« Previous | Contents | Next »
Listen
7. The Scottish Household Survey
7.1 General information
The Scottish Household Survey ( SHS) started in February 1999. Each year, about 15,500 private households across Scotland are interviewed. The SHS is designed to provide results which are representative of Scotland as a whole for each quarter, and for each local authority every two years. Higher sampling fractions are used for Council areas with small populations, to ensure a minimum of about 550 household interviews per Council every two years.
The survey interviews, which lasted an average of 44 minutes in 2001/2002, are carried out in respondents' homes using Computer Aided Personal Interviewing ( CAPI) by TNS and MORI Scotland. The interview has two parts. The first part is with the Highest Income Householder or his/her spouse or partner. This collects mainly factual information about the household and its members. The second part is with a randomly-chosen adult (aged 16+) member of the household. This focuses more on individual attitudes and behaviours. Both parts of the interview may be with the same person. After reweighting, the results from the first part interviews should be representative of households, and the results from the second part interviews should be representative of adults. The response rate for the first part was 67% for the 2001/2002 two-year "sweep". Among households which participated, the response rate for the "random adult" part of the interview was 93%.
The data are reweighted to take account of the over-sampling of the Councils with smaller populations, the under-sampling of adults living in multi-adult households (they are under-represented, relative to their share of the population, because only one "random adult" is interviewed in each household), and the uneven spread of interviews over the days of the week (which would affect the representativeness of the Travel Diary results) .
More detailed information is available from the publications mentioned in section 7.3.
7.2 Sampling variability
The likely extent of sampling variability can be quantified, by calculating the "standard error" associated with an estimate produced from a random sample. On average:
- only about one sample in three would produce an estimate that differed from the (unknown) true value by more than one standard error;
- only about one sample in twenty would produce an estimate that differed from the true value by more than two standard errors;
- only about one sample in 400 would produce an estimate that differed from the true value by more than three standard errors.
By convention, the "95% confidence interval" is defined as the estimate plus or minus about twice the standard error, because there is only a 5% chance (on average) that a sample would produce an estimate that differs from the true value of that quantity by more than this amount. The table on the next page shows the "95% confidence limits" for estimates of a range of percentages calculated from sub-samples of a range of sizes.
As an example of the use of this table, Table 11 shows that there were 881 households with an annual net income of "over £40,000" in the survey in 2004, an estimated 51% of which had an "up to 3" minutes walk to the nearest bus stop. The entry in the cell at the intersection of the "50%" row and the "1,000" column shows that the confidence limits for an estimate of 50% based on a sample of 1,000 cases are about +/- 3.7%-points. Similarly, an estimate of 50% based on a sample of 700 cases has confidence limits of +/- 4.4%. So, the confidence limits for an estimate of 51% based on a sample of 881 cases would be between the two values - say +/- 4.0% (as 881 is closer to 1,000 than 700). This means that there is a 1-in-20 chance that the estimate differs from the true value by more than about 4.0%-points, and roughly a 1-in-3 chance that the estimate differs from the true value by more than about 2.0%-points.
"95% confidence limits" for estimates of various percentages which are calculated from SHS sub-samples of various sizes
Estimate *(%) | Sub-sample size ( i.e. the " n =" value which corresponds to 100%) |
|---|
100 | 300 | 500 | 700 | 1,000 | 2,000 | 5,000 | 10,000 | 15,000 |
|---|
percentage points (plus or minus) |
|---|
5 or 95 | 5.1 | 3.0 | 2.3 | 1.9 | 1.6 | 1.1 | 0.7 | 0.5 | 0.4 |
|---|
10 or 90 | 7.1 | 4.1 | 3.2 | 2.7 | 2.2 | 1.6 | 1.0 | 0.7 | 0.6 |
|---|
15 or 85 | 8.4 | 4.8 | 3.8 | 3.2 | 2.7 | 1.9 | 1.2 | 0.8 | 0.7 |
|---|
20 or 80 | 9.4 | 5.4 | 4.2 | 3.6 | 3.0 | 2.1 | 1.3 | 0.9 | 0.8 |
|---|
25 or 75 | 10.2 | 5.9 | 4.6 | 3.8 | 3.2 | 2.3 | 1.4 | 1.0 | 0.8 |
|---|
30 or 70 | 10.8 | 6.2 | 4.8 | 4.1 | 3.4 | 2.4 | 1.5 | 1.1 | 0.9 |
|---|
35 or 65 | 11.2 | 6.5 | 5.0 | 4.2 | 3.5 | 2.5 | 1.6 | 1.1 | 0.9 |
|---|
40 or 60 | 11.5 | 6.7 | 5.2 | 4.4 | 3.6 | 2.6 | 1.6 | 1.2 | 0.9 |
|---|
45 or 55 | 11.7 | 6.8 | 5.2 | 4.4 | 3.7 | 2.6 | 1.7 | 1.2 | 1.0 |
|---|
50 | 11.8 | 6.8 | 5.3 | 4.4 | 3.7 | 2.6 | 1.7 | 1.2 | 1.0 |
|---|
* the confidence limits are the same for estimates of x% and for (100-x)%
The above information relates only to sampling variability. The survey's results could also be affected by non-contact / non-response bias: the characteristics of the people who should have been in the survey but who could not be contacted, or who refused to take part, could differ markedly from those of the people who were interviewed. If so, the SHS's results will not be representative. Comparison of SHS results with information from other sources suggests that any non-contact or non-response biases are not large overall - but they could, of course, be more significant for some sub-groups of the population or in certain Councils, particularly those with the highest non-response rates. In addition, as a survey of private households, the SHS does not cover some sections of the population - e.g. many students in halls of residence.
7.3 Further information about the SHS
Further information about the survey can be found on the SHSwebsite at http://www.scotland.gov.uk/shs
This website provides some background, information about the progress of the survey, and the published results, including Scotland's People - the SHS annual report, which provides many tables of SHS results, and other documents, which contain information about the survey methodology, the fieldwork outcomes, and an edited version of the questionnaire. Copies of these are available from Blackwell's Bookshop.
Three series of Transport Statistics bulletins, available from Blackwell's Bookshop and from the Scottish Executive web site, provide some Transport-related SHS results:
- Household Transport: some Scottish Household Survey results;
- Transport across Scotland: some Scottish Household Survey results for parts of Scotland; and
- Scottish Household Survey Travel Diary results.
Details of each of these are given in the section on Scottish Executive Transport Statistics publications.
SHS results also appear in other Executive publications, e.g.Scottish Transport Statistics.
General enquiries about the SHS should be addressed to:
Scottish Household Survey Project Team,
Area 1-F
Scottish Executive,
Victoria Quay,
Edinburgh,
EH6 6QQTel: 0131 244 8420
E-mail: shs@scotland.gsi.gov.uk
« Previous | Contents | Next »