Continuous Household SurveyMethodologyEach year CSU sets the content of the questionnaire in consultation with client departments. The questionnaire consists of both a household interview and an individual interview with each person aged 16 and over. Both the household and individual questionnaires consist of core items that are included each year, modules that recur on a regular cycle and ad hoc modules. Core items include accommodation, tenure, type of heating and fuel usage, employment status, employment activity, educational qualifications, adult and child health and family information. Non-core items include prevalence of smoking, level of alcohol consumption, and sports and leisure activities.
Sample
The CHS is based on a systematic random sample of 4,500 addresses drawn each year from the Land and Property Services Agency’s (LPSA) list of domestic addresses. The LPSA addresses are sorted by district council and ward, so the sample is effectively stratified geographically. Data is collected by personal interview using CAPI, and the interviews are spread equally over the 12 months from April to March.
Sample
Design and Response
The sample of addresses used for CHS is obtained
from the Land and Property Services Agency
Valuation List. In 2007/08 a simple
random sample of almost one per cent of domestic
properties on the list was selected.
The information covered by the survey is collected
by personal interview throughout the year,
allocated on a monthly basis. Interviews
were sought of all the adult members (those
aged 16 and over) of 4,500 addresses in 2007/08.
Table
A.1 Response from all households.
|
Sample
|
2007/08
|
|
Addresses Issued
|
4,500
|
|
Ineligible
|
529
|
|
Effective Sample
|
3,971
|
|
Response
|
Percentage
|
|
Fully Co-operating including Proxy & Partial
|
65
|
|
Refusals
|
26
|
|
Non-contact
|
9
|
|
Base (100%)
|
3,971
|
Of
the addresses issued 12 per cent were excluded
from the sample as ineligible through absence
of any household at the address or because the
address no longer existed.
The effective sample was 3,971 addresses.
The minimum information accepted from
a household was a fully completed household
schedule including basic demographic information
on all individuals in the household.
A subset of information was accepted
by proxy for individuals with whom it was not
possible for the interviewer to make contact
within the allocation period.
Outright refusals were obtained from
26% of addresses in the effective sample (Table
A.1).
A further 9% of the effective sample
addresses could not be contacted during the
allocation period.
Sampling
Error
Estimates
quoted in CHS tables are based on data collected
from a sample of the population and are therefore
subject to sampling error.
To illustrate this point the confidence
limits relating to a selection of CHS percentage
figures at 95% confidence interval are shown
in table A.2.
It is important to note that some trends
observed in the data may be attributable to
sampling error particularly where the numbers
in the cells are small.
Table
A.2 Sampling error associated with CHS estimates.
|
|
Percent
%
|
95%
Confidence interval & Standard error
(SE)
|
|
|
Lower
|
Upper
|
SE
|
|
Percentage
of households with:
|
|
|
|
|
|
CD
Player
|
72
|
71.11
|
72.89
|
0.89
|
|
Tumble
Drier
|
60
|
59.03
|
60.97
|
0.97
|
|
Computer
|
61
|
60.04
|
61.96
|
0.96
|
|
Base
= 100%
|
2,560
|
|
|
|
|
|
|
|
|
|
|
Percentage
of adults consulting a GP
|
|
|
|
|
|
All
16 and over
|
15
|
14.45
|
15.55
|
0.55
|
|
Base
= 100%
|
4,271
|
|
|
|
|
|
|
|
|
|
|
Percentage
of all persons aged 16 and over smoking
cigarettes
|
|
|
|
|
|
All
16 and over
|
23
|
22.28
|
23.72
|
0.72
|
|
Base
= 100%
|
3,403
|
|
|
|
|
|
|
|
|
|
|
Percentage
all of persons aged 16 and over with
a mobile phone
|
|
|
|
|
|
All
16 and over
|
84
|
83.44
|
84.56
|
0.56
|
|
Base
= 100%
|
4,272
|
|
|
|
|
