Tuesday 19 February 2008
The British Polling Council (BPC) is currently investigating the way in which findings have been published from a survey conducted by Ipsos MORI for Transport for London (TfL). Ipsos MORI does not accept that it is in any way at fault in the way these findings have been published, or that the BPC has jurisdiction over the publication of this survey.
Contrary to a report in the London Evening Standard, the survey was not conducted wholly or partly for the Mayor of London. The survey in question was conducted entirely for TfL, was entirely concerned with transport policy, and included no questions asked on behalf of the Mayor. The questionnaire from the survey included no party-political questions.
The purpose of the BPC, and the purpose for which Ipsos MORI joined the BPC, is to ensure transparency of methodology in political polls, and its disclosure rules are designed with this end in mind. The survey conducted for TfL is part of ongoing research our organisation undertakes on behalf of TfL. It is not by any standard a voting intention or other 'poll', but rather a survey which is part of a wider programme of consultation in line with TfL's statutory responsibilities. We do not consider that it comes within the jurisdiction of the BPC, or that the disclosure procedures laid down in the BPC code are appropriate to this research. It was conducted under the normal contractual arrangements for such surveys, which give the client the absolute right to decide what and how to publish, subject to our professional responsibilities under the Market Research Society (MRS) Code of Conduct.
All Ipsos MORI's research is governed by the MRS Code of Conduct, which includes provisions to prevent the publication of research in a dishonest or misleading way. We are confident that all the requirements of the MRS code with regard to this survey have been met, in particular that all the relevant technical details needed to interpret the research or assess its validity have been made public, and that there is nothing unpublished which would change the interpretation of the published figures.
Full technical details of the survey are included in the report which was published on TfL's website on 13 February:
These include fieldwork dates, sample size, demographic and geographic breakdown of the sample, and the full wording of all questions from which findings have been released.
Ipsos MORI's record on transparency in political polling is a very strong one. We will continue to make available the fullest details of our methodology and findings, as has been the case since MORI was founded in 1969, to a greater standard of openness than that prescribed by the BPC's own rules. However, the bulk of Ipsos MORI's business consists of surveys that are not political opinion polls. We apply the strictest professional standards to these surveys, in line with our responsibilities to the MRS, and ensure to the fullest of our ability that findings from such surveys are not published in a misleading way; but the disclosure procedures designed for the specific case of political opinion polls would neither be appropriate nor would serve the legitimate interests of our clients.