Monday, July 4, 2016

CCHS 2012 - Mental Health variable, Household Income


My researcher initially asked if a 15-year-old could have been responsible for answering questions about household income, but I'm wondering, more broadly, if someone who is under 18 or even someone who is not considered one of the heads of the household could be responsible for answering household questions.

It started with this question, "I am using the Canadian Community Health Survey - Mental Health (2012; CCHS-MH, 2012) data set, and I have a question. The age of respondents begins at 15, and goes to 90 and above. I note in the user guide, it says no proxy interviews were permitted. I am wondering, for the income variables ( ex. total income in the household), were 15 year olds responsible for providing this?"

After some back-and-forth, it turns out that she's interested in people aged 15-29, and wondered why she had so many of them seem to be in a higher-than-expected income group (using household income). I replied that although she's able to pull out the age group in question doesn't mean they were living alone (i.e., every member of the household influences household income) and I have suggested she might want to consider the individual income variable. I'm waiting to hear from her to see if there is any follow-up to that suggestion.

So barring anything unexpected, I think I've actually found a solution for her research (not that hard, I admit), but a general question remains unanswered in my mind: is there an age-cutoff for answering the household questions? I looked at the questionnaire and the data dictionary, and see that when someone starts answering the questions they are rejected if they are under 15. I also see that in section 5.4 of the User's Guide, "The interview," there is a statement indicating that "a knowledgeable household member was asked to supply basic demographic information on all residents of the dwelling. One member of the household aged 15 years or older was then selected for a more in-depth interview, which is referred to as the C2 interview." In 5.4 e) Proxy interviews, the UG says that no proxy interview were permitted. This, I believe, is where my researcher's confusion stems.

Can someone confirm for me that the income questions would have been asked of the person most knowledgeable of the household? I feel like a dork for even asking, since it seems pretty obvious (I've read the Income Imputation section, btw), but I'd rather be safe than sorry.


This pertains to CCHS Mental Health, so this response does not apply to the Annual.

As stated in the User Guide, the beginning of the survey is done by a knowledgeable person in the household. The interviewer is instructed to ask for a knowledgeable person when they first make contact with the household. This person completes the household roster so we get the most accurate information on name, age, sex, & marital status (the “basic demographic information”) of all household members. Then we do the selection of the survey respondent. The respondent does the rest of the survey themselves and income is part of that. Therefore, income is asked to 15-year-old respondents. If they don’t feel qualified to answer the question, they do have the option to say “Don’t know” or “Refuse”, as with any other question on the survey. Income was imputed when there was a non-response.