When reporting opinion polls, the BBC’s Editorial Guidelines set down in some detail the way information should be presented and the type of language which is appropriate. The Guidelines say that “Accuracy, credibility and impartiality are as important when the BBC reports on ‘polls’ and ‘surveys’ as elsewhere....(When they are) commissioned by the BBC (they) carry reputational risk, so care must be taken to ensure that the audience can trust their findings, and that we do not give them undue weight when reported… For any BBC-commissioned opinion poll, the methodology, data and accuracy of the language used to report it, must stand up to scrutiny.”
The full guidelines on Opinion Polls, Surveys and Votes are available here, but these are some key extracts which all output producers reporting the polls should know:
- the result of an opinion poll should not be the lead or be headlined in broadcast or other output, unless it has prompted a story which itself merits being the lead or headlined and reference to the poll’s findings is necessary to make sense of the story
- language should not give greater credibility to polls than they deserve. For example, polls ‘suggest’ and ‘indicate’, but never ‘prove’ or ‘show’
- the BBC should report the methodology used, the organisation which carried out the poll and the organisation or publication which commissioned it. Such polls should not be described as ‘a BBC poll’. All relevant details, including the questions, results and sample size, should be made available so the audience can understand the methodology and results where editorially relevant, dates of the fieldwork and subsequent events which may have shifted opinion should be reported.
Further advice is available from Ric Bailey or Matthew Eltringham.