Select Committee on Public Administration Minutes of Evidence

Examination of witnesses (Questions 140 - 159)



  140. What does that mean?
  (Mr Rees) It means that the more information Ministers and indeed MPs have when they take the decisions, the better the decisions are likely to be. Let us take the example of electronic government. Government has made it perfectly clear that it wants to see a much greater takeup and availability of electronic channels, means of communication. It is actually rather important to know how many people have access within their homes to the internet, how many people have access to phones, as we drive that programme forward, see how that changes and see how their attitudes change. That is part of the work we shall be doing.

  141. It is not to find out what people want so that governments can do what people want?
  (Mr Rees) What people say they want is one of the factors that clearly governments will take into account. Coming back to my example of electronic government, one of the issues we need to take into account is that if you have never seen a kiosk, you do not know what an electronic kiosk is, then that is going to condition your attitude. It cannot simply be that people are not interested, old people do not like the new technology therefore let us not give old people new technologies. It is part of the package which needs to be built into the policy process.

  142. So the people can be wrong.
  (Mr Page) Absolutely.

  143. That is the point, is it not?
  (Mr Rees) No, I do not think the people are wrong.

  144. You have a difference of opinion there.
  (Mr Rees) The point is that the information you get from people does not dictate policy making it has to be part of an overall package of information.

Mr Oaten

  145. I am keen to understand the process of how the subjects and the questions are actually put together because that seems to be quite critical. Take me through how that happens. Is somebody sitting in a dark room thinking of these questions? How are questions generated?
  (Mr Rees) There are two stages to the process really. Firstly, let us take the next large quantitative wave we are doing. We will ask departments, including Ministers, including people at the centre of government, whether there are issues on which they would like to see some questioning done.

  146. So each of the Government Ministers may have a civil servant come up to them, say this is about to happen and ask whether they have any questions on their mind which they would like to test out.
  (Mr Rees) Yes. Indeed when we started the process the then Minister, Peter Kilfoyle went round, spoke to his colleagues, told them there was this new mechanism, said if they wanted to use it this was the way to go about it. The second way is that we have set up a group which is designed to embrace some of the key research departments within government, but also some of the people outside government. We will always show them the questionnaires which we are proposing. That is an opportunity for them to say that is all very interesting, but we would actually be interested in adding two or three questions on that particular theme, if you are opening that up. It is our contribution to trying to join up government.

  147. Do MORI then, when you have this list and all these ideas for questions, say you cannot actually ask that?
  (Mr Rees) Yes.
  (Mr Page) Oh, yes.

  148. You have quite a strong input.
  (Mr Page) Yes. We will not go and ask questions that the public are going to misunderstand completely or are too long or just not being asked in the right way. We will try to ask questions in ways where the person who is asking the question and the person answering can both understand exactly what it means and in language which the average reading age will actually understand. We do not go and talk about Best Value because they do not know what it is.

  149. I can imagine Jack Straw thinking six months, a year ago, that he wondered what would happen if we got rid of jury services, what the public would think about that. However, he is not likely or would not be keen to come to you with that because that would become public information. How many times do Ministers say to you that they really want to know this but not if it is going to be public?
  (Mr Rees) They have never actually said that. Whether they have thought it I do not know. Our starting point when we set up the Panel was that we were going to start with public services rather than for instance questions such as "Is jury service a good thing or not".

  150. It is a public service, is it not?
  (Mr Rees) It is a public service, but it is not so much the service delivery. Maybe I have got that example wrong but the starting point is to talk to departments about how they are delivering services, the different ways they are doing it and that is where most of our questions have come. We have also asked questions about local democracy because DETR said that they wanted to ask questions. May I go back a stage in terms of the way the process works? On something like what is the public's demand for 24 x 7 services, which is a very difficult question to ask people, we held some focus groups with MORI first to try to get a better understanding of what the right way to put the question is before we then go out to ask 1,000 people. It is an iterative process. We do rely on their advice; we also rely on other advisers within government.

  151. Would you rule in or out a question which was put to the Panel, which was specifically about an Opposition Party's policy, for example people's attitude to putting extra pennies on tax for various public services? Would that be a question you would not want to ask because it is potentially political?
  (Mr Rees) Our general rule has clearly not been to ask political questions. Equally, all of the questionnaires are approved by Ministers and Ministers are even more concerned that they should not be used for the wrong purposes. I would give you a pretty categoric assurance that we would not ask that question.

  152. The folk at Millbank would have as equal access to this information as the folk at Central Office, would they?
  (Mr Rees) All the information is totally published, people can get the different breakdowns. The only issue is whether people know where to find it. I hope hearings like today will spread the message a little bit, but no, we do not keep any of the information secret.

Mr White

  153. I am interested in the statistics of this because it seems to me that you can get national statistics, but if you go back to your example of immigration in Newcastle, how statistically valid are the two people in Newcastle who have used the Immigration Service?
  (Mr Page) It is not. I am merely using it as an example. What I am trying to say is that it is completely meaningless in statistical terms, but if you wanted to draw together for some sort of qualitative exercise, be it a community workshop or focus group or something like that, people who have had experience of services at a particular regional level, MORI would be able to draw down their telephone numbers and invite them to that meeting. Obviously the sample is representative, so if as a percentage of the public we have the correct proportion of the public who have used the Immigration Service in Newcastle, obviously only X per cent of the public as a whole in this country live in Newcastle and only X per cent—

  154. You presumably put the appropriate health warnings on.
  (Mr Page) Of course.

  155. How does somebody looking at the web page understand?
  (Mr Page) Sure. Any statistics which are being published here will all be ones which have a sensible margin of error, three per cent, two per cent, four per cent, something like that. We would not publish anything with sample sizes of below 100.

  156. Do you provide that information to the departments?
  (Mr Page) The department has the data. For a number of reasons we do not analyse data which is going to be so wildly erroneous because the sample sizes are so small that it would be silly to look at it. I can go and analyse groups of 50 people here or 20 there but MORI would not encourage people, government or any other client, to look at that because the margin of error on 50 interviews is about 20 per cent or something. You get 50 people and 50 say they are happy, it could be 70 or it could be 30. It is not very helpful. As researchers we would not encourage any client in any situation to look at that and we certainly would not be publishing it. If we were publishing the results from qualitative research, which is about why people think things but is based on smaller numbers, it would come with a caveat at the front which says this is not statistically reliable, you cannot extrapolate from this, but it does give you a very good indication of how people think about this issue.

  157. One of the purposes of the local authority panels was to transform the culture of local authorities, but you specifically excluded that for government with the People's Panel. Does that either suggest that there needs to be a parallel citizens' panel which is not about changing culture or does it say that there is an issue which you have not addressed?
  (Mr Rees) We are trying to do two things. What local authority panels are about in many cases is indeed trying to bring the local authority closer to the citizens by getting those citizens who are interested in those types of issues. We specifically did not go down that road for the People's Panel, so in that sense it is a representative panel rather than a citizens' panel. It is one of the things we have done quite a lot of work with local authorities on, just to see what the lessons are which can be learnt from the local authority style panels and read across to the national panels which work best and which do not. It is fair to say that some of the local authority panels are miniature versions of the People's Panel. In other words I could not go on one just because I fancied it and they are also recruited in a representative way.

Mr Browne

  158. The conversations and evidence we had last week tended to suggest to us that the benefit of some of these local authority panels was that people did become interested in these issues and expanded their interest and their involvement. Clearly you are trying to avoid that. How do you actually measure whether people are going native other than measuring this base line which you have which is now two years old against other independent surveys?
  (Mr Page) You are absolutely right, it is all about clarity, about what the purpose of the Panel is. Some local panels are deliberately designed to engage local people. This one is not, but it is designed to encourage civil servants across Whitehall to collaborate and to come out of their silos and actually work on things together and to talk about cross-cutting issues. That is in response to Mr White's point. On this one, some of the slides I have put up, what we have done is not only at the base line did we compare their views at the time of recruitment, but subsequent to that, on each consecutive wave, we have put on control questions so we can see that the Panel is saying this but a separate survey we asked on this at the same time came up with the same results. At the time of recruitment we checked but we are also checking on an ongoing basis.

  159. Can you give me an example?
  (Mr Page) For example, on the next survey which is coming out, the consumer test survey, we will have asked about, for the sake of argument, perceptions of local government, their views of their local council, which is something we asked at the time of recruitment. MORI works extensively for local authorities and independently every six months conducts a representative survey of the general public about their views of local authorities. What we will be able to see is whether the members of the Panel's views of their local council have changed in the same way as it has amongst the general public.

previous page contents next page

House of Commons home page Parliament home page House of Lords home page search page enquiries index

© Parliamentary copyright 2000
Prepared 15 May 2000