A survey receives a thousand answers from people attending a cycling festival. That is a substantial stack of responses. It still does not automatically describe everyone who travels through the city.
Sample size and sample coverage are different questions. Counting more responses does not, by itself, tell you who had a chance to appear in the data.
Name the population before judging the sample
Is the study intended to describe festival attendees, local cyclists or all residents? Those groups are not interchangeable. A sample can be useful for one question and poorly suited to another.
Look for the population definition in the report rather than inferring it from a broad headline.
Trace the route into the dataset
Ask how people were invited, where recruitment happened and whether participation depended on access to a particular service or location. In the invented festival example, people who did not attend could not answer an on-site questionnaire.
Also distinguish those invited from those who responded. A high number of completed forms does not explain the views of everyone who declined or was unavailable.
Read the limits before extending the claim
Check whether the authors describe exclusions, nonresponse or adjustments. Do not assume a weighting method solves every coverage problem; read what it was designed to address.
If the necessary information is missing, say that the sample’s relationship to the broader population is unclear.
Write a narrower sentence
“Among respondents at the festival” may be an honest description where “city residents believe” is not supported. The narrower statement can still be interesting and useful.
This is a conceptual reading exercise, not a claim about a real survey. For the next layer of checking, see the dataset guide. Good interpretation starts by respecting who is actually represented before admiring how large the number is.