Project Canada 2005
CitationBibby, R. W. (2021, March 16). Project Canada 2005.
SummaryThe 2005 survey was the seventh in the Project Canada national adult surveys. It was carried out by mail between approximately July 15 and December 15 of 2005. Reginald W. Bibby was the principal investigator, assisted by Project Manager Reggie Gordon Bibby, Jr., Jim Savoy, and a number of student research assistants. The initial Project Canada 2005 sampling frame consisted of a list of 1,729 people who had participated in one or more of the six previous Project Canada adult surveys (1975 through 2000). Previous experiences suggested that as many as 50 percent of these individuals (about 900) would participate once again. To produce an eventual participating sample of at least 1,500 cases, this core was supplemented with an initial new sample of some 1,500 people and a projected 500 participants, resulting in a total 2005 sample of around 1,500 people. In addition, to facilitate comparisons with Alberta in the province's centennial year, Alberta was oversampled to bring its total to around 600 people - permitting generalizations within about +/- four points, 19 times in 20. This involved adding about 400 people beyond the 232 Albertans who had participated in Project Canada 2000 and selecting some 1200 additional people from the province. Our projections were reasonably accurate: the total number of 2005 participants consisted of 2,400 people, including 655 Albertans. With oversampling compressed, the Alberta sample is an appropriate 160 (10 percent), and the total national sample 1,600.
The ARDA has added three additional variables to the original data set to enhance the users' experience on our site.
Data FileCases: 2400
Weight Variable: WT, AWT
Alberta's sample of 655 people - weighted to 625 - is highly representative of Albertans with a similar level of accuracy in generalizing to the province. That weighted sample can be accessed by simply applying the weight 'AWT'.
Data CollectionJuly 15th - December 15th, 2005
Original Survey (Instrument)Project Canada 2005 Codebook
Funded ByThe funding for the survey was provided by the Louisville Institute (Lilly Endowment), along with Alberta Children's Services.
Collection ProceduresIn 1995, participants dating back to 1975 were pursued; this emphasis was repeated again in 2015 with the goal of including as many people as possible from the previous six surveys. That effort began in the spring of 2005 and continued throughout the project. It appears that some 1,300 of the 'Project Canada 2000' respondents were located (about 75 percent).
Follow-up procedures were the same as used in 1980 through 2000. Approximately two to three weeks after the initial mailing, the pre-paid post-card-functioning as a reminder and asking about the status of the questionnaire-was sent, followed about three weeks later by a second copy of the questionnaire, with 'Second Request' stamped on its cover letter. In the fall of 2000, participants were sent a thank-you post-card, informing them how the results were being disseminated, and welcoming further inquiries.
As with the earlier Project Canada surveys, the 2005 questionnaire was constructed to provide comprehensive data pertaining to three main subject areas being investigated: social issues, intergroup relations, and religion. In order to monitor change, many of the items earlier six were included in the questionnaire, with new items added to clarify the earlier findings as well as update the content more generally.
Sampling ProceduresBased on previous survey experiences, a projected required supplemental sample of about 1,500 cases was drawn in the spring which, when combined with the core from 2000, would produce a final sample of adequate size and sufficient representativeness to permit high accuracy generalizations to the Canadian population (typically, about 25 percent cannot be located). As before, Canada was stratified by province and community size (greater than 100,000, 99-10,000, less than 10,000), and the overall sample drawn proportionate to the national population, with 355 communities involved. Potential participants were randomly selected using telephone directories. As in the past, it was anticipated that weighting for provincial and community size, as well as gender would be required, with age weighting required since the ongoing cores would have aged anywhere from five to twenty-five years - and also to correct for the growing underrepresentation of people under the age of 35. As just noted, a special oversampling of Albertans was undertaken, with the goal of producing some 625 participants from that province.
A total of 692 people from 2000 and earlier (40 percent of those 1,729 participants) submitted usable questionnaires, as did 1,708 new individuals, resulting in a 2005 total sample of 2,400. The overall return rate was about 60 percent. Alberta's total came to 655, including an oversample of 337 people.
In 1995, participants dating back to 1975 were pursued; this emphasis was repeated again in 2015 with the goal of including as many people as possible from the previous six surveys. That effort began in the spring of 2005 and continued throughout the project. It appears that some 1,300 of the 'Project Canada 2000' respondents were located (about 75 percent).
So weighted, the sample is highly representative of the Canadian population (see Table 1). A representative sample of this size should be accurate within about four percentage points on most items, 19 times in 20 similar surveys.