3.15 Permission to release personal census information after 92 years

In June 2005, an amendment to the Statistics Act received Royal Assent.  This amendment permitted the transfer of personal census records 92 years after the taking of the census to Library and Archives Canada (LAC).  For censuses up to and including the 2001 Census, explicit consent is not required.  Starting with the 2006 Census explicit consent is required.  Only those personal records where explicit consent has been received will be transferred to LAC.  This information will serve future generations by providing them with historical and genealogical data.  Nationally, 55.6% of respondents replied yes, they agreed to make their 2006 Census information available for public release in 2098 (92 years after the census); the remainder either said no or left the box blank.  For those who did not give permission, their personal information will not be transferred.

(a) Question wording

The question wording was changed in the alternate questionnaire.  The differences are reflected in Figures 3.52 and 3.53.

Figure 3.52  Public in 92 years in the control questionnaire

Description

53 The Statistics Act guarantees the confidentiality of your census information. Only if you mark "YES" to this question will your personal information be made public, 92 years after the 2011 Census. If you mark "NO" or leave the answer blank, your personal information will never be made publicly available.

Does this person agree to make his/her 2011 Census information available for public release in 2103 (92 years after the census)?

Ο Yes
Ο No

__________

Ο Yes
Ο No

Figure 3.52

Source: Statistics Canada, 2008 Census Test, 2B.1 questionnaire.

Figure 3.53 Public in 92 years in the alternate questionnaire

Description

58  Only if you mark "YES" to this question will your census responses and family history be part of the historical record of Canada. A "yes" means your census responses will be available to family members and historical researchers, 92 years after the 2011 Census, in 2103.

If you mark "NO" or leave the answer blank, your census responses will never be made available to future generations.

Does this person agree to make his or her 2011 Census information available in 2103 (92 years after the census)?

Ο Yes Ο No   |   Ο Yes Ο No

Figure 3.53

Source: Statistics Canada, 2008 Census Test, 2B.2 questionnaire.

Although the question was the same on both the 2A and 2B alternate questionnaires, the instruction in the banner was slightly different.  Figure 3.54 shows the instructions used in the alternate questionnaires.

Figure 3.54 Public in 92 years — instructions in the alternate questionnaires

Description

2A.2
Note: If you are answering on behalf of other people, please consult each person.

2B.2
The following questions are for all persons who usually live here including those less than 15 years old. If you are answering on behalf of other people, please consult each person.

Figure 3.54

Source: Statistics Canada, 2008 Census Test, 2A.2 and 2B.2 questionnaires.

The banner was placed before the first permission question (Elections Canada) on both the 2A and 2B alternate questionnaires.  The message on the 2B.2 questionnaire reminded respondents to include their children, as the majority of the long-form questions related only to persons 15 years of age and over.

(b) Question placement

In the 2006 Census, the 'Public in 92 years' question was placed on the inside pages of the 2A and 2B questionnaires.

In the 2008 Census Test, the question retained its position on the inside pages of the control and alternate 2B questionnaires, but only on the control version of the 2A questionnaire (2A.1).  The question was placed at the back of the alternate 2A questionnaire (2A.2) after the Elections Canada question because the space was needed for Question 6 (relationship to Person 1).  The question placement at the back combined with its size also affected the format of the question.  As shown in Figure 3.55, responses were placed over two rows.  This is different from the preceding question which listed all responses vertically.

Figure 3.55 Public in 92 years (alternate questionnaire)

Description

9  Only if you mark "YES" to this question will your census responses and family history be part of the historical record of Canada. A "yes" means your census responses will be available to family members and historical researchers, 92 years after the 2011 Census, in 2103.

If you mark "No" or leave the answer blank, your census responses information will never be made available to future generations.

Does this person agree to make his/her 2011 Census information available in 2103 (92 years after the census)?

Person 1
Ο Yes
Ο No

Person 2
Ο Yes
Ο No

Person 3
Ο Yes
Ο No

Person 4
Ο Yes
Ο No

Person 5
Ο Yes
Ο No

Person 6
Ο Yes
Ο No

Figure 3.55

Source: Statistics Canada, 2008 Census Test, 2A.2 questionnaire.

Following are some of the key findings and observations for the 'Public in 92 years' question.

(c) Weighted response rates by panel

  • 92.4% for 2A1 and 82.7% for 2A2
  • 80.6% for 2B1 and 83.5% for 2B2
  • 85.5% for 3B1 and 89.1% for 3B2
  • The weighted response rates for 2A and 3B panels were statistically different between the control and alternate versions of the question, but this was not the case for the 2B panels.
  • Response rates were higher for Internet respondents than paper respondents.
  • The distribution of the responses shows higher rates of 'Yes' responses for the alternate version of the question.

(d) Question placement and format

  • Response rates to this question were the same for the control and alternate versions of the 2B questionnaires, whether the response was obtained by Internet or paper.
  • There was no difference among the 2A control and alternate versions for Internet respondents.  For paper returns, the weighted response rates were 91% for the 2A.1 and 79% for the 2A.2.  The location and format on the paper questionnaire may have contributed to the lower response rate for the 2A.2.  The question, located on the back of the questionnaire, may have been missed by respondents.  The format of the response categories split over two rows, as seen in Figure 3.55 above, compared to the Elections Canada which listed the household members vertically, may have caused respondents to enter the response for Person 2 in the space for Person 4 (directly below Person 1) which would have been removed during clean-up if the household size was two or three.
  • For persons 14 years old and under, a large rate of non-response (close to 50%) was observed for both the control and alternate versions of the 2B paper questionnaire.

(e) Wording of Public in 92 years, following the Elections Canada question

  • The alternate panels always had a higher rate of 'Yes' among respondents to the question (the difference in distribution among the control and alternate questions is statistically significant).  This suggests that the wording and the placement of the Elections Canada question before it (perhaps putting the question in context), had a positive impact on obtaining a 'Yes' response from those who responded.
  • Among those persons linked to 2006, those in the alternate panel had more changes of 'No' (2006) to 'Yes' (2008) and fewer changes of 'Yes' (2006) to 'No' (2008) than those in the control panel.

(f) Impact on Step F (questions about the dwelling)

  • The control versions performed worse for Step F than the alternate versions, therefore there is no reason to find that the alternate version had a negative impact.

Previous page | Results of analysis | Next page