Expert Panel on Access to Historical Census Records

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.

Census Expert panel

Expert Panel on Access to Historical Census Records

1. Introduction

2. Issues and Considerations

3. Recommendations: Ways and Means for Achieving the Release of Historical Census Records

Annex A: List of Expert Panel Meetings

Annex B: Detailed Opinion Poll Questions and Results

1. Introduction

1.1 Mandate and Work of the Panel

1.1.1 Mandate

On November 5, 1999, the Honourable John Manley, Minister Responsible for Statistics Canada, appointed an Expert Panel on Access to Historical Census Records to report regarding the legal, privacy and archival implications of providing access to historical census records. The Panel was asked to examine the following issues:

1.1.2 Scope

The Panel was asked to consider the public release of historical census records. It did not consider the public release of other survey and administrative data records. It has dealt with the release of records for all census periods including the future.

1.1.3 Members

The Panel was chaired by Dr. Richard Van Loon, President of University, and members included: the Honourable Lorna Marsden, President and Vice-Chancellor of York University, Professor Chad Gaffield, University of Ottawa, Professor John McCamus, Osgoode Hall Law School and the Honourable Gérard La Forest, retired Supreme Court judge.

1.1.4 Work of the Panel

The Panel, unlike a Commission of Inquiry or a Parliamentary or Senate Committee, did not hold public meetings; rather Canadians were invited to write to the Panel regarding their views on access to historical census records. As well, the Panel asked certain interested groups to explicitly express their views on the matter.

The Expert Panel held five working meetings during which time a number of sessions were held with specific persons or groups (see Annex A). Two telephone conferences were held with provincial genealogical associations. The Panel Chair and Dr. Gaffield also had the opportunity to meet with the Chief Statistician of Australia who made time available during his scheduled visit to Ottawa in February 2000.

Over 2,500 letters, faxes and e-mails were sent to the Minister responsible for Statistics Canada, the Chief Statistician and numerous Members of Parliament prior to the announcement of the Expert Panel. These were made available to the Panel members. During the seven month duration of the Panel, a further 1,055 letters were received from Canadians active in genealogical societies. The Panel received 95 submissions from historical, archival and genealogical associations as well as other interested Canadians.

As part of its deliberations, the Panel members also considered the draft legislation and motions before the Senate and the House of Commons together with related speeches and documentation.

Further, the Panel considered the public opinion research commissioned by Statistics Canada. The firm Environics conducted six focus groups (two in Ottawa, two in Toronto and two in Montreal). Nine questions were asked on an opinion poll conducted by Environics in March 2000 and seven were asked in a follow-up survey conducted at the end of April. Panel members observed the focus groups. The questions asked in the first survey were developed by Panel members and the staff of Environics. The questions asked in the second survey were developed by Environics and Statistics Canada.

All of the material received and considered by the Panel is available for consultation at the Statistics Canada Library and Information Centre, R.H. Coats Building, Ottawa. As well, a limited number of submissions have been posted on the Statistics Canada Web site. Please note that federal government official languages policy prevents the posting of unilingual materials on government maintained Web sites and, for that reason, only a limited number of submissions were posted.

The Panel would like to thank all those individuals and organizations who sent written submissions, participated in the teleconferences and attended their meetings. The interest, dedication and efforts of many Canadians were of considerable assistance to the Panel.

1.2 Organization of the Report

In the first section, we frame the major issues as we see them and consider the positions of the major proponents for and against release. The report then turns to a consideration of the history of census in Canada, the international context, the undertakings regarding confidentiality and release, the nature of the questions asked of respondents, current public concerns and views. After some consideration of the legal situation, the report considers the important practical issues of how the principles we enunciate can be put into practice.

2. Issues and Considerations

2.1 The Frame of Analysis

The Panel was created because there is a difference of views between those who believe that individual census records should routinely be made available, after a time, to the public and those who believe such records should never be released. Advocates of the former view include archivists, historians, genealogists and others interested in historical research. Advocates of the latter include Statistics Canada, the National Statistics Council and the Privacy Commissioner.

Those who advocate perpetual confidentiality are concerned that any releases of individual census records after the 1901 Census will require retroactive changes to legislation and that this will be considered by many people to constitute the breaking of a promise made by the federal government to Canadians. They fear that any such action will lead to doubts being cast on the confidentiality of all statistical data collected or held by Statistics Canada and will result in a decline in the response rate for future surveys and censuses. Since these data are essential to the management of the economy and to the operation of government, they assert that this risk far outweighs any benefits to be gained from release. They also assert that the moral and legal right to privacy of individual information accrues in perpetuity.

Those who advocate release believe that the value of the historical information to be gained by release outweighs other considerations. They believe that no guarantee of perpetual confidentiality was ever made, and they cite both historical precedents in Canada and the practices of Britain and the United States as supporting the view that the release of individual historical census records is not problematical and will not be of concern to the public.

2.1.1 Statistics Canada and Data Collection in Canada

The primary role of Statistics Canada is to establish and maintain a national statistical system for Canada. This objective is achieved by collecting, compiling and publishing information about the economic, social, cultural and demographic conditions of Canadians. As a statistical agency, Statistics Canada has a world-renowned reputation for the accuracy and quality of its statistical output.

Statistics Canada undertakes its work by conducting large sample surveys, by accessing administrative data sources and, of course, through the census. The importance of the census in fulfilling governance requirements is matched in importance by Statistics Canada’s surveys in areas such as business, labour force, population health or household income.

Given the importance of the work of Statistics Canada for Canadian society and government, it is essential not to harm or damage the Agency’s ability to undertake the successful collection of census or other data collection activities. The taking of a census can be viewed as an intrusive activity. The trade-off for requiring compulsory participation in the census has been a guarantee of confidentiality. Quite aside from any moral or ethical concerns in this regard, then, any change to this balance which might lead to a lower response rate, increased collection costs or deterioration in the quality of response would be viewed with intense alarm by Statistics Canada and its data users.

2.1.2 Users of Historical Census Records

Historians and genealogists are the major users of historical census records. Historians note that the formation of households and families, the organization of work among family members, the socio-economic and spatial mobility of ordinary Canadians and the growth and decline of rural and urban areas are all essential parts of the fabric of Canadian history. The census, as a unique source of information on all Canadians and on the patterns of Canadian settlement, is highly valuable in understanding our past and historians assert that only through access to individual census records can they properly conduct their research.

The large group of Canadians who are interested in genealogy requires access to individual census records to determine family lineage and change over time. The census is a particularly important source as it reveals the names, ages and characteristics of all of the family and household members. It provides unique information on the characteristics of the individual such as place of birth and immigration status, education and economic status. Only by tracing the life histories of individual family members is it possible to explore the lineage of family units in Canada.

Historians and genealogists demonstrated a need for access to the individual level information and neither group could identify another data source that would meet their research and information requirements. Both groups stressed that historical census records must be accessible and available to everyone, not just academics and scholars. This was a point equally emphasized by members of the archival community.

2.2 The Census in Canada: Historical Perspective

Canada has had a long history of census-taking. Jean Talon is considered to have held the first modern census in New France in 1666. Pre-confederation censuses were held also in British North America. Since Confederation, a national census of the population has been held every ten years until 1956, when Canada’s national census began to be taken every five years. During the period beginning in 1906 to 1946, mid-decade censuses were held only in western Canada.

Table 1 summarizes the general nature of all post-Confederation Canadian censuses.

The major objective of a census has always been to obtain a highly accurate count of the number, characteristics and distribution of a population living permanently within pre-determined borders. A census provides a snapshot of the country and its population for a given point in time. This is done in order to support legal requirements such as the establishment of electoral boundaries or the distribution of transfer payments. For these reasons, it is vitally important that a census counts virtually all residents at one point in time, and a near-perfect response rate is thus essential.

Some guarantees of confidentiality of data have accompanied every census for which we have detailed information but all individual census records up to and including the 1901 Census have been publicly released in Canada.

Today’s census is Canada's largest peacetime operation, involving, at its peak, over 38,000 enumerators. It requires about eight years from the question concept development stage to the final release of information and with respect to the 1996 Census, the total cost was in excess of $347.5 million.

2.3 The International Context

The practices and legislation governing the release of historical census records differ among countries holding censuses. The practices in some countries normally considered benchmarks for Canada are summarized in Table 2.

Summary of International Situation Regarding the Release of Historical Census Records

The two countries most commonly used as a benchmark for Canada, Britain and the United States, routinely release historical census records. In France, release is very limited and other countries, not in the table, show a mixture of practices. Australia, on the other hand, destroyed census records for the periods between 1905 and 1996, but has recently decided to hold and then release the 2001 Census records after 99 years. The release will be subject to respondent consent.

In both Great Britain and the United States, the wording of laws governing census confidentiality is quite similar to that in our Statistics Act. Their provisions, like ours, contain no mention of time limits, forbid access to records to anyone but census employees, limit use of data to aggregate statistical purposes and prohibit release of data that could result in individual identification.

The public release of census records in both the U.S. and Britain is actually governed by legislation relating to their national archives. This legislation is assumed to apply after the records are transferred to those repositories.

The U.S. legislation with regard to confidentiality of data in the hands of census officials is set out in Section 9 and 214 of Title 13 of the United States Code. The census records are transferred to the National Archives in a manner similar to that which is mooted in the Canadian Instructions to the Census Commissioners and Enumerators for the period up to and including the 1946 Census and is effected pursuant to an exchange of letters between the heads of the U.S. Bureau of the Census and the National Archives. The release of the American historical census records after 72 years is then conducted by the Archives in accordance with its procedures and is not contentious.

Similarly British legislation, in the Census Confidentiality Act 1991 (Schedule 12) provides guarantees of confidentiality and penalties for breach of confidentiality while the data are in the hands of the census officials and is silent on the matter of length of time. Census records are transferred to the Archives with the understanding that they are to be impounded for 100 years. Release of British census materials after 100 years through the National Archives and subject to the legislation governing that institution is then routine and uncontroversial.

The Panel’s view is that the international precedents in the two countries most often used as benchmarks for Canada are consistent with the claim that the guarantees provided to Canadians with respect to confidentiality were not intended to be perpetual, that the records would eventually be deposited with the National Archives and that, once there, the National Archives Act would apply.

2.4 Canadian Undertakings Regarding Confidentiality

But what of the specific undertakings made to Canadians regarding confidentiality? Since 1871, all census enumerators have taken the oath of secrecy and have been instructed to keep the results of the census confidential. Review of the enumerator instruction manuals used in census-taking from 1871 to the most recent 1996 Census Guide reveals a continuity of concern by Statistics Canada to assure respondents that their personal census information would be revealed only to those doing census work. The secrecy requirements appear to have been a response to a continuing fear and suspicion among respondents that their census information could be used against them personally.

For example, the 1871 Census Manual of Instructions for Commissioners and Enumerators stated that: "A census is not taken for purposes of taxation, as, unfortunately, many persons imagine." Therefore, enumerators were told that: "Persons having apprehensions, or showing hesitation in giving their answers, must be assured that no information they may give, and that nothing taken down in the schedules, can possibly injure, or in any way affect their standing or their business." Similarly in the 1911 Census, enumerators were told to give positive answers and assurance in the course of their work: "If a fear is entertained by any person that they may be used for taxation or any other object."

The 1996 Census questionnaire tells all respondents: "Your personal census information cannot be given to anyone outside Statistics Canada - not the police, not another government, not another person."

It is impossible, of course, to know what was communicated to individual respondents regarding confidentiality by enumerators over time and across the country, particularly in the early years of the census. But it is striking that all of the instructions and all of the examples used refer to contemporary concerns such as taxation, immigration status or business data. Words like "perpetual" or "eternal" or "for ever" were used neither in the legislation nor in the more colloquial instructions to enumerators and are never found in the debates surrounding the 1905 and 1906 Census and Statistics Act and the 1918 Statistics Act.

While the assurances given to respondents have been similar over time, the legislative framework in which they have been set has evolved very considerably. The most significant change came with the passage in 1918 of the Statistics Act. It sets out explicit guarantees in legislation as follows:

Thus while it remained the case after 1918 that what was said ‘at the door’ may have varied from individual to individual, the Parliament of Canada set out much clearer and broader public assurances in law than had previously been the case.

2.5 Historical Census Records Transfer, Storage and Use

At the same time as respondents were given assurances of confidentiality, there is also evidence of the intent to preserve this information for the use of future generations. The same instructions to the enumerators which dealt with confidentiality also contained reference to the returns being kept in the archives of the Dominion and used for future reference. For example, in 1911 enumerators were enjoined to "write legibly" since the records would eventually be transferred to and stored at the Dominion Archives. This intention was evident in all of the instructions up to the 1946 Census. And, since the mandate of the Dominion (later National) Archives has always been the storage of information for some present or future use, it is reasonable to assume that the eventual release of individual record data stored there has generally been contemplated. Certainly, by 1942, the first release of historical census information had occurred as the 1871 Census records were made available to the public by the National Archives and release continued, somewhat irregularly, up until the release of the 1901 records in 1993.

2.6 The Questions

The nature of questions asked on Canada's census have remained broadly similar over time. For example, the early censuses asked for the name of every household member, the address of the residence, the sex, date of birth, marital status, birthplace, citizenship, year of immigration, religion, number of children born to married women, language spoken, level of schooling, occupation, place of work, income and sources of income for all household members. In addition, questions were asked about the dwelling, such as type of tenure and costs. Many of these same questions are still asked today.

There were, occasionally, questions asked in earlier censuses which might be viewed as highly sensitive today. For instance in 1911, a question about infirmities included the categories "crazy or lunatic" and "idiotic or silly". But similar questions were asked in the 1901 Census and release of these census records has elicited no complaints.

Since 1971, the short-form questionnaire, answered by all respondents, has covered only "tombstone" information such as name and date, sex, relationship to persons in the household and marital status. The long-form questionnaire, answered by only a sample of households, asks detailed socio-economic questions including income and shelter costs. In fact, then, in some respects, the response burden and overall intrusiveness was greater in the earlier censuses. Moreover, during these early censuses, the answers were given directly to an enumerator who was, in most cases, known to the respondent. While respondents today might know the enumerator who leaves the form at their residence, this is much less likely to occur as the majority of Canadians live in major urban centres and Statistics Canada makes every effort to assign census workers to neighbourhood areas located at a distance from where they live.

In the Panel’s view, most of the information collected by the census is not of a highly sensitive nature. Moreover, that information which might be sensitive for some respondents, such as income data, is likely to lose its sensitivity over time and reaction, or lack of it, to release of previous censuses seems to confirm this conclusion.

2.7 Current Public Concerns

The Environics opinion poll conducted in March 2000 suggests, even in the current privacy-sensitive environment, the majority of those asked supported the release of their personal census information 100 years after the census. And, when participants in the focus groups were advised of the nature of census information, they generally appeared to think that disclosure of such information after the passage of a lengthy period of time was not problematic. Some participants, indeed, were pleased by the thought that their descendants would take such an interest in the facts of their ancestors’ lives. Due to the nature of the information collected and given the passage of time, the privacy interest of respondents does not appear to be a particularly intense one.

Most Canadians surveyed in the poll consider that Statistics Canada keeps their personal information confidential although over a third do not. Moreover, a change to legislation which would permit the release of their own personal census information about 100 years hence would not seem to change the way that the majority of Canadians would answer the next census or participate in Statistics Canada surveys (see Annex B).

However, the situation is less clear with regard to any legislation that might be necessary to release census records that were subject to earlier guarantees of confidentiality. The Opinion Poll demonstrates that, while 68 percent of Canadians would agree with such legislation, 30 percent would disapprove and in total 21 percent would do so strongly (see Annex B, Table 4).

The focus groups were also instructive in this regard. The question used in the first national survey did not emphasize that one could construe legislative change to release historical census records as the government "breaking a promise" but all the focus groups, more or less quickly and on their own, came to construe such legislation in that way. Once they did, they became very hostile to the idea. It was not clear that this would affect their response to future censuses but it is safe to say that they were displeased at the concept.

In recognition of this, Statistics Canada commissioned a second national survey in which the notion of "retroactivity" was explicitly introduced. Table 5 (Annex B) indicates that a question focussing on a retroactive amendment is greeted with less approval compared with one asking only about changing the law, though even, in this second survey, over half of the respondents would approve of such a change. Even so, the notion of retroactive legislation is of considerable concern to the Panel, as it is to Statistics Canada.

Overall, however, the Panel does not think it likely that the eventual release of census records will materially affect census participation. Our survey data suggest it is not a problem, and we subscribe to the view of the U.S. Congressional Committee that investigated the problem. With regard to the fear that such release would damage participation the Committee noted:

This impression is supported by the historical record in Canada. Early census records began to be released by the National Archives when the 1871 Census was made available in 1942. The 1881 Census records were opened to the public in 1979 and following the passage of the Privacy Act, the 1891 and 1901 Censuses were made public 92 years after each census was taken. Meanwhile, all pre-Confederation censuses in Newfoundland were released shortly after Newfoundland joined Canada. The aftermath of the Privacy Act suggests that the government believed it could and should release census records, on a regular basis, after 92 years. Certainly, for all censuses undertaken prior to the changed legal situation created by the 1906 Census and Statistics Act and the 1918 Statistics Act, the government was entitled to do so simply by the understandings surrounding the Privacy Act and without further legislative action.

There have been no complaints whatsoever to the Government of Canada or Newfoundland about the release of this information. In fact, when the United States and British releases are included, there has now been census records released regarding over 620 million individual records in the three countries and there has not been a single complaint made to the National Archives, census offices or privacy commissioners of any one of them.

Overall, the Panel concludes that access to census records after the death of the respondent or after the passage of a long period of time is not likely to be or to have been in conflict with the actual expectations of past respondents or at odds with the views of the present ones.

It seems to the Panel that where the government can release census records without recourse to any legislation which could be construed as the retroactive removal of a guarantee of confidentiality, there is little danger of an inflamed public opinion that would affect future censuses. But if legislative change was necessary and particularly if it were an amendment to the Statistics Act, there may be some risk to the census, and, quite possibly, to the other activities of Statistics Canada.

2.8 The 92-Year Rule

We have seen that most countries releasing historical census records keep the personal census records confidential for a specific period of time, usually ranging from 70 to 100 years, after which time the records are released to the public. While the time period for the release of historical census documents in Canada varied up until the passage of the Privacy Act, the period of confidentiality has since been established to be 92 years and is enshrined in the Privacy Act.

It is not at all clear why 92 years was originally chosen to be the interval between the taking of a census and its public release. In practical terms, it would appear that 92 years permitted the immediate release of the 1891 Census records once the Privacy Act was passed in 1983.

It has been argued that the release of the historical census information, even after 92 years, violates the fundamental principles of personal privacy because respondents were not explicitly informed that their information would eventually be released to the public. The basic tenets of privacy legislation mean that Canadians have the right to know why personal information is being collected, how it will be used, how long it will be kept and who will have access to it. Uses of information for purposes other than for which it was collected should be subject to consent.

This perspective is at odds with the view that the passage of time or death of an individual diminishes the privacy rights of that person, a view that is also enshrined in the Privacy Act.

The Panel has carefully considered these competing views. Our conclusion is that the commitment to confidentiality of census data was not intended to last indefinitely. Our view is that the passage of 92 years is sufficient time to allay concerns regarding individual privacy.

2.9 The Legal Situation

Although the Panel contains two members of the legal profession it is not our intention to give legal advice to the government. However, we are persuaded that perpetual confidentiality was not likely either assumed or intended by lawmakers. This view is derived from the following considerations:

Public reaction (or lack of it) to release of earlier census records suggests that our view is shared by the public.

We have reviewed legal opinions provided to Statistics Canada and to the National Archives by the Department of Justice and we recognize that legal minds can differ regarding the legal standing of various assurances given respondents versus the indication that records were to be transferred to the Archives. However, while we find the legal situation ambiguous, we find no convincing evidence that Parliament intended to create perpetual confidentiality.

We have come to the view that the release of pre-1906 census records constitutes a particularly important precedent particularly when combined with the fact that release of the 1891 and 1901 census records occurred in concert with the 1983 Privacy Act. We further believe that the passage of time - 92 years in this case - is an important legal and moral consideration and that the release of census records after 92 years in no way violates the original intent of those who developed the census in Canada. We do recognize that the passage of the Statistics Act in 1918, with its encoded guarantees of confidentiality, adds an element of uncertainty as does the disappearance, after 1946, of the requirement that census records be transferred to the National Archives but this does not change our view of the spirit and intent.

2.10 Release of Future Census Records

It will not have escaped the reader’s attention that we have said little about the situation with respect to the release of the records for future censuses. We do not consider this to be a difficult matter. Provided that respondents are advised that individual census records will be released after a 92-year period we do not foresee a problem. Our public opinion data suggest the vast majority of Canadians are untroubled by this prospect and will not consider it an impediment to response. It is incumbent on Cabinet to keep this in mind when it approves questions for the census but we believe the addition of this consideration to Cabinet’s concerns is a proper exercise of the executive power.

2.11 Issues and Considerations: A Summary

In the view of the Panel, its recommendations must achieve a balance between the privacy concerns of individuals and the value of public access to census records, with due regard for the viability of future censuses and of the other activities of Statistics Canada.

It is the Panel’s view that the passage of time diminishes the concerns about individual privacy. Thus, in the matter of privacy concerns of individuals versus the value of public access, the Panel believes the value of public access after a sufficient period of time to be pre-eminent. We believe that 92 years is a sufficient elapse of time.

But the matter of the viability of future censuses and surveys is also a vital consideration. The Panel is concerned that public reaction to anything which could be construed as removing a guarantee made by the Government of Canada to its citizens, even 92 years ago, could create significant problems for Statistics Canada and its activities. While we believe this construction of the situation would be wrong, since no perpetual guarantees were intended, we do have to recognize that some could characterize legislative change this way.

The Panel therefore tempers its recommendation in favour of release with a call for careful consideration by government as to whether legislation is required and a thorough evaluation of the potential risks if it is. The likelihood that legislation could be required is greater for the release of data collected after passage of the Statistics Act in 1918 and the Panel has therefore been more cautious in its advice for that period.

3. Recommendations: Ways and Means for Achieving the Release of Historical Census Records

Our fundamental recommendation is simply that census records should be publicly released through the National Archives 92 years after a census is taken. The means by which the release of historical census records can be achieved varies with the historical period in which the census was and will be taken.

3.1 2001 and Subsequent Censuses

The Panel recommends that all Canadians be informed that the guarantee of confidentiality with regard to all future censuses endures for a period of 92 years after which time individual census records are to be released by the National Archives.

The Panel recognizes that any legislative change to enable release of personal census information is unlikely to be in place in time for the 2001 Census. The Panel also recognizes the long planning horizon required for undertaking any census and in no way wishes to jeopardize its success.

Equally, the Panel holds the view that Canadians should be informed at the time of the 2001 Census and for all subsequent ones that the guarantee of confidentiality is interpreted to endure for a period of 92 years after which time all individual records may be transferred to the National Archives for public release. It believes that an effective communication strategy must be mounted to ensure that Canadians are aware of this situation.

3.2 Consent

The Panel does not recommend that each individual’s or respondent’s consent be sought for future release of census records.

The Panel is not convinced that the provision of consent, for example, as proposed for the 2001 Australian Census, would achieve an adequate result in Canada. The notion of "group consent" whereby the individual who completes the household census form provides consent on behalf of all household members is not a form with which Canadians are familiar; nor is it founded in Canadian privacy law and practice.

Rather, the Panel recommends for the 2001 Census that Canada adopt the practice currently in place in the United Kingdom and the United States. For the 2001 Census and all others where a household questionnaire is used, respondents would be informed that their individual census information would be kept secret for 92 years and after this time, the information would be publicly released by the National Archives.

3.3 Achieving Release of Censuses Taken Prior to the 1918 Statistics Act: 1906-1911-1916 Censuses

The Panel recommends the immediate public release through the National Archives of the 1906 Census with a scheduled release of the 1911 Census in 2003 and the 1916 Census in 2008.

The Panel has carefully evaluated the moral, historical and legal arguments surrounding release of these records. Given the considerations set out in the preceding section of this report the Panel recommends that Statistics Canada immediately effect the transfer of the 1906 Census records to the National Archives for public release. The 1911 Census should be released to the National Archives in 2003 and the 1916 Census records in 2008.

The Panel is not persuaded that a legislative amendment to the 1906 Census and Statistics Act is necessary to effect the release of the remaining censuses undertaken prior to the enactment of the 1918 version of the legislation, particularly given the precedent of the release of earlier census records after passage of the Privacy Act in 1983. We believe that the National Archives Act governs the census records once it has been transferred and we believe the intent to transfer the census records to the Archives was clear.

Should there be legal doubt sufficient to cause the Government of Canada to believe legislation is necessary, the American and British precedent of clarification via the legislation governing the National Archives is considered to be the appropriate route but we wish to reiterate our view that any such clarification in no way would vitiate the spirit and intent of early censuses.

3.4 Censuses Taken Under the 1918 and Subsequent Statistics Acts

The Panel recommends that the government commit itself to the public release of census information arising from the censuses held from 1921 onwards and determine what, if any, legislative steps might be required to do so.

While the Panel believes the principle of release of census information required that public release should occur in a scheduled manner 92 years after the census is held, it also notes that the transport of confidentiality guarantees into legislation in 1918 may be interpreted as making it more difficult to do so for post-1918 census records. The Panel urges the government to take any legislative steps which may be required after the 1916 Census is released to the public in 2008 and with due regard to a more deliberate assessment of the potential impact on the activities of Statistics Canada than has been possible to date.

We recommend that, should any legislative change be required, it be to the National Archives Act and clearly refer only to the release of census records.

3.5 Summary

The Panel is firmly convinced of the benefits of the release of historical census records. The Panel is of the view that with the passage of time, the privacy implications of the release of the information diminishes and that the passage of 92 years is sufficient to deal with such concerns. We are persuaded that a guarantee of perpetual confidentiality was not intended to apply to the census. We believe that the indication of transfer to the National Archives also implied an intention that the census records would eventually become public and we would not view any legislation deemed necessary to do so as a breaking of a promise to respondents. We view the historical and international precedents as fully supportive of this position. The Panel is equally convinced of the value of the census and other work of Statistics Canada and is unwilling to make any recommendation which it believes will jeopardize this work. It is for that reason that we recommend release of the pre-1918 Census records and post-2001 records on a 92-year cycle, while advising some caution regarding any legislative steps that might be thought necessary to effect the release of those census records for the period 1921 to 2001.

Date	Submission
January 6, 2000	Statistics Canada Dr. Ivan Fellegi
February 9, 2000	Canadian Historical Association Dr. Irving Abella Dr. William Waiser Dr. Joanne Burgess Privacy Commissioner of Canada Mr. Bruce Phillips National Archivist of Canada Mr. Ian E. Wilson Telephone Conference Call Participants: British Columbia Genealogical Society Mr. Gordon Watts Mr. Bob Paulin Alberta Genealogical Society Ms. Carol Anderson Saskatchewan Genealogical Society Ms. Marge Thomas Manitoba Genealogical Society Dr. Geoffrey Lambert Ontario Genealogical Society Ms. Marjorea Roblin Genealogical Institute of The Maritimes Dr. Allen Robertson New Brunswick Genealogical Society Mr. Calvin Craig Genealogical Society of Nova Scotia Ms. Karen McKay P.E.I. Genealogical Society Inc. Mr. Richard Savidant Newfoundland & Labrador Genealogical Society Inc. Mr. Patrick Walsh NWT Genealogical Society Mr. Robert Carr
March 10, 2000	Telephone Conference Call Participants: Quebec Family History Society QFHS), Mr. G. Schroder Société généalogique canadienne-française Mr. Marcel Fournier La Société de généalogie de Québec Ms. Mariette Parent
April 7, 2000	Environics Dr. Donna Dasko
May 1, 2000	National Statistics Council Dr. Ivan Fellegi Dr. Lloyd Barber Dr. Derek Hum Dr. Arthur Kroeger

Annex B: Detailed Opinion Poll Questions and Results

Table 1: "Thinking about yourself, would you strongly agree, somewhat agree, somewhat disagree or strongly disagree with allowing your personal census information to be made available to historians, to those interested in family history and to the public about 100 years after a census is held?"

Table 2: "From now on, would you approve or disapprove of the government releasing your personal census information to the public after about 100 years?"

In the first survey, the question for each option was asked as follows: "If the government were to change the law so that your personal census information would be available about 100 years after a census, would be more likely/less likely/make no difference to your answering/giving accurate answers/participation/trusting Statistics Canada?"

In the second survey, the question for each option was asked as follows: "If the government were to change the law to allow the eventual release of personal census information that they had guaranteed would remain confidential, would be more likely/less likely/make no difference to your answering the next census/giving accurate answers/trusting Statistics Canada?"

Table 4: "The law guaranteeing the confidentiality of individual names and collected in the census has been in place for every census since the 1906 Census. Would you strongly approve, somewhat approve, somewhat disapprove or strongly disapprove of the government changing the law to allow the release of personal census information from past censuses after a delay of about 100 years?"

Table 5: "In order to release past personal census information, the government will have to change the law that guaranteed this confidentiality. Do you approve or disapprove of the government changing the law to permit personal census information that was collected with a guarantee of confidentiality to be released to the public after about 100 years?"

Archived Content

Census Expert panel