Thursday, February 6, 2014

CIUS 2012

Question

I had a look at the files from the CIUS, and I am quite disappointed in the quality of the SPSS setup files supplied for this survey. There's many variables that haven't even been given the courtesy of a variable name, let alone value labels. I also had a quick look at the SAS setup and it looks to be the same, and I'm going to go on the assumption that it holds true for Stata.

Here's a list:
Household file:
G_URBRUR
G_FAMTYPE
G_HHSIZE
G_HEDUC
G_STUD
HA_02G
G_HQUINT
WTHP

Person file:
PUMFID_P
G_URBRUR
GCAGEGR6
CSEX
G_CEDUC
G_CSTUD
G_CLFSST
GFAMTYPE
G_HHSIZE
G_HEDUC
G_HSTUD
CU_G07
CU_11G
CU_12G
HA_02G
G_HQUINT
WTPP

I'm not sure why variable names and value labels haven't been supplied as this has never been the case with Statcan setup files before.

Answer

Regarding CIUS 2012, we have been supplied with an updated codebook and syntax cards as indicated to the list yesterday. For more information, the author division advised that:

"A review of the codebooks’ metadata indicated that some concept fields (descriptions) were possibly truncated due to the limited characters allowance. Accordingly, this would also impact the value label in the syntax cards.

We are able to provide updated codebooks that address the variable labels for CU_G07 and G_HQUINT. As for the variable G_CLFSST, the description provided is based on the groupings of answer categories to protect the confidentiality of the information. Thus, “able to work but not in the labour force” is the correct assumption."

The updated versions of the syntaxes for the Canadian Internet Use Survey (CIUS), 2012 can be found on the EFT site:

EFT: /MAD_DLI/Root/other-products/Canadian Internet Use Survey - cius/2012/doc/cius_ecui_2012sas-spss-statav2.zip