Examination of Witnesses (Questions 240-259)
MS KAREN
DUNNELL, MS
JIL MATHESON
AND SIR
MICHAEL SCHOLAR
28 JANUARY 2008
Q240 Ms Keeble: Have you hit all
the points on your critical path? Do you have all the traffic
lights green that you should have?
Ms Dunnell: At the moment, yes,
we are doing quite well. Our most important one coming up is the
procurement to finalise who is going to take forward this big
task, and we are expecting to get to that point in May. We have
an OGC check coming up just before that, so we are working very,
very hard to make sure we get through that.
Q241 Ms Keeble: You said just now
that you are going to rely on local authorities to deliver some
of the details of what you needed with the hard-to-engage groups.
You said previously also that the lists provided by the local
authorities were not adequate. What is the position there? Since
you can get computer programmes that will do some of the work
on hard-to-reach groups and where they are, could those not be
used?
Ms Dunnell: We are not relying
on local authorities to do this. What we are doing is working
much more closely with them than we did in 2001 because, of course,
they are the main users of the results at the end of the day and
the first people to jump up and down and complain if they think
we have not got the answers right. So we are working very closely
with them on whether they have data sources that can help us plan
the census and whether they have data sources that can help us
quality-assure them and whether they can do all those things like
help us find enumerators, identify community groups, provide resources
and so on. That is the kind of thing we are doing. We are not
just handing the problem over to them.
Q242 Chairman: Just clarify for me
the total amount that you got in the whip round the other departments?
You got the £25 million for the fourth page but that is not
the total of your bid?
Ms Dunnell: For the census?
Q243 Chairman: For the additional
money that you were short of.
Ms Dunnell: The additional money
for the census, yes.
Q244 Chairman: You only need £25
million in total?
Ms Dunnell: Yes.
Q245 Jim Cousins: Ms Dunnell, can
I take you back to the point you made about the fact that you
had 1,500 interviewers going out day by day? How many of those
people experience abuse, verbal or physical, in the course of
their work?
Ms Dunnell: I do not actually
have the figures to hand but what I do know is it is quite rare
actually. They are very highly trained. I think they probably
get more difficulty from dogs than people actually. We do, of
course, send letters in advance saying that somebody is coming,
so if people really have a very strong aversion to somebody coming
round to interview them, they tend to phone up or write in so
that we can prevent it. It very rarely comes to my attention that
they have difficulty.
Q246 Chairman: I think the Committee
would probably be grateful for any hard information about that.
Ms Dunnell: Yes, we can certainly
do that for you. [1]
Q247 Chairman: Do you get any reports
through your permanent team of interviewers of neighbourhoods
where they do not feel safe?
Ms Dunnell: Yes. We have an awful
lot of inside information about that because interviewers, by
and large, get used to the areas that they work in, so they develop
an enormous amount of local knowledge about how to get into gated
communities, how to get through security systems, what are good
times, et cetera. As I said, we do write out to people in advance
so that if there are problems we try, where we can, to get telephone
numbers and so on. An awful lot of their training is in how to
actually find people in difficult circumstances. We are using
some of the information that our interviewers have, of course,
on the Census to identify just those areas where we need to put
extra resources in.
Q248 Jim Cousins: Just to be clear
about this, you are saying that there is some information you
have about areas where interviewers do not feel safe. You just
now made reference to the use of telephone numbers. Do you accept
that the number of people who have declared information about
landlines is now going down rapidly?
Ms Dunnell: Yes, we do believe
that the number of people who have landlines is going down.
Q249 Jim Cousins: Not the people
who have a landline, but the people who have a landline and are
prepared to have the number made available.
Ms Dunnell: Listed, yes. That
is definitely going down and we do not rely on it. What I am referring
to is that when we write out to people, if people are going to
have a problem helping us then we encourage them—
Q250 Jim Cousins: What percentage
of people have landlines?
Ms Matheson: 90% of households
have a landline and that has gone down.
Q251 Jim Cousins: How many of those
landlines are, as it were, open access landlines where the information
is declared?
Ms Matheson: I do not know.
Q252 Jim Cousins: I would be grateful
if you could get that information for the Committee.
Ms Dunnell: The proportion of
them that are x-directory is what you are looking for?
Q253 Jim Cousins: Yes.
Ms Dunnell: I am sorry, we do
not know that offhand. [2]
Q254 Jim Cousins: Do you accept that
the areas where people are, for a variety of reasons, hard to
measure, where there is considerable churn and transience of population,
and where there is a great deal of volatility about the numbers
of people, are not spread randomly across the country, they are
concentrated into particular local authority areas, particular
neighbourhoods?
Ms Dunnell: Yes.
Q255 Jim Cousins: What impact does
that have on the margin of error of your overall statistical information,
particularly about mid-year estimates?
Ms Dunnell: We have just published
indicators of churn which suggest that some local authorities
are experiencing more than a quarter of their population changing
each year and relatively stable communities where it is 5% or
less. There is a huge variation and it does make a difference.
One of the reasons we developed that measure was because it is
an extra bit of information on top of the population estimates
about the kinds of communities that live in those places. I do
not think it is really new to areas like Camden and Islington,
for example, that they have got a higher level of population churn,
but having statistics about it makes it very clear and obvious
that they are in the top ten, as it were, and they can use that
information and government departments can use that information
in bringing an indicator like that to bear on decisions that they
may make about resource allocation and so on.
Q256 Jim Cousins: What is the margin
of error about the mid-year estimates bearing that point in mind?
Does that margin of error vary between areas of low churn and
high churn?
Ms Dunnell: What we know about
our population estimates is that there is three elements to it:
births, deaths and migration. As we have discussed, the migration
of all of those is the one we estimate rather than count because
births and deaths we have 100% registration of in the United Kingdom
so we can be absolutely sure about those. Migration is something
that we have to estimate based on samples. Obviously we are less
confident in statistical terms about that. The interesting thing
is not only has international migration increased but also so
has internal migration, which is why we are very interested in
tapping into other sources of information which will help us get
a better picture of who is where local authority by local authority.
Q257 Jim Cousins: So what is the
overall margin of error in your mid-year estimates of population
and does that vary according to the area?
Ms Dunnell: Yes. It will vary
by area because—
Q258 Jim Cousins: What is the margin
of error and how much does it vary?
Ms Dunnell: I am not sure that
we have got exact measures of this because it is very, very complicated
to do and I am not sure that we have got the methodological wherewithal
to do that. We do acknowledge that it is the migration part of
it which is the most difficult to establish. This year, we have
only just introduced the new method of apportioning international
migrants using the Labour Force Survey. I think what we can do
now is a bit more work on trying to establish the reliability
of that in statistical terms. [3]
Q259 Jim Cousins: I want to come to that,
but before I do I just want to be clear about this. You have just
set out very clearly the difficulties you have with certain key
factors in components of the population and you have just told
the Committee you do not have any sound margin of error for your
mid-year estimates, nor how that margin of error might vary taking
into account the factors that you have just referred to, that
you do not have that.
Ms Dunnell: We do not publish
estimates of the statistical error, no.
1 Ev 282 Back
2
Ev 283 Back
3
3 Ev 286 Back
| 
