Publication - Research and analysis

Time use survey 2014-2015: results for Scotland

Published: 5 Mar 2019

Analysis of the Scottish results of the 2014-2015 time use survey by Centre for Time Use, Oxford University.

Contents
Time use survey 2014-2015: results for Scotland
5. Methodology

5. Methodology

This report uses data from the 2000-01 and 2014-15 Time Use studies. The 2000-01 time use survey was funded by the UK Government and undertaken by the IPSOS-RSl.[7] The 2014-15 survey was commissioned by the Centre for Time use Research (CTUR) at Oxford University, with fieldwork carried out by NatCen and the Northern Ireland Statistics and Research Agency (NISRA). Both surveys followed the Harmonised European Time Use Survey (HETUS) guidelines,[8] ensuring their compatibility across time.

Data Collection

Fieldwork for the UK 2014-15 Time Use Survey, was carried out between April 2014 and December 2015. The 2014-15 survey used a multi-stage stratified probability (random) sampling design. The small users Postcode Address File (PAF) was employed as the sampling frame for households in England, Wales and Scotland and the Land Property Services Agency (LPSA) national list of domestic properties was used in Northern Ireland.

Demographic data was collected by interviewers, who initiated contact with the households and explained how to use the time use diary. Within a participating household, each respondent was asked to complete the same diary days, including one weekday and one weekend day. Respondents wrote in what they were doing at ten minute intervals during the day, using their own words, and were asked to include as much detail as possible.[9] These diary entries were assigned codes as part of the analysis process. Diaries were then collected in person by interviewers, or posted back by respondents if this was not possible.

In the full UK study, 11,860 sampled households resulted in 4,238 household interviews with a total of 10,208 eligible respondents. The achieved sample for Scotland specifically was 799 individuals and a total of 1,052 diary days, comprising 554 diary days for women and 498 for men. The total number of diary days collected in the rest of the UK was 13,662, made up of 7,395 women and 6,267 men.

More information about the sampling strategy and coding process can be found in NatCen’s technical report concerning the research project.[10]

Participants and Data

Within the Scottish sample, the distribution of participants amongst the age groups was as follows:

Age

16-24

100

25-44

308

45-64

342

65+

302

The proportion of the sample that reported a disability or long-term health condition in the Scottish sample was as follows:

Presence of disability / long-term health condition

Has disability / long-term health condition

444

No disability / long-term health condition

606

Disability status was determined on the basis of the following question:

“[Do/Does] [you/Name] have any health problems or disabilities that [you/he/she] [expect/expects] will last for more than one year?”

The proportion of respondents answering this in the affirmative was approximately 42% of the sample. For context, the 2017 Scottish Health Survey found that, in Scotland, 45% of adults aged 16 and older reported that they had a long-term health condition.[11]

The distribution of annual incomes in the achieved Scottish sample was as follows:

Income

Bottom 25%

226

Middle 50%

420

Top 25%

160

Missing

246

For a full description of the process by which the data was weighted, please refer to the Technical Guidance provided by NatCen.[12]

This Report and Interpreting the Results

For this report, the Scottish Government asked the CTUR to analyse the Scottish component of the time use data for 15 combined variables. These combined variables were composed from a selection of the over 300 activities identified in the full time use study. A full list of the activities contained in each of the 15 combined variables can be found in the data tables accompanying the report.

The findings presented here relate to the average number of minutes per day spent on an activity. This is worked out as (the average time per day for all people divided by the proportion of people who participated in the activity) multiplied by 100. The amount of time spent per day on an activity might be lower than expected when compared to a hypothetical person’s activity. For example, in Scotland, an average of around 2.5 hours is spent on paid work per day. However, this must be understood as the amount of time spent across the whole Scottish sample, not just the proportion of the Scottish sample participating in paid work on a given day.

The number of people who, based on the survey results, could be expected to be participating in a given activity on a randomly selected day is the daily participation rate. In the example of paid work, the daily participation rate is 35%. In analysing the results of the survey, we report on the average time spent on an activity by those participating where this appears useful and to further clarification. Full data on the participation rates for each activity amongst each group can be found in the data tables accompanying this report.

Establishing Significance

In relation to bivariate variables such as gender, disability and trends over time, statistical significance was established on the basis of comparative t-tests. In most cases, a significant result on t-tests coincided with non-overlapping confidence intervals between the variables, but this was not always the case. In analysing gender differences, the Scottish findings are analysed in the report alongside the results for the rest of the UK with which they are largely, although not exclusively, congruent.

In all other cases with multiple variables, significance was established on the basis of non-overlapping confidence intervals between the averages. This was the case regarding age groups and income groups, as well as comparisons analysing gender across age groups and gender alongside disability status.

In cases of multiple groups, significance is analysed between the groups in question by one group with another, rather than analysing the variance within the group as a whole. So, for example, comparing the top 25% of the income distribution to the bottom 25% of the income distribution, rather than a multi-variate comparison between all three groups.


Contact

Email: Claire McHarrie