Appendix B:
Survey Methodology

View the most recent version.

Archived Content

Information identified as archived is provided for reference, research or recordkeeping purposes. It is not subject to the Government of Canada Web Standards and has not been altered or updated since it was archived. Please "contact us" to request a format other than those available.

Archived

This page has been archived on the Web.

1.0 Data Quality

The intention of the Survey of Older Workers was to augment information currently collected by the Labour Force Survey (LFS) as it pertained to older workers in Canada. As a live LFS supplement,¹ the target population consisted of persons between the ages of 50 thru 75 who were currently in the labour force or who had performed paid labour in the previous 2 years. The Survey of Older Workers (SOW) was designed to assess labour market intentions and transitions of older Canadians. The aim of the survey was to identify factors that influence one's decision to retire or remain working. In this context, pensions, general finances, the role of dependents, the nature of work, health considerations etc., were of primary concern in trying to understand workers intentions and motivations.

Collection was non-proxy and completed using Computer Assisted Telephone Interviews (CATI) only. One respondent was randomly selected per household. To ensure sufficient sample for the projected data needs the survey used 5 rotation groups of the LFS. The survey was conducted during the months of October and November 2008 following completion of the LFS.² The survey was conducted across Canada in all ten provinces but excluded the northern Territories, people living on Native reserves or military installations and Canadians who were residents of institutions (penal, palliative or medical) during the period of collection. Collection activities for the 2008 Survey of Older Workers officially ended on December 1st, 2008. In all, we have calculated a national response rate of 79.5 percent.

The response rate is based on the total number of complete and partial responses divided by the total sample sent into the field, minus the out-of-scope sample (e.g. vacant or destroyed households or households that don't contain a respondent matching the sample or sub-sample criteria). In Canada, the total number of valid responses received was 10,034. The total size of the sample sent into the field was 12,629, of which only 11 have been considered out-of-scope. Thus, the response rate is 10,034/ (12,629-11)*100 or 79.5 percent.

Implications of using the LFS

As a supplement to the Labour Force Survey, the Survey of Older Workers was able to obtain the majority of demographic and labour force information directly from the LFS. The LFS also provided advantages in terms of reducing information items on the SOW and having eliminated the concept of a "cold call". While certain household information for the LFS might be collected using the "household reference person" the SOW itself was non-proxy.

1.1 Response Rates

Across Canada, the total number of respondents was 10,034 valid responses. The total size of the sample sent into the field was 12,629. Thus, the response rate is 10,034/ (12,629-11)*100 or 79.5 percent.

The following tables summarize the number of in-scope persons, number of respondents and resulting response rates to the Survey of Older Workers (SOW).

Table B.1 Provincial response rates

Response rates across the provinces were reasonably consistent. Saskatchewan achieved the highest response rate at 83.7 percent, wile the lowest response rates were observed in British Columbia (73%). Residence in an urban or a rural part of Canada made minimal difference to the response rate of the survey. Respondents residing in urban areas obtained a 79.8 percent response rate while those located in rural communities responded at a rate of 78.9 percent.

The total sample was made up of 81% English speaking respondents, 18% French speaking respondents and 1% percent of the sample spoke another first language. While response rates did not vary much between English (80%) and French (82%) speaking respondents, only 15% of the sample who spoke a language other than French or English responded to the questionnaire.

---

Notes:

1. A "live" supplement refers to the methodology wherein when possible, the questions making up the Survey of Older Workers were asked immediately following the LFS interview.
2. In instances where the selected respondent was not available, a call-back was attempted during the week following the LFS.