Wednesday, January 22, 2020

Variable issues with GSS Cycle 30

Question:

A researcher reports something strange with two variables from the 2016 iteration (Cycle 30- Canadians at Work and Home) of the GSS.

Both variables AGEFHSHG (Age of respondent’s father in household) and AGEMHSDG (Age of respondent’s mother in household) show answer categories that are very perplexing if you consider that these are supposed to be PARENTS:

0 to 9 years and
10 to 19 years

Granted, there are young parents that would be in the second category, but the first one does not make sense at all and I wonder if the second one is meant to start as early as 10 years old. And there are actually answers within the 0 to 9 years old category! This must be a mistake or there is something really strange about how these derived variables are obtained.

Answer:

Thank you for bringing this discrepancy to our attention. After some investigation, we discovered the following two issues: 

1.       There was 1 case where all foster children were inadvertently coded as foster parents. This has been recoded.

2.       Labels in the PUMF were incorrectly applied for “Age of respondent’s mother in the household” and “Age of respondent’s father in the household”.

The way it appeared in PUMF--incorrect
Age of respondent’s mother in household

0 to 9 years
10 to 19 years
20 to 29 years
30 to 39 years
40 to 49 years
50 years and over

Valid skip


To rerun the data yourself, please use the following coding:

 AGEMHSDG in groupings of 10 years */

/* Age group of respondent’s mother in household */

/****************************************************************************************/

/* Values for : */

/* 1 : Less than 40 */

/* 2 : 40 - 49 */

/* 3 : 50 - 59 */

/* 4 : 60 - 69 */

/* 5 : 70 - 79 */

/* 6 : 80 years and over */

/* 96 : VALID SKIP No Mother in household */

/* 97 : DON'T KNOW */

/* 98 : REFUSAL */

/* 99 : NOT STATED */

Our Processing Team is making all appropriate corrections and is rerunning the data. The data dictionary will be reissued in the near future.

.....

...the changes apply to both the mother and father variables.


Once we do the corrections on our end, the folks at Odesi will be able to take the data to restage for their purposes.