Scottish household survey 2016: key findings

Technical notes for interpretation

Interpretation of change

All changes highlighted in this report are statistically significant – which means they are very unlikely to have occurred by chance.

Survey design

The Scottish Household Survey has a high sample size (10,700 households, 9,800 adults) which allows annual Local Authority level data and sub-group analysis including a breakdown for some equalities groups.

The survey has a fully unclustered core and modular structure. It includes a "core" set of 20 questions which have been designed to be asked in consistent ways with other surveys, such as age and gender. "Modules" of questions have been designed to be flexible in terms of topic, frequency and geography. For example, questions asked of the "full" sample and asked on an "annual" basis would be able to provide local authority level data on an annual basis. Similarly, questions might only be asked of "1/3" of the sample on a "biennial" basis ( i.e. asked every second year). Such questions could only get national level estimates every second year.

Sampling

The SHS sample has been designed by the Scottish Government and allows the annual publication of results. Nationally representative samples of private households and of the adult population in private households are produced.

It is drawn from the small user file of the Postcode Address File ( PAF).

The sample design is coordinated with the other two Scottish Government population surveys.Samples of the three surveys can be pooled for further more detailed analysis. From January 2012, a number of core questions are asked the same way in all three Scottish Government cross-sectional surveys, although in a small number of specified instances the questions may differ somewhat but still provide harmonised outputs. Data from all three surveys for these questions is pooled, to provide better estimates at national and local level (Scottish Survey Core Questions). The target sample size for Scotland was 10,678 household interviews.

Response Rate

The SHS response rate declined from 67% in 2014 to 64% in 2015 and 2016.