Polling sounds precise. It is one of the easiest tools to get wrong. OZJF uses polls to help readers read the public mood. We do not treat a single headline number like a verdict.
How we read polls
Field date matters. Question wording matters. The sample frame matters. A poll run during a war, a terror attack, or a campus crisis may tell you something real. It may not tell you what a calmer survey would tell you. So we care more about patterns over time than about a single number pulled out for effect.
What this section should help with
Before you absorb a headline claim, you should see four things. What the poll asked. When it was run. Who was surveyed. What the main caveats are. This matters most on questions about Hamas support, Palestinian governance, antisemitism, campus climate, and U.S. views of Israel. Loose poll language in these areas often turns into bad argument.
Our target matches the norms from AAPOR and the survey practice at Pew Research Center. If a page uses a number, readers should be able to find the field dates, the sample, and the question wording. They should also see enough method context to judge whether the number is solid, narrow, or easy to abuse.