As a longitudinal study, consistency is key. However, from time to time we implement changes to make improvements to the way we collect data and the resources we provide for data users. What is new this time?
Changes to the questionnaire
As a longitudinal study, Understanding Society routinely collects the same information from households at regular intervals, so that change (or stability) over time can be tracked. The Study has to balance the needs of researchers with the burden on survey participants, which is why we don’t include all questions in all waves of data collection. Alongside the questionnaire content that is included in each wave (“annual content”), there are questions that are asked less often (“rotating content”).
Previously rotating content questions were on 2-, 3-, 4- or 6-year rotations. Recently, to streamline survey length some modules were moved to a three-year rotation and a few to a six-year rotation pattern. Additionally, the long annual history module (where questions about changes since the last interview are included) was split into six separate annual histories to make it easier for survey participants and data users:
- Annual Health Conditions History
- Annual Partnership History
- Annual Residential History
- Annual Education History
- Annual Employment History
- Annual Fertility History
Our long term content plan outlines what’s being covered in each wave, and changes to the questionnaire since the last wave is available here.
Questions about families and households
The concept of families and households has been changing, and we wanted to make sure we were collecting data that reflects different types of families and households. For example, individuals who consider themselves to be part of a household may not be living together. Parents may not live with their children all of the time but still have close contact with each other and play a very important part in their lives. So, questions on parents living apart (PLA), and non-resident children were reviewed, new questions suggested and tested. To improve the information that we collect about families and households, some changes were introduced in Wave 14 and some further were made in Wave 15. Read this Understanding Society Working Paper that provides a detailed description of this exercise. In Wave 15,
- We added new content and revised existing questions on separated families, non-resident children and parents living apart, identifying new pregnancies and household leavers.
- We asked for the names and details of all children ever born to the respondent included in a new child summary, to check that we were including children within and outside the household.
- We collected contact information about significant others parents living elsewhere and partners.
Families with children at key ages were given longer interviews in this wave, but for everyone else the questionnaire length remained the same. An additional £5 thank you voucher was offered to these families.
Environmental modules
We also reviewed our questions about environmental behaviour and attitudes. You can read more about the review and the new content in the Working Paper Review of environmental attitudes and behaviour questions in the Understanding Society survey. After a review, it was decided to drop the environmental attitudes module and introduce a new environmental identity module based on European Social Survey questions.
Changes to the data
Employment data
As data users are aware, following a consultation with stakeholders from Wave 13 onwards, we started asking about all current jobs, rather than just the main and second current job, to reflect the changes in work patterns today. As a result, some respondents reported up to 10 jobs. Note even though multiple jobs could be reported, respondents were asked which was their main job and information about this job was recorded as main job variables to maintain longitudinal consistency. So, users only interested in the main job information can continue to use the same variables across all waves. You can read more about these changes here.
As was the case for the main job, each subsequent job mentioned was followed by multiple questions about that job. Until now, these additional variables were included in the main respondent data file, w_indresp, resulting in a large number of variables, which was set to inapplicable for a majority of cases. To make the datafile easier to use, it was decided that starting from the Wave 15 data release, these additional variables will be recorded in a long format file, w_employment, each row uniquely representing a job for a respondent. Note, w_indresp will continue to include information about the main job.
Weights
Some changes were made to the weights provided as well. Why did we make these changes?
With the inclusion of Wave 15 data, for the first time longitudinal analysis can be conducted on the sample including the Wave 14 boost sample. So, a new series of longitudinal weights was started, w_inding2_lw.
As around 90% of interviews are now conducted on the web, which is by design entirely self-completion, there is little need for self-completion weights which adjust for the non-response to the self-completion part of the questionnaire among face-to-face respondents. Similarly, the number of proxy respondents has dropped to very low numbers, reducing the need for separate proxy weights (for analysis using direct and proxy respondents). As a result from onwards Wave 15, self-completion and proxy weights will not be produced.
Other changes
As always, you can read about regular changes and improvements made to previous releases of the data here. For example, two datafiles were renamed with suffix 2, w_nonresch2 and w_natchild2, to reflect small changes to the wording of the questions in those modules.
For those interested in geography data, a new user guide is available.
Get started with the data!
Each release of a new wave of data is a milestone for Understanding Society! Now that it’s out with you we hope that Wave 15 adds to your understanding of life in the UK and we look forward to seeing the new research that comes from it.
Please remember to cite all the datasets you have used and let us know when you’ve used Understanding Society in your research. If you would like to share your syntax for producing new datasets or variables using our data that may be useful for other data users, please let us know.
And as always, if you need help with using the dataset, our user support forum is always available.
Authors
Alita Nandi
Alita Nandi is Senior Research Fellow and Associate Director for Understanding Society
