Statistical Bulletin: Transport Series: Trn / 2013/ 2: Scottish Household Survey: Travel Diary 2012

2. Introduction to the Scottish Household Survey and the Travel Diary

2.1 The Scottish Household Survey (SHS) is a continuous survey based on a sample of the general population in private residences in Scotland and has run since 1999. As part of the survey a randomly-chosen adult in the household is asked about their travel on the previous day. This component is known as the Travel Diary and is the basis of the information presented in this publication.

2.2 Each adult in the household had an equal chance of selection. The results have been weighted to take into account other differences in selection probabilities. Un-weighted sample numbers are given in italics.

2.3 Results are subject to sampling variability and care should be taken when interpreting year on year changes - confidence limits for the estimates are presented in Table 17 and should be considered alongside the data itself. A detailed background of the SHS methodology is given in the Appendix.

2.4 Table 10 contains statistics which underpin Scotland's National Indicator on congestion. More information can be found on the Scotland Performs website.

Methodological Changes - 2007 onwards

2.5 A number of changes were made to the Travel Diary from 2007 onwards, as follows:

  • Journeys less than ¼ mile or shorter than 5 minutes on foot are now reported - reduces under reporting of short (likely to be) walking journeys;
  • Introduction of improved weighting system accounting for non-response bias.
  • ¾ of the main SHS are now asked the Travel Diary element - previously all the households had been asked.
  • Improvements to wording of questions - reduces under reporting of journeys.
  • Improvements to data quality - e.g. duration of journey asked instead of calculated.

2.6 The main impact was a higher proportion of short, mainly walking, journeys with mode, distance and duration tables being most affected. For this reason, most time series comparisons mainly focus on those from 2007 onwards. Care should be taken when making time series comparisons prior to 2007. Further details of these changes are discussed in the Appendix.

Additional changes - 2012 onwards

2.7 From January 2012 a new SHS went in to the field which had a substantially restructured sample design and integrated the previous Scottish House Conditions Survey. The new survey uses a fully un-clustered core and modular structure with some questions being asked of the full sample and others of a one-third sub-sample. The overall sample size has reduced from around 14,000 household interviews to about 11,000 though improvements in efficiency of the survey design mean it will be possible to attain local authority estimates on an annual basis where analysis permits. While the overall sample size of the survey has reduced, the survey design improvements has meant that the precision of estimates have not been affected significantly. More detail regarding the changes to the survey methodology can be found on the SHS pages of the Scottish Government website.

2.8 Improvements were also made to the structure of the Travel Diary section of the interview in 2012, particularly around return journeys. These changes simplified the survey for respondents and have improved the quality of the origin and destination data collected. The new methodology has resulted in more walking journeys being recorded and more journeys recorded as 'go home'.

2.9 'Go home' is recorded as the purpose where no direct return to home is made, for example if someone goes to work, then to meet friends, then home. Previously the journey home would have been allocated either to commute or meet friends.

Distance calculation

2.10 Travel Diary distances are calculated using the straight-line method (see Appendix) and are therefore likely to be underestimated. A piece of work was undertaken to investigate the extent of the underreporting of Travel Diary distance1. A comparison between distances calculated by the straight-line method and distances calculated using GIS software confirmed that the distances reported through Travel Diary are underestimates of actual distance travelled.

2.11 From 2012, the Travel Diary data includes additional distance estimates incorporating GIS information for distance travelled by road (where mode of transport makes measurement appropriate). Distance tables in this publication use straight line distance. Further analysis will be published later in the year comparing straight line and road distances.