Don’t Get Worked Up About That New Jersey Poll Just Yet

Like electoral clockwork, the first day of the new month brought with it an October surprise. But this one didn’t come in the form of a remark about the “47%” or an old DUI charge; it was a poll (this shouldn’t come as a surprise to anyone reading this blog). More specifically, a poll of New Jersey.

Monday morning, Stockton University released a survey of the NJ Senate race, asserting that Incumbent Democrat Bob Menendez was two points ahead of his Republican challenger, Bob Hugin. For an increasingly blue state in an election year where Democrats are already playing the Senate from behind, that’s bad news.

At first glance, this poll seems pretty reliable — Stockton’s live interviewers used landlines and cell phones to survey 531 likely voters, with a +/- 4.25 percentage point margin of error. Great methodology, right?

Well, no. Patrick Murray, director of Monmouth University’s Polling Institute, put out a long tweet, where he noted many key problems in the survey. I’ll spare you the time and give a rundown of his thoughts:

  1. 92% of the registered voters they asked were deemed likely voters, which is uncharacteristically high.
  2. The sample of this Stockton poll underrepresented young and minority voters, both of which are groups that tend to vote Democratic. Thus, in the weighting process, one “strange” response will skew the results pretty heavily. This has happened before, after all.
  3. Stockton University is not a member of the American Association for Public Opinion’s Transparency Initiative, meaning we may never know what the true raw data looks like.

Murray’s criticisms give us valid reason to discredit this New Jersey survey. I also want to remind you of a constant theme in my posts —  trust an aggregate of polls, not just one. So, what do those averages say?

Though there haven’t been many polls released in New Jersey, Menendez saw a 43-37 lead in an August survey from Quinnipiac. That’s a far cry from his nearly 20-point victory in the 2012 election, and that’s probably due to the numerous scandals and corruption allegations he’s faced since then.

Other indicators are also looking good for Menendez; race raters and forecasters still have him with a comfortable advantage, with FiveThirtyEight giving him an 11-in-12 chance of winning, and others rating the election as either Solid or Likely Democratic.

Therefore, this one survey from Stockton University shouldn’t worry Democrats just yet. If future gold-standard pollsters begin putting out similar results, that might change. For now, though, New Jersey’s Senate race is tightening, but it will take a bigger October surprise to truly shift the tides.

One thought on “Don’t Get Worked Up About That New Jersey Poll Just Yet

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 )

Facebook photo

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

Connecting to %s