Support #1560

Project Query

Added by Casper Gill about 2 years ago. Updated almost 2 years ago.

Start date:
% Done:




Firstly, I would like to make it clear that my knowledge regarding Understand Society's data as well as STATA is limited, so I apologise in advance for asking what might be perceived as an obvious question.

I am currently conducting a project into the long-term care that stroke patients/survivors utilise. As both waves 7 and 9 collected observations that are relevant to this topic e.g. as they have variables that distinguish stroke patients as well as the formal, informal and social care that they use, they are consequently very much suited to investigating this issue.

Currently I am determining my wave 7 sample size from those understanding society wave 7 participants who have 'mentioned' having a stroke in the past year (since the last questionnaire) using variable hcond7. In total there are 109 patients. I however noticed that 2831 are proxies with values of -7. Is it safe to assume that these 2831 individuals have 'mentioned' having a stroke in a previous wave's questionnaire. Would there be any other reasons as to why they have proxy answers? I only ask as I was trying to find a way of incorporating previous stroke patients from other waves into waves 7 and 9, and a larger sample size would greatly help with my project.

Any feedback or opinions would be greatly appreciated.


Updated by Understanding Society User Support Team about 2 years ago

  • Status changed from New to Feedback
  • % Done changed from 0 to 80
  • Private changed from Yes to No


If a variable has a valud of -7 it means that this person did not complete the interview themselves, but someone else completed a shorter Proxy questionnaire on their behalf, and this question was not included in this short Proxy questionnaire. On a more general note, the negative values -21 -20 -11 -10 -9 -8 -7 -2 -1, reflect that this variable has no valid value, and the specific negative values shows the reason why this is missing.

For example, -7 means that the value is missing because the quesiton was not included in the Proxy questionnaire and this was an interview by proxy; -8 means the person was not asked the question because they did not eligible for the question; -2 means the person refused to answer and -1 that they didn't know. Here is more about missing values in the User Guide.

To get all the people who said they had a stroke by Wave 9: include anyone for whom hcond7=1 in any of the previous waves (this was in almost all previous waves), see here:

As the questions asking about health conditions has changed across the waves, please take a look at the FAQ.
Then look at the category "Questions about variables" and then look at the question "How were health questions asked"

Best wishes,
Understanding Society User Support Team


Updated by Understanding Society User Support Team almost 2 years ago

  • Status changed from Feedback to Resolved
  • % Done changed from 80 to 100

Also available in: Atom PDF