A poll appears in today’s news beside a major announcement. The layout encourages a quick conclusion: people have reacted to the announcement. But the interviews may have ended before it happened.
Find the fieldwork dates before making that connection. The Roper Center’s survey terminology guidance distinguishes the period when responses were collected from the date results were released or added to an archive. Publication tells you when the findings became available, not necessarily when people expressed the opinions.
Put the event on the same timeline
Imagine a fictional poll conducted from Monday to Wednesday, released on Friday. A policy announcement on Thursday cannot explain answers already collected earlier in the week. A responsible description would say that the poll preceded the announcement, rather than calling it an immediate public verdict.
If fieldwork straddles an event, the interpretation is more complicated. Some respondents may have answered beforehand and others afterwards. Unless the data and study design support a comparison, do not invent a before-and-after effect from the combined result.
Longer collection periods can also blend changing circumstances. This does not automatically make the survey poor quality. It changes the time span the result can reasonably describe.
Dates are necessary, but not sufficient
Check the population surveyed, question wording, collection method and sample information as well. A recent date cannot make two differently worded questions directly comparable. Likewise, a large sample does not resolve a mismatch between the respondents and the population named in a headline.
When sharing a result, include the fieldwork period near the percentage. Keep the release date separately if it matters to the news story. This small distinction stops an old snapshot from being mistaken for a live reaction and gives readers a fairer basis for comparing the next poll that appears.