Polls that are commissioned by advocacy groups are always skewed. Sometimes subtly, sometimes grossly, but they can never be trusted without a deep dive into their methodology, the specific questions being asked and the wording of the questions.
I already wrote about how CAIR used an absurdly poor methodology to conclude that more American Muslims would vote for Jill Stein than Kamala Harris, something that exit polls showed was not close to true.
J-Street on Thursday issued results of their comprehensive survey of American Jews on how they voted, and found that Jews preferred Kamala Harris by a 71-26 margin, a 45 point margin. But the Fox News exit polls found a 66-32% split, a 34 point margin.
That's a huge difference. What accounts for it?
It comes down to methodology.
Fox News asked people after they voted, "What is your present religion, if any?" and reported on how many people said their religion was Jewish.
The J-Street survey asked the identical question in their survey of American Jews, yet only 76% of them answered "Jewish." The follow-up question for those who answered negatively: "Even though you do
not consider your religion to be Jewish, do you consider yourself Jewish?" and the answer was 100% "yes."
Fox News asked about religion; J-Street asked about what they call themselves.
A similar thing happened in 2020: J-Street found 77%-21% for Biden; news organization exit polls found 68%-30% among Jews. But their definitions of who is Jewish are vastly different.
Notice that J-Street didn't point out that even according to their own methodology, the gap between Jews voting Democrat and Republican this year narrowed by 11 percentage points.
The 26% of self-described Jews who do not consider their religion to be Judaism is an accurate number, according to Pew. But it shows how polls can be skewed by changing the definition of what a Jew is. And it has a huge impact on the answers to other questions as well.
For example, J-Street asked "Thinking about negotiations between Israelis and Palestinians to resolve their conflict, do you think the U.S. should act as a fair and even-handed broker between Israelis and Palestinians or do you think the U.S. should side with Israel?" They found 51% do not want the US to side with Israel. If you take away the 26% who do not consider their religion to be Jewish, the numbers would flip in the other direction.
Other polls have consistently found that the stronger the Jews identify with Judaism, the stronger they support Israel. Orthodox Jews overwhelmingly showed support for Trump while exit polls show that atheists were strongly pro-Harris.
The atheists are J-Street's target Jewish audience - people who are Jews in name only.
That is not the only problem with the J-Street survey. Like their previous ones, use loaded questions as well to get the responses they want. Any question that begins with "As you may know..." subtly instructs the person being surveyed with information that is highly selective and possibly wrong. Here's one:
When, in the past several years, have there been any negotiations on that level of detail? The last time Israel and Palestinians negotiated even indirectly together for a final peace agreement was over ten years ago, when the Palestinians rejected the US framework for a potential deal. The question does not mention that the negotiations "fell short" because Palestinians rejected even a John Kerry plan that went beyond what Israel offered. Wouldn't that affect the answers?
As always with J-Street, it offers highly selective "facts" to prompt an answer that they want.
Imagine if a question was prefaced this way:
As you may know, for over 80 years Palestinians have rejected every attempt to support a peaceful two state solution between themselves and a Jewish state, no matter what borders and terms were offered. Bill Clinton and John Kerry's plans for peace were rejected. Meanwhile, surveys show that Palestinians consistently and overwhelmingly support terror attacks against Jews in Israel and the Palestinians even even pay the terrorists salaries. Should Israel be pressured to give more concessions to them in order to create a Palestinian state that could very easily turn into an Iranian proxy like Lebanon and Yemen?
This all brings up a much bigger problem - news organizations blindly report polling data, even from the sketchiest sources, as if they are science. The many ways they can be manipulated are not considered. (I found an egregious example, where an anti-Israel organization that innocuously calls itself "Listen to Wisconsin" made it sound like voters in swing states consider support for Gaza to be a huge issue, when in fact their survey only included people who are already strongly anti-Israel.)
Commissioning polls is an easy way to issue press releases and get in a news cycle. Partisan organizations can manipulate even polls done by solid survey organizations by how they word the questions, how they choose the people being polled and even the order of the questions.
The bottom line is, don't trust a news report of a poll by an organization that has reason to manipulate it.
