We have a new website go to gov.scot

SHS 2017 Consultation

Consultation Analysis and Scottish Government Response

The Scottish Government published the analysis of the SHS consultation findings, alongside the Scottish Government’s response to the consultation, on Friday 19th August.

Scottish Household Survey (SHS) 2017 and Beyond Consultation

As part of Scotland’s Spending Plans and Budget for 2016/17, the Scottish Government requires to make savings on SHS 2017. Savings of the level sought cannot be achieved without significant changes to the design of the SHS.

The Scottish Government is seeking local authorities (LAs) and other users’ views on two alternative options for how the savings can be achieved:

Option A          Biennial topics, i.e. halving the number of topics covered by the survey each year and collecting data on each topic every second year, with a small reduction in sample size (from 10700 to 10100). Maintains full topic coverage, albeit every two years.

Option B          Reduction of the overall survey sample size by around a third, from 10700 to 7450, with a small reduction in topics covered by the survey.

The consultation can be accessed at and responded to via this link https://consult.scotland.gov.uk/scottish-household-survey/consultation-on-2017-survey. Alternatively responses to the questions in Annex B and the Respondent Information Form in Annex D should be returned by email to shs@gov.scot . The closing date is Tuesday 19th April 2016.

Respondents to the consultation may also like to look at the short paper on SHS 2017 Options A and B -  sample sizes and confidence intervals for national and local authority level estimates.

This note provides some illustrative confidence intervals for SHS 2017 Options A and B, and explains how confidence intervals can be calculated for different questions. Annex A of the paper provides sample (base) sizes for all local authorities under the current SHS and options A and B.

The paper includes the reference to a Significance spreadsheet for looking up confidence intervals for different sample sizes.