Examination of Witnesses (Questions 220-239)
MS KAREN
DUNNELL, MS
JIL MATHESON
AND SIR
MICHAEL SCHOLAR
28 JANUARY 2008
Q220 Mr Love: Can I ask you about
the trade-off, to come back to my original question? There is
obviously to some extent a trade-offand I know you can
do it with layout, making it simplebetween length and the
return that you will get. How are you trying to deal with that?
Ms Dunnell: Actually, in the test
that we did in 2007 we found that there was very little difference
in response between a third page and a fourth page because we
did test the different lengths. We are pretty confident that the
four-page one will work. The temptation of courseand this
goes back to an earlier answer I gave youis to cram all
the questions on to the pages regardless. That would still cost
extra money because it is extra processing, but it does not increase
things like printing and postal costs and so on. Our evidence
suggests that it is not a good idea to cram all the questions
on to these three pages because we then get lots of what we call
question non-response; people will skip questions or misread instructions,
and that is no good either.
Q221 Mr Love: I have just tried to
fill in a bank form and that was bad enough, I can tell you. You
do need to keep it simple. If you cannot get the fourth page and
you do not want to cram, what criteria are you going to use as
to what is in and what is out? What will be the most important
features for you?
Ms Dunnell: The most important
thing about the census is to actually get our count of the population
right. Maybe, Jil, you can talk a bit about the other criteria.
That is our key one.
Ms Matheson: We have had lots
and lots of public consultation about this and set out a set of
criteria that we will use to judge what we will recommend is in
and out. The first one, as Karen says, is does it contribute to
understanding and counting the population. That is priority number
one. The other criteria are, are there alternative sources? Are
there other places, either from surveys or administrative data
or wherever, where that information can be provided? Is the information
required at very small geographic levels? That is what the census
does that no other source can do. Is it required in order to understand
something in relation to other things that are on the census,
i.e., is it a freestanding piece of information that could be
collected in another way or is it integral to the census itself?
There is a whole set of criteria. Acceptability: will people answer?
Q222 Mr Love: Let me ask you two
questions on that list of criteria. You talked about the small
area of statistics that are not available anywhere else and we
have talked to others about the importance of that. I represent
a London constituency where turnover of population is very high,
so the statistics you are talking about are out of date within
two years. You have hardly published them before they are out
of date. What use is that to the people that would normally use
the statistics when they get five, six or seven years out? Whereas
they can use those statistics, they do not have any confidence
that they are now accurate.
Ms Dunnell: That is one of the
reasons we believe why the census is such an important baseline
to do, at the moment, every ten years and that is why increasing
the effort that we put into population estimation in between,
particularly capturing all this information about people coming
into the country, is so important.
Q223 Mr Love: But you do not think
there is another way of capturing those small areas?
Ms Dunnell: Again, if we had really
good administrative data which registered people's changes of
addresses, that would be very helpful but that is exactly what
we do not have. For example, at the moment we use the National
Health Service registers to determine movements within the country.
That is, after many years of research into different registers,
the best possible source that we have at the moment but it is
dependent on people registering their change of address and re-registering
with a doctor when they move and, as we know, there are certain
groups of people who do not bother to do it until they get ill
maybe ten years down the line. This is why our programme of looking
into administrative records is so important and why we are of
the view which we have expressed that until we have much better
mechanisms for recording change of address in these big registers,
they are not helpful for even internal migration.
Q224 Mr Love: The other question
we keep asking people is what is the most important thing and
overwhelmingly they say they want to get population accurate.
Is that the overwhelming priority?
Ms Dunnell: Absolutely.
Q225 Mr Love: If so, should we really
be dispensing with some of the complexity of the form?
Ms Matheson: I think there are
two things. It is the overwhelming priority and one of the reasons
that we think we need a fourth page is because we need extra questions
in the three pages in order to reflect that new complexity: second
homes, citizenship and all of those other things that we really
need in order to be able to understand the population, some of
which we did not include in 2001. The other partand this
goes back to your earlier question about the potential trade-off
between response and lengthwe carried out a postal test
last year of 10,000 households, so it was a good size, and showed
no difference between response where there were three pages per
person and four pages. That mirrors international experience.
One of the things that is important that a fourth page would allow
us to do is to do something which I think will help with response,
which is relevance, making sure that in there there are questions
that are really important and really valued and that you can explain
why they are there, such as carers, and that are relevant to those
groups in society is a benefit of having a fourth page.
Q226 Mr Love: This is coming back
to a question that Mr Ainger asked earlier on, and that is about
public confidence, not for the reasons he suggested but you will
know that there have been large-scale leaks of confidential information
from government over the last few months. Do you think that calls
into question public confidence? Will people be prepared to give
you what is in effect quite confidential information based on
their experience so far with these leaks? Perhaps I can also ask
Ms Dunnell first and perhaps Sir Michael would like to respond
as well.
Ms Dunnell: Obviously, we are
extremely concerned about the loss of data from government departments
and of course we have all been charged by Sir Gus O'Donnell to
put our house in order and it is very reassuring when we have
done a very extensive emergency audit to find that in fact we
are in the clear there. That is very reassuring. The other thing
that we were very concerned about, because of course we have nearly
1,500 interviewers out in the community all over the UK every
day, collecting information on things like the Labour Force Survey
and the General Household Survey, that actually our response rates
were not affected by these losses, which we thought they might
be. That is, of course, largely down to the skills of our interviewers
in explaining the confidentiality of the information that they
provide. Of course, that is much harder to do on something like
a census, although of course you do have that opportunity to have
a really big national campaign about it. It obviously is a major
concern and making sure that everybody understands the confidentiality
with which we hold census data is a very important part of that
big public relations exercise. We are not under-estimating how
difficult it is but hopefully time will be on our side and we
will not have a crisis like we have had just recently in 2011.
Q227 Mr Love: You mean we will have
all forgotten about it by then. Can I ask Sir Michael about security?
Is there a need for greater security to reassure the public and
is that something you will look at?
Sir Michael Scholar: I think the
maintenance of public trust in the confidentiality which the ONS
and statisticians generally in government treat the information
they get from the public is of vital importance. The events in
the last few months have no doubt dealt a blow to that confidence
and I very much hope that the Government can recover from that
and regain the confidence which is so necessary.
Q228 Ms Keeble: I want to ask a bit
about confidence. Sir Michael, you implied just now that it was
down to the Government to regain confidence but in terms of the
census, is it not down to yourself to be able to ensure that the
public and also your customers have confidence in the census results?
Sir Michael Scholar: Yes.
Q229 Ms Keeble: How do you intend
to achieve that?
Sir Michael Scholar: You are absolutely
right. After 1 April it will be the responsibility of my Board
to secure that confidence and maintain it. At the moment, of course,
it is the responsibility of government because the ONS reports
to Ministers and through Ministers to Parliament but I am quite
sure that my Board will take this very seriously. I certainly
do.
Q230 Ms Keeble: Can I just ask, Karen,
just now you set out quite a comprehensive work programme in terms
of what had to happenit is pretty basic stuff, whether
it is a three or four-page questionnairehow you engage
with different authorities and so on to get the hard-to-reach
groups. Do you have a critical path programmed out for that? Is
that formally mapped out and set out?
Ms Dunnell: Our whole strategy
for ensuring response? Yes, it is.
Q231 Ms Keeble: No, I mean for actually
delivering the census.
Ms Dunnell: Yes, it is all part
of the very detailed planning. Jil, can you say a little bit more
about that?
Ms Matheson: The census planning
is being done under normal project management arrangements and
there is a critical path and a set of deliverables between now
and 2013, when the results finally ...
Q232 Ms Keeble: With the kind of
traffic lights?
Ms Matheson: Yes, we use the traffic
light system.
Q233 Ms Keeble: Who is in charge
of that? Who formally heads up that particular process?
Ms Matheson: There is a senior
responsible owner, supported by a project director and the normal
techniques of project management.
Q234 Ms Keeble: Has there been a
change in staff there recently?
Ms Matheson: Yes, there has.
Q235 Ms Keeble: The director in charge
of it? When was that?
Ms Matheson: About two weeks ago.
Q236 Ms Keeble: But the new person
presumably is going to carry it through?
Ms Matheson: That is the intention.
Indeed, that is the commitment, that he will be there until the
census is delivered.
Q237 Ms Keeble: Are you satisfied
with that process? You are satisfied you have the personnel, and
the experience and the skills there now to deliver it?
Ms Matheson: Absolutely, yes.
Ms Dunnell: That is one of the
things that we recognized early on in our planning, that we would
need a Director-level post to lead that work and that person is
in place and is now on our Management Board.
Q238 Ms Keeble: So you did not have
somebody before?
Ms Dunnell: We did not have somebody
at that level. We had Jil at that level before she got her recent
promotion but she did the census plus a lot of other things and
we now have a Director level person whose sole job is the census,
which was always the plan.
Q239 Ms Keeble: This is no comment
on Jil's work, clearly, but do you think that you had enough attention,
if you had one person doing this and quite a lot of other things
at the early stage of the delivery of the 2011 census? With hindsight,
do you think she was overloaded?
Ms Dunnell: She is a very capable
person. I do not think she was particularly overloaded and we
had always planned to have a person at that level when the time
came. The time came and we found that person.
Ms Matheson: Can I just add to
that? I will not say whether I was overloaded or not but it was
also, again, looking back to 2001, and the full-time Director-level
post is now in place a lot earlier than it had been before the
2001 census.
|