Tuesday, December 19, 2006

Participation and Activity Limitation Survey Question Re: Children


A UBC hospital researcher is planning to use PALS, but her interest is in the children surveyed, not the adult population. This presents a problem because the PUMF in the DLI collection does not include children 5-14. To quote the User's Guide:

For purposes of the PUMF, analysts based themselves on the 28,908 persons aged 15 and over who responded to the survey.

The researcher has not been able to contact the person identified in the release in The Daily or that person's successor.

She would really like to know whether there will be a PUMF for the children's file. She has been in contact with the local Research Data Centre and will work there if necessary, but wants an answer as to the eventual availability of a children's PUMF.


There is no PUMF for the PALS children's file.

Some results are available in "A profile of disability in Canada, 2001", catalogue number 89-577-XIE. The Profile of disability among children is available at the following link:

More than likely, the researcher may need additional information not covered in the previous profile. The researcher's options are:

1) Access the PALS file via the RDCs
2) Order custom tabulations

Index to Senate Debates


Does the Parliamentary website have an index to the Senate debates? If so, where is it tucked away?


Here is the link to it from the DSP:

Amount of unpaid work at retirement


We have a student looking for how many years of unpaid work that people have when they retire and what the reason for this unpaid work is.


Two possibilities:

1) We measure the amount of unpaid work through the Census.
2) The GSS Time Use cycle has some goodies in terms of unpaid work.

These are cross-sectional data though and not cumulative over a period of a lifetime. I am not familiar with any source of data that provides the number of unpaid work years people have when they retire.

Thursday, December 14, 2006

Attitudes toward homosexuality


A couple of students here have worked with U.S. General Social Survey data to look at attitudes toward homosexuality. There are a few variables in that survey that get at respondents' attitudes:


They are hoping to continue this work using Canadian data, but I haven't been able to find any like variables in our GSS cycles or other surveys available in the DLI -- there are surveys dealing with sex and with whether sexual orientation is a factor in victimization, but I can't find any that measure the attitudes of the respondents themselves. Does anyone know if Statistics Canada collects, or has collected, this kind of information? The students want to relate attitudes to religion and ethnicity.


1) The World Values Survey has questions that get at this, and are comparable across a number of countries. The 2000 wave includes questions:

v76 People the respondent would not like to have as neighbours: homosexuals
v208 Behaviour ever justifiable: homosexuality

The data is available from http://www.worldvaluessurvey.org/, as well as ICPSR.

2) There's also the Charter of Rights and Freedoms data collected by CRIC. These deal with gay marriage, equality rights and
discrimination. In addition, there are some Globe and Mail polls that might be helpful, also from CRIC.

Queen's has a huge number of public opinion polls in their CORA archive. Go to:

3) Access to the first three waves of the World values surveys (1981-1995) is available unrestricted, via an SDA interface at:

Limited online analysis of all 4 waves is available at the WVS web site:

It looks as if you can obtain all the data there as well.

Wednesday, December 13, 2006

Updated Products - PSES / CIES / AWP

Public Service Employee Opinion Survey - 2005

Data is available on the opinions of employees in their work environment, job satisfaction, career movement, equipment needs and special needs. Tabulations are available at the department level or at the Public Service level.

FTP: /dli/pses/2005
WEB: http://www.statcan.ca/english/Dli/Data/Ftp/pses/pses2005.htm

Changes in Employment Survey - Cohorts 1 to 10 (CIES)

The primary objective of the Changes in Employment Survey (CIES) is to evaluate the impact of Bill C-12 on the Employment Insurance legislation and the degree to which program objectives have been achieved. Bill C-12 was introduced into legislation in part in July 1996, with the remainder coming into effect in January 1997.

The legislation was designed to better reward work effort, to ensure adequate benefits by targeting those most in need, to encourage job creation, and to improve the perception of fairness. Specific aspects of these objectives were addressed in the survey. In addition, the survey attempts to get a measure of the aggregate impact of the legislation.

Secondary objectives of the survey include the continuation of the information collected in the 1993 and 1995 Canadian Out-of-Employment Panel Surveys. This includes collection of background demographics on the individual and the household, as well as information on job search activities and outcomes, assets and debts, expenditures, and utilization of Employment Insurance and Social Assistance.

Microdata files for the Changes in Employment Survey (CIES) are now available for five reference periods. They are: Cohorts 1 and 2: January 1995 to September 1996; Cohorts 3 and 4: July 1995 to February 1997; Cohorts 5 and 6: January 1996 to September 1997; Cohorts 7 and 8: July 1996 to February 1998; and Cohorts 9 and 10: January 1997 to September 1998.

FTP: /dli/cies
WEB: http://www.statcan.ca/english/Dli/Data/Ftp/cies/cies-cohorts1-10.htm

Survey of Annual Work Patterns 1978 - 1985

FTP: /dli/awp
WEB: http://www.statcan.ca/english/Dli/Data/Ftp/awp.htm

Monday, December 11, 2006

Income Variables in CCHS 2.1


There's a curious difference between two income variables in CCHS 2.1; according to the codebook, for the total household income

INCCGHH=1 represents 'NO INCOME AND LESS THAN $15,000',

whereas for the personal income variable

INCCGPER=1 represents 'NO INCOME' and
INCCGPER=2 represents 'LESS THAN $15,000'

A student currently using the PUMF was wondering why one had the two categories collapsed into one, but the other, not.


The reason provided by the division was indeed for confidentiality reasons.

Friday, December 8, 2006

Seafood Consumption by Ethnic Group


Can someone help me to find data on fish and/or seafood consumption (in Canada) by ethnic groups?


The Family food expenditure surveys 1992 and 1996 - summary files don't have much, but they do have:
- country of birth (more like continent of birth) and mother tongue (English, French, and other)
- weekly food expenditure on fish and other marine products

Unfortunately, the country of birth variable is no longer included in the 2001 pumf.

There is also the pdf file:
Influence of Immigration on The Ethnic Food Market in Canada / Agriculture and Agri-Food Canada in the November 2005 edition of Canada food stats (23F0001XCB)

I checked the Canada health survey (1978-79) and the Nutrition Canada survey (1970-1972) without success - but then, these are quite old by now and there have been substantial changes to the ethnic composition of Canada since then.

You may need to send the user the RDC-route, since the CCHS 2.2 nutrition component should include other foods than just the fruits and vegetables that are in the pumf, as well as ethnicity variables in the common and optional content file.

Monday, December 4, 2006

Questions about the General Social Survey 2005, main file

The weight variable for variable EPI677 is not stated - Would it be WGHT_PER?


The Coverage is not stated for variables TCS_Q180 to TCS_Q200, VCG_Q300 - In all cases would it be all respondents?


Is a weight variable supposed to be associated with variable INCMMEMC? None is defined. WGHT_PER is used for INCM, but WGHT_HSD might be used - which weight variable (if any) should be assigned?


What is definition of WGHT_CSP and WGHT_SNT?

WGHT_CSP: This is the weighting factor for analysis at the person level created using the sample of persons asked the questions in Section 10A - culture, sports participation and physical activity. For example, to estimate the number of persons who used library services as a leisure activity in the last 12 months, WGHT_CSP should be summed over all records with this characteristic. This weight is zero for respondents who were not asked this section i.e. completed Sections 10B and 11. - Used for participation in Culture activities.

WGHT_SNT: This is the weighting factor for analysis at the person level created using the sample of persons asked the questions in Sections 10B and 11. To estimate the number of persons with a particular characteristic, WGHT_SNT should be summed over all records with this characteristic. This weight is zero for respondents who completed Section 10A. - Used for participation in Sport activities.