their weekly poll, The Economist and YouGov found a clear partisan divide over what statistic best measures Trump’s economic performance as president. Forty-five percent of Republicans felt “the unemployment rate and jobs reports” are the best way to judge Trump’s work on the economy, while only 16 percent preferred “the prices of goods and services you buy” (the remainder answered “the stock market index,” “personal finances” or didn’t know). Meanwhile, 36 percent of Democrats said the price of goods and services are the most important way to rate the president’s economic performance compared to 19 percent who preferred the unemployment rate and jobs reports (the remaining respondents chose one of the other three options).
• Morning Consult also asked Americans about whether 99 different activities ought to be considered sports and what attributes are most important to determining whether an activity is a sport. For example, Americans uniformly agreed that basketball is a sport — 97 percent said it was — but were divided over how to define miniature golf. Fifty-two percent said it was a sport, while 46 percent said it wasn’t. The polls also found a nearly even three-way split over what aspect is most important in determining if something is a sport: 34 percent said the physical effort required, 34 percent said the skill involved in playing and 32 percent said the competitive element of the activity.
• Ever wondered what Americans’ preferred sandwich is? YouGov recently found that 79 percent of Americans “really like” or “somewhat like” a grilled cheese, which tops the list of 15 sandwich options. Bringing up the rear was the French dip, with 46 percent of respondents expressing fondness for it.

Trump approval

According to FiveThirtyEight’s presidential approval tracker, 42.1 percent of Americans approve of the job Trump is doing as president, while 53.2 percent disapprove (a net approval rating of -11.1 points). At this time last week, 42.5 percent approved and 52.9 percent disapproved (for a net approval rating of -10.4 points). One month ago, Trump had an approval rating of 42.6 percent and a disapproval rating of 52.5 percent, for a net approval rating of -9.9 points.

Generic ballot

In our average of polls of the generic congressional ballot, Democrats currently lead by 6.2 percentage points (46.1 percent to 39.9 percent). A week ago, Democrats led Republicans by 6.1 points (46.1 percent to 40.0 percent). At this time last month, voters preferred Democrats by 6.2 points (46.3 percent to 40.1 percent).