On this page:

Scotland's People: Results from the 2001/2002 Scottish Household Survey (Volume 8: Technical Report)

« Previous | Contents | Next »

Listen

Scotland's People: results from the 2001/2002 Scottish Household Survey
Volume 8: Technical Report

5. Confidence intervals and complex standard errors

Data collected in surveys are always an estimate of the true proportions in the population. The accuracy of these estimates - the sampling error - can be calculated for any estimate in the survey using information about the proportion of people giving the response and the number of people in the sample (or sub-sample). The sampling error can be expressed as a 'confidence interval', which can be added to and subtracted from the survey estimate to give a range within which it is fairly certain that the true value lies.

Since the survey is not a simple random design, the confidence intervals need to take account of the impact of clustering and stratification. The SHS, therefore, has what is known as a 'complex standard error'. While for some variables the design of the sample improves the precision of the survey estimates compared with a simple random sample, the overall effect of the survey design is to reduce the precision of the estimates. The relationship between the complex standard error and the theoretical simple random sample standard error for a sample of the same size is summarised in the 'design factor'.

Design factors were calculated for a range of measures using a 'jack-knifing' technique. As can be seen from Table 5-1 below, these ranged from 0.73 to 1.84. The overall average is 1.15, but that should not be taken as a 'typical' value, given the distribution of values across different variables. However, it suggests that the original assumption of a design effect of 1.1-1.2 was reasonable and using a value of 1.2 as a 'rule of thumb' for adjusting the standard errors of the survey data would account for the design factors associated with most variables in the survey.

The 95% confidence intervals shown are based on complex standard errors.

Table 5-1 Design factors and confidence intervals for key variables in 2001/2002 data

Characteristics
Estimate
95% Confidence Intervals
Theoretical simple random sample error for the same size of sample
SHS Complex Standard Error
Design Factor

Lower

Upper

Tenure

Owner-occupied

64.0

63.5

64.5

0.27

0.28

1.02

Social-rented Sector

28.4

27.8

29.0

0.26

0.29

1.12

Privately rented

6.0

5.7

6.3

0.14

0.16

1.19

Below bedroom standard

3.2

3.0

3.4

0.10

0.11

1.10

Property type

Detached house

19.8

19.3

20.3

0.23

0.26

1.13

Semi-detached house

21.9

21.0

22.8

0.24

0.44

1.84

Terraced house

22.2

21.5

22.9

0.24

0.38

1.59

Flat/maisonette

35.8

35.0

36.6

0.27

0.40

1.47

Economic status of working age adults

Full time employee

48.0

47.2

48.8

0.36

0.39

1.09

Part time employee

13.5

13.2

13.8

0.24

0.18

0.73

Self-employed

5.8

5.5

6.1

0.17

0.17

1.02

Unemployed

4.6

4.2

5.0

0.15

0.18

1.22

HiH or partner has a bank/ building society account

87.1

86.7

87.5

0.19

0.20

1.02

Marital status of all adults

Married/cohabiting

49.1

48.6

49.6

0.19

0.23

1.25

Separated/divorced

5.6

5.4

5.8

0.09

0.09

1.02

Single/never married

38.3

37.9

38.7

0.18

0.23

1.24

Widowed

7.0

6.8

7.2

0.10

0.11

1.18

Access to the internet

34.2

33.6

34.8

0.28

0.29

1.02

Travel to work in a car

63.3

62.5

64.1

0.43

0.40

0.93

Require regular care or help

12.0

11.6

12.4

0.19

0.22

1.20

Reporting long-standing illness, disability or health problem

31.0

30.6

31.4

0.26

0.22

0.84

« Previous | Contents | Next »

Page updated: Friday, March 31, 2006