Tuesday, November 13, 2018

Agegr10 - GSS 28: Victimization

Question:
I received the following question from a researcher below about Aboriginal identify in the General Social Survey, Cycle 28: Victimization.

I have a question regarding their measure of Aboriginal identity.  In the answering category, there are 6884 valid skip.  My understanding is that the contrast is between Aboriginal v. Non-Aboriginal.  I am reviewing a paper which claims that it is a contrast between Aboriginal and White Canadians.

Question: Am I right that it is a contrast between Aboriginals v. NonAboriginals?

Question: NonAboriginals includes all other Canadians, both visible minorities and non-visible minorities, right?

What is the category of “valid skip”?

Answer:
We’ve received the following response from subject matter:

“For questions AMB_Q01 – Are you an Aboriginal person, that is, First Nations, Métis or Inuk (Inuit)? First Nations includes Status and Non-Status Indians.

This question was asked of respondents who were born in Canada (DV_CNTRYCODE_B01
= 11124), the United States (DV_CNTRYCODE_B01 = 11840), Germany (DV_CNTRYCODE_B01
= 21276) or Greenland (DV_CNTRYCODE_B01 = 11304) , if they did not meet this criteria they are not asked this question and are therefore a valid skip (they wouldn’t get asked this question)

For AMB_Q002

This question was asked of respondents who were born in Canada (DV_CNTRYCODE_B01
= 11124), the United States (DV_CNTRYCODE_B01 = 11840), Germany (DV_CNTRYCODE_B01
= 21276) or Greenland (DV_CNTRYCODE_B01 = 11304) and who responded "yes" to
the question ""Are you and Aboriginal person, that is, First Nations, Métis or Inuk (Inuit)?
(AMB_Q01 = 1).

if they did not meet this criteria they are not asked this question and are therefore a valid skip (they wouldn’t get asked this question).

I hope this helps answer what ‘valid skips’ mean.”

Follow-up Questions:
I have a follow up question for the same survey, GSS, Cycle 28: Victimization:

“When I used age (agegr10) subtract age mean, and then square the results (see the following equation), there are suddenly a lot of missing data for the new variable.

Compute agesquare = sqrt (agegr10 – 4.04).

How do I avoid this?”

Follow-up Answer:
Subject matter has replied with the following:

“Agegr10 is an age group variable with categories 1 through 7.  Use age instead.”

Follow-up Question:
Can we get further clarification and explanation from subject matter? It appears there is no other variable age. Agegr10 is the only variable for age.

Answer:
Subject matter has responded with the following:

“If this is the PUMF, the client will not be able to use Age as only age groups are available due to confidentiality.

They would need access to the analytical dataset to have access to the Age variable.  Or they would need a custom request which would be done on a cost recovery basis.”