Imagine that a school asks pupils which lunchtime activity they would like next term. Of the 120 pupils invited to answer, 60 return the survey. Thirty of those respondents choose a gardening club.
One report says, “Half of the pupils want gardening.” Another says, “Half of the pupils who answered chose gardening.” The second statement describes the result more precisely. The survey does not reveal the preferences of the 60 pupils who did not answer.
This difference matters when the school makes a decision. Thirty responses support an interest in gardening, but they do not establish what every pupil wants. Staff could ask non-respondents again or discuss why some pupils did not take part. Perhaps the form was easy to miss, or perhaps those pupils were unsure. Those are possibilities to investigate, not facts supplied by the result.
The wording of the question also affects what can be concluded. If pupils may choose only one activity, the survey identifies their first choice among the listed options. It does not show every activity they would enjoy. If an activity is absent from the list, its popularity cannot be measured by that question.
A useful report gives the question, the number of responses and the limits of the evidence. Careful wording does not make the result less valuable. It helps readers understand what the numbers actually show.