Magazine article Marketing

Research Slams Polls Practices

Magazine article Marketing

Research Slams Polls Practices

Article excerpt

The inability of leading pollsters to predict the outcome of the 1992 General Election was largely due to technical errors, it was concluded last week.

The verdict is a damning indictment of the market research industry's practices.

An exhaustive report, commissioned by the Market Research Society, identified three main reasons why the leading polling organisations were four percentage points away from the actual election outcome.

First, this discrepancy was largely due to the pollsters' sampling techniques which did not accurately reflect the social profile of the electorate at the time; second, the pollsters failed to compensate for Conservative supporters' reluctance to reveal party loyalty.

And third, the late swing to the Tories caught them all completely unaware.

To avoid similar mistakes being made again, the MRS is recommending the use of more accurate sampling methods to ensure the surveys are as representative as possible. It also suggested the use of "adjusted" polls based on previous election patterns and secret ballots to overcome people's reluctance to state their voting intention.

Some pollsters have already experimented with new techniques, as results from the June European election showed. …

Search by... Author
Show... All Results Primary Sources Peer-reviewed


An unknown error has occurred. Please click the button below to reload the page. If the problem persists, please try again in a little while.