2967.0 - Information Paper: Parliamentary Inquiry into the Treatment of Census Forms: Submission from the Australian Bureau of Statistics , 1997
ARCHIVED ISSUE Released at 11:30 AM (CANBERRA TIME) 18/07/1997 First Issue
Page tools:
 Print Page
 Print All
| ||||
OVERVIEW
"A disposal authority allowing the destruction of name-identified records for 1996 and beyond will be issued. If circumstances alter for subsequent Census records then these disposal arrangements can be reviewed". Therefore, the Committee would need to be convinced that significant changes in circumstances had occurred, before it recommended the practice of form destruction, adopted and endorsed by successive governments, be altered. ABS considers there have been no significant changes in circumstances which would warrant any change in the current policy of census form destruction. Indeed to the contrary, new evidence shows that public opposition to the retention of census forms is very much higher than expected. This evidence reinforces the judgement that that if census forms were kept the cooperation of the public, in terms of the level of response in the census and the quality of data provided, would suffer. Very serious implications follow for all users of census data and other official statistics, particularly for users of population estimates. Serious impacts would be likely on electoral redistribution's, on the distribution of funds to the States and Territories, and for the host of government and private sector organisations who use census data for a variety of purposes, including infrastructure planning, service delivery, and program evaluation. CURRENT ISSUE The current practice is for census forms to be destroyed once statistical processing is completed. For about 12-18 months following the census the data on the forms is electronically captured, edited and the original responses coded into statistical classifications (eg industry, occupation, qualifications, family type). Names and addresses are not kept or put onto the computer record. Once these processes are completed the forms are pulped and then recycled. The Australian Bureau of Statistics (ABS) retains a complete, electronic copy of the unidentified statistical information relating to each person responding to the census. This computer file is used to produce statistical tables and other products from the census. The computer file is kept permanently. HOW DID THE CURRENT PRACTICE START Prior to 1971, census forms were retained for a time until it was judged they would not be required for any further cross-tabulations which might appear worthwhile after the initial analysis. Only forms for the two previous censuses were retained at any one time. In 1971 the then Treasurer (Mr Snedden) ordered the destruction of all census forms held by the ABS. He also directed that the census forms for the forthcoming 1971 Census be destroyed as soon as the information contained in them had been transferred to magnetic tape. He reassured the public that names and addresses on the census forms would not be transferred to the tape. This action was prompted by privacy concerns. These had been stimulated by the newly formed Australia Party, which had attacked the requirement for names and addresses on census forms, and by earlier publicity about anti-census campaigns overseas. Since then, successive governments have consistently maintained the policy of destroying census forms in Ministerial Statements to Parliament announcing the conduct of the Census (for example 20 November 1979, 25 February 1985, 4 May 1989; and 20 September 1994). EFFECTS OF RETENTION OF CENSUS FORMS Privacy is a sensitive issue for many Australians The initial decision in 1971 that census forms be destroyed was driven by a public campaign on privacy issues led by a new political party, the Australia Party. The public debate on the Australia Card that took place in the 1980's shows how easily the population's attitudes can be inflamed over privacy concerns - the privacy climate can change very quickly. Market research commissioned by ABS before each census over the past 20 years identified privacy and confidentiality as key issues of public concern that needed to be addressed in the public relations campaign. There is significant public opposition to the retention of census forms Recent research commissioned by the ABS, and conducted by AGB McNair after the 1996 Census, indicates that there is significant public opposition to the retention of census forms. Key findings of the research are:
These findings are consistent with the qualitative evidence obtained from ABS consultations with the community about the census, feedback from census collectors and other empirical indicators of community attitudes to privacy and confidentiality, such as the increasing proportion of people and households opting to use privacy envelopes. THE ACCURACY OF CENSUS DATA WOULD BE ADVERSELY AFFECTED The research suggests that if census forms were retained ABS could expect a significant reduction in the level of public cooperation in the census. Non-response rates of the order of 10% or more are possible and, should civil liberty and privacy groups mount a substantial campaign, they could be even higher - it should be remembered that the AGB McNair research was conducted following the 1996 Census, a census which ran smoothly with only a low key privacy debate. A reduction in the quality of data provided by householders could also be expected. This would be disastrous for the census. A high level of non-response would significantly reduce the quality of all census data and, most importantly, the accuracy of the crucial State/Territory population estimates over the intercensal period. THE QUALITY OF OTHER ABS STATISTICS MAY ALSO BE ADVERSELY AFFECTED ABS is concerned that a change in the policy on census form retention could undermine confidence in the community, or in parts of the community, about ABS resolve to protect the confidentiality of other data that it collects. There is no doubt the high level of cooperation that ABS receives in all its statistical activities, which in turn underpins the high quality of Australian statistics, is based in no small part on the trust the respondents have in the ABS. The ABS has a very solid compact with its respondents: response rates are very high and respondents provide accurate data, and in turn the confidentiality of the data is completely protected. This level of cooperation is the envy of most overseas statistics agencies, and it should not be put at risk by any reduction in the public's confidence in the ABS. IMPACT ON ELECTORAL DISTRIBUTION In 1975, the High Court held that the Constitution required that the population of the various States needed to be ascertained during the life of each ordinary Parliament for the purpose of determining the number of members from each State in the House of Representatives. An opinion of the then Law Officers, Attorney-General (Mr Ellicott) and Solicitor-General (Mr Byers), on the High Court decision in February 1976 said: ".... it necessarily follows that the State's respective populations be reliably determined (emphasis added). For this some method of counting the population such as a periodical census is essential." The Law Officers' opinion led to the provisions in the Census and Statistics Act, 1905 which require 5-yearly censuses, and quarterly population estimates. On the question of accuracy of population estimates in the context of quinquennial censuses, the Law Officers said: "The temporal disparity between quinquennial censuses and triennial elections means that the statistical population estimates which are based on the antecedent census tend to become unreliable and thus to afford ground for a court to hold the number of each State's members in the Representatives is not in fact in proportion to its population." The accuracy of State population estimates relies in large measure on the accuracy of the census counts. If the retention of census forms led to poor quality population counts from the census, then the accuracy of the State/Territory population estimates would be certainly reduced, perhaps significantly. The Law Officers' opinion would suggest that a reduction in the accuracy of State/Territory population estimates could afford ground for the High Court to hold that the number of each State's members in the House of Representatives is not in proportion to its population, as required by the Constitution. The consequences of such a serious possibility would need to be taken into account in any consideration of census form retention. IMPACT ON THE DISTRIBUTION OF STATE GRANTS State/Territory population estimates are used in the distribution of about $21 billion each year in the Commonwealth Grants to the States. Current inaccuracies in those estimates already are a matter for concern amongst some of the States and Territories. Increased inaccuracy would be viewed very unfavourably. OTHE IMPORTANT USES OF CENSUS DATA WILL BE ADVERSELY AFFECTED There are many other important uses made of census data and the related State/Territory population estimates. They are used by many government agencies in the planning, delivery and evaluation of services to regions and to various population groups such as migrants, the aged, children, and Indigenous peoples. They are also used extensively by business and community groups for their planning and research. Privacy concerns relating to storage and use Perceptions about the capacity of computers to store and manipulate vast amounts of information, and fears about data matching, contribute to an increasing concern in the community about confidentiality and privacy. Fears about potential future mis-use of census information will fuel these concerns. As the former Privacy Commissioner, Mr O'Connor put it
The 'value' of keeping name-identified census records In the past no demand has been put to the ABS for the use of census forms for medical or historical research. ABS considers no substantive case has yet been made for the use of census forms for genealogical research which demonstrates substantial social and community benefit. ABS believes there are many other sources of information which will be accessible to the public in 100 years, which should satisfy family history research needs. Can retention of census forms be 'sold' in terms of special interest for the centenary of Federation? Obviously all the privacy issues raised above apply to the 2001 Census. Therefore, the basic question still has to be addressed for the 2001 Census. Two additional points arise. Does the research value of the 2001 Census forms increase due to the fact that the census will coincide with the centenary of Federation? On the cost side, can a special case be made for the government to meet the costs of retaining those census forms in terms of the special interest for the centenary of Federation? The AGB McNair study showed that 89% of the Australian public agrees that "census forms should be destroyed to protect people's privacy and confidentiality", and 73% disagreed that "researchers should be given access to census forms including names and addresses". In order to run the 2001 Census of a similar quality to previous censuses the ABS would have to mount a successful campaign to convince about three-quarters of Australians to change their mind, based on the importance to the community of the centenary of Federation. If the ABS were only 80% successful in this task, which in itself would be a remarkable achievement, then 15% of the public would be likely either to not respond or to respond poorly in the census. In either case, the value of the census would be severely diminished. ABS considers that a public relations campaign based on the importance of the centenary of Federation is unlikely to succeed. It is also difficult to see how, should a precedent be established in 2001 for the centenary of Federation or for some other reason, it could be argued in a subsequent census that forms should not be retained. Costs of retention The retention of census forms would involve significant additional costs to the ABS to conduct the census, and for the long term administration and management of the stored records, including providing a right of access as is required by the Privacy Act, 1983. Also, it is doubtful that the level of future access in 75 or 100 years would be sufficient to meet the costs of retention. ABS sees no reason for government to provide what in effect would be a substantial government subsidy to support what is essentially private research. CONCLUSION The evidence shows there is significant public opposition to the retention of census forms. This evidence also shows that if census forms were kept, the cooperation of the public in terms of the level of response in the census and the quality of data provided would suffer. Very serious implications follow for all users of census data and other official statistics, particularly for users of population estimates. Serious impacts would be likely on electoral redistribution's, on the distribution of funds to the States and Territories, and for the host of government and private sector organisations who use census data for a variety of purposes, including infrastructure planning, service delivery, and program evaluation. The value for research purposes of retaining census forms is tenuous, at best. Many judgements have been made over the years supporting the destruction of census forms by:
In the view of the ABS, there have been no significant changes in circumstances which would warrant any change in the current policy of census form destruction. Indeed to the contrary, the AGB McNair research conducted after the 1996 Census suggests community opposition to any change in policy is now stronger that ever. The ABS strongly holds the view that all information in the form of name-identified or identifiable documents containing information collected under the Census and Statistics Act, 1905 should be defined as exempt from the provisions of the Archives Act. The retention of census forms or identified statistical records raises significant privacy issues. The retention of census forms would also have a significant impact on the capacity of the ABS to conduct an effective census, and indeed all its other statistical collection activities. For these reasons the ABS considers that decisions on the destruction or retention of census forms are matters for the Government and the Parliament. ABS considers the best way of ensuring this is for:
Australian Bureau of Statistics July 1997 INTRODUCTION THE BASIC ISSUE In order to conclude that the population census forms should be retained, the Committee has to take the view that:
The Committee should be aware that the Director-General of the Australian Archive, who addressed the same question for the 1996 Census, said on 18 February 1994 that: "A disposal authority allowing the destruction of name-identified records for 1996 and beyond will be issued. If circumstances alter for subsequent Census records then these disposal arrangements can be reviewed". Therefore, the Committee would need to be sure that significant changes in circumstances had occurred, before it recommended the practice of form destruction, adopted and endorsed by successive governments, be altered. BACKGROUND BRIEF DESCRIPTION OF CENSUS What is the Census? The Census of Population and Housing is the largest statistical collection undertaken by the ABS. The census is conducted every 5 years and the information is generally collected by self-enumeration. The forms are filled out by all householders in Australia, with only limited help from census collectors. The main objective of the census is to measure the number and key characteristics of people in Australia on census night. This provides a reliable basis to estimate the population of each State, Territory and local government area, primarily for electoral purposes (including the determination of the number of seats allocated to each State and Territory in the House of Representatives) and for the distribution of government funds. The census also provides the characteristics of the Australian population and its housing for small geographic areas and small population groups. This information supports the planning, administration and policy development of governments, business and other users. Who is counted? The census includes all people in Australia on census night, with some exceptions. Foreign diplomats and their families are excluded, as are foreign crew members on ships. Visitors to Australia are counted regardless of how long they have been in the country or how long they plan to stay. Australian residents out of the country on census night are generally not included in the census unless they have left the country without being required to undertake departure formalities (for example, naval personnel on board ships away from Australia). Since 1996, the territories of Cocos (Keeling) Islands and Christmas Island have been included in the census. When is the Census held? Since the 1961 Census, Australia has had a census taken every five years. A 1977 amendment to the Census and Statistics Act, 1905 requires that 'the census shall be taken in the year 1981 and in every fifth year thereafter, and at such other times as prescribed'. This amendment followed a 1975 decision of the High Court on the Representations Act, 1905 (see Section 3.5 for details). The last census was held on Tuesday, 6 August, 1996. What is the basis of the Census? While the Census collects information relating to each person and household in Australia, it is run primarily on a household basis. It aims to capture information for all members of a household staying at a particular address on census night. It is not a family-based collection as more than one family can share a household and, generally, information about family members absent on census night are not included with the family census record. Each household is asked to fill in the details required on the census form - this is done either by one householder, on behalf of the other householders, or by each person in the household. The household form contains space for up to six people. Personal forms are available to supplement the household form where there are more than six members of the household, or where members of the household, or visitors to the household, require privacy. Interview-based forms are used in the enumeration of the some of the Indigenous population, generally in outback areas. What data are collected? Since the first national census in 1911, the content of censuses has varied to reflect contemporary data needs. Many topics have been included in each census in that time (for example, age, marital status and religion). In the 1996 Census, a total of 52 questions was asked on the following topics: Name Basic Demography (sex and age, relationship to person one on the form, marital status, usual residence at census time, address of usual residence one and five years ago, number of children ever born) Religion (an optional question) Ethnicity (citizenship, Aboriginal/Torres Strait Islander origin, country of birth, year of arrival, country of birth of parents, language spoken at home/proficiency in English) Education (attendance at an educational institution, age left school, educational qualifications) Labour force (labour force status, occupation, industry, hours worked) Transport (journey to work, mode of travel to work) Income (individual) Only the question on religion was optional, as provided by Section (14) (2), Census and Statistics Act, 1905. Name and address information is used to help in the management and operations of the census. It is also used with the results of a small survey taken immediately after the census to check how many people were missed in the census and how many were counted more than once. Names are also used in the coding of some complex family relationships within households. Names and addresses are not stored on the computer files, nor are any other identifiers. How are the topics determined? The ABS undertakes extensive consultations with both users and members of the public to gather views about what topics should be included in the census. The process begins with the publication of ABS views on census content and an invitation for anyone interested to make a submission on the proposed content and procedures. This is followed by a series of consultation meetings around the country to discuss options. The recommendations which ABS drafts as a result of these consultations are then considered by the Australian Statistics Advisory Council, prior to the preparation of a submission to the Government. The final decision is one for Government and documents outlining the content and procedures for the next census, along with relevant regulations, are tabled in Parliament. Where are people counted? The census counts people where they were on census night and provides a 'snapshot' of the population at that time. The Australian census is based on actual location, or place of enumeration. The records of persons away from home on census night are not subsequently merged back with their home census form. About 5% of the population are counted at a place other than their usual place of residence. What data are released? Once processing of the census forms has been completed, the outcome is a file of coded records for each person, family, household and dwelling enumerated in the census. The file contains no personal or household identifiers. The release of census information is constrained by the confidentiality provisions of the Census and Statistics Act. Section (12)(2) requires that the information not be published or disseminated in a manner that is likely to enable the identification of a particular person or organisation. A number of techniques are used to ensure the confidentiality of data that is released, including making random adjustments to counts in statistical tables that are less than four. How long are Census records kept? In the past, when statistical information was extracted from census forms manually or by less sophisticated automated means, the forms themselves were crucial for producing new statistical information. Once they were destroyed, the raw data was effectively lost, leaving only the statistical tables that had been prepared. Because of this, the forms were only destroyed when it was believed that no more tabulations would be required. However, nowadays an electronic record of the 'raw data' is kept. This is made up of the fully coded responses (ie the statistical codes, not the original responses written on the census form) from each individual census form, stripped only of the name and address and other identifiers. A file of these unidentified unit record data is retained indefinitely. Therefore, it is possible to generate further statistical tables from the file at any time, as required. This ensures that this invaluable data set continues to be available to support research into the future. The original, name-identified, census forms are destroyed (they are pulped and then recycled) once the unit record file has been created and the forms are no longer needed for statistical processing. This normally takes place within one and a half years of the Census. Definition of "Census Forms" "Census forms" comprise household forms and personal forms: Household forms: Personal forms: About 1 million personal forms (including Special Indigenous Personal forms) will be used in the next census. Personal forms are used for people in non-private accommodation like hotels, hospitals etc and where needed in private dwellings. Special Indigenous personal forms are used in mostly remote Indigenous communities. Each form is 6 pages, in the form of a 3 sheet booklet, approximately 230 mm wide and 330 mm long. Successive Australian governments have supported the destruction of census forms Prior to 1971, census forms were retained for a time until it was judged they would not be required for any further cross-tabulations which might appear worthwhile after the initial analysis. Only forms for the two previous censuses were retained at any one time. In 1971 the then Treasurer (Mr Snedden) ordered the destruction of all census forms held by the ABS (Commonwealth Treasury, Press Release No. 28, 'Destruction of Census Forms", 1 June 1971 - Attachment 6). He also directed that the census forms for the forthcoming 1971 Census be destroyed as soon as the information contained in them had been transferred to magnetic tape. He reassured the public that names and addresses on the census forms would not be transferred to the tape. This action was prompted by privacy concerns. These had been stimulated by the newly formed Australia Party, which had attacked the requirement for names and addresses on census forms, and by earlier publicity about anti-census campaigns overseas. Since then, successive governments have consistently maintained the policy of destroying census forms in Ministerial Statements to Parliament announcing the conduct of the census (for example 20 November 1979, 25 February 1985, 4 May 1989; and 20 September 1994, Attachments 7, 8, 9 and 10 ). In making these decisions governments have considered relevant such factors as:
In 1979 the Australian Law Reform Commission (ALRC) expressed a provisional view in its report Privacy and the Census that "identified information should not be destroyed but should be transferred in an appropriate form to Archives" with access forbidden for 75 years. The then Treasurer (Mr Howard) rejected the ALRC recommendation, making the following statement to the Parliament on 20 November 1979:
Director-Generals of Australian Archives have supported the destruction of census forms Since the Archives Act, 1983 the Director-General of Archives has been responsible for making the decision to retain or destroy census forms. In February 1994 the Director-General advised the ABS that: "a disposal authority allowing the destruction of the name-identified Census records for the 1996 Census and beyond will be issued. If circumstances alter for subsequent Census records then these disposal arrangements can be reviewed". This authority was given after the Director-General had conducted a review into the research value of name-identified forms from the 1996 Census and concluded that the value of the forms did not outweigh the Government's interest for an effective census. It should be noted that this authority covers censuses beyond 1996. Similar reviews were conducted by the Australian Archives with respect to the 1986 and the 1991 Census forms. On both occasions the then Director-General concluded that the census forms should be destroyed. Perhaps it should be noted that throughout that time the Australian Archives Council consistently advised the Director-General that the census forms should be retained. The Australian Statistics Advisory Council includes representatives of each Premier and Chief Minister, Federal Government, business, academics and community groups. The Council has discussed the destruction of census forms on a number of occasions and has declared its support for the current policy in its Annual Reports for 1978-79, 1980-81, 1988-89, 1992-93, 1993-94, 1995-96, and will again in the forthcoming 1996-97 annual report. At its meeting in March 1997 the Council reaffirmed its strong support for the current policy of census form destruction. Council considered it unlikely that the benefits of retaining census forms would outweigh the privacy and confidentiality concerns that could be expected to arise. It also noted that little or no factual evidence has been produced to date to support a contrary view. The previous Privacy Commissioner has supported the destruction of census forms The previous Privacy Commissioner, Mr Kevin O'Connor, has also consistently supported the ABS policy of census form destruction. In a letter to the Sydney Morning Herald, dated 12 July 1996, (Attachment 3), Mr O'Connor said:
The census asks for a wide range of detailed and in some cases sensitive personal information. The Bureau of Statistics fully complies with the privacy principle that personal information should only be used for the purpose for which it is obtained, and this has contributed to a high level of community trust. If the forms were to be retained, that trust would most likely be significantly diminished, leading to a reduced level of accuracy, and undermining the important objectives of the census. It would also, in my view, be an undesirable intrusion into the privacy of all Australians. However strong the assurances initially given about confidentiality during a 'closed access' period, the Bureau would inevitably come under severe pressure to make exceptions for other public interests. There would also be differing opinions about the length of time before the forms were made available. .....The aggregate census results are of course already a major and valuable research resource. Long term access to identifiable details would be an additional use at the margin. In my view, the present policy strikes the right balance of public interests." Australian Statisticians have consistently supported the destruction of census forms All Australian Statisticians since 1971 have strongly supported the policy on the destruction of census forms. In an address to the Australian Archives Council in November 1995 (Attachment 1) the current Australian Statistician, Mr W. McLennan, set out his views on the current policy. Mr McLennan expressed serious concerns that a possible adverse public reaction to the retention of name-identified census records would affect public confidence and trust not only in the census, but in other data collected by the ABS, and thereby affect public cooperation in the collection of official statistics. This would affect the quality not only of census data, but also of data from other ABS collections. A reduction in data quality from the census would diminish its value for these important purposes and possibly put some of them at risk.
|
||||