CENSUS ORDER 1975

HL Deb 17 March 1975 vol 358 cc557-89

5.25 p.m.

Lord WELLS-PESTELL rose to move. That Articles 7, 8, 9(d) to (j), 10, 11, 14 and 15 of Part I, and Articles 17, 18(a)(vi), 19, 20 and 21 of Part III, of Schedule 2, and Articles 6, 8, 10(c) to (i), ll(a)(vi), 12, 13, 14, 16 and 17 of Schedule 4 of the draft Order in Council entitled the Census Order 1975, be approved. The noble Lord said: My Lords, we are concerned this afternoon with the draft Order in Council for a mid-term Census to be taken in 1976. Your Lordships are probably aware that we have had 10 yearly Censuses of the population of Great Britain in most recent years whose date ends in 1. This will be only the second time that we shall have had a midterm Census. In 1966 we had a sample Census of 10 per cent. of the population, but the proposals in the present draft Order are for a 100 per cent. Census. The proposals for this Census have already been made public in the Government's White Paper on the 1976 Census of Population published on the 20th February. Most of what I have to say reflects what has already been said there.

The need to take Censuses more frequently than in the past arises mainly because of the enormous amounts now spent by Central and local government -now over 45 per cent. of the gross national product: to spend this properly up-to-date facts are important because of the speed with which local populations can alter: moreover, the whole structure of local government has itself changed since the last Census and so have the authorities administering the health service. These are the main reasons why information such as the Census provides needs to be as up to date as possible. Without a mid-term Census we should have to wait until 1982 or 1983 for new figures to replace those derived from the 1971 Census. Figures out of date by as much as 11 or 12 years are just inadequate as a basis for planning and policy for- mulation. Because of the changes in local government and the increasing demand for statistics relating to fairly small areas, sample figures on subjects such as population and housing would not be adequate.

I should like briefly to explain the provisions of the draft Order. It prescribes the date for the Census, who has to make the returns and who they should cover. It also sets out the particulars required, some of which are the subject of the motion. The Order as a whole is subject to Negative Resolution, but the items in the motion require approval by Affirmative Resolution because they are not specifically authorised in the Census Act. I shall come on to the details of the items in a few moments.

My Lords, the date prescribed for the Census is Sunday, 16th May; we want to continue the tradition of holding Censuses on Sundays because most people are to be found at home on Sunday evenings. Recent Censuses have been held on the last Sunday in April, but in 1976 that date would be too close to Easter and the local elections in England and Wales rule out earlier dates in May.

Your Lordships will see from Article 5 of the draft Order that returns will be required from every household and from each individual in a non-private establishment—that is, hotels, hospitals, prisons and Forces establishments—or on board a vessel. As in previous censuses, the responsibility for making the return for a private household rests with the head of the household, but this time we acknowledge the possibility of there being joint heads. Individual members of households will also be able to obtain their own form to fill in if they prefer to do this rather than to give personal details to the head of the household The people to be included in the returns for private households are listed under Group I in Schedule I to the draft Order

My Lords, if I may refer to the position in England and Wales, the particulars to be stated in the returns are prescribed by Article 6 of the draft Order in conjunction with Schedules 2, 3 and 4. One of our aims is to cut down the burden on the form-filling public, and I should like to explain briefly how we propose to do this. Almost as much information is required from this Census in England and Wales as in 1971. We have cut out one or two subjects, including some of the most sensitive ones, but more, has been achieved by introducing a new sampling scheme. We have divided the questions into two groups, some to be asked on all forms and others to appeal on only some of the forms. Every form will contain the basic set of items together with a number of additional questions

Your Lordships will find the basic set in Parts 1 and 2 of Schedule 2 and the additional particulars in Part 3. Schedule 3 shows how the additional questions will be spread over 10 variants of the Census form. Article 6 (5) of the Order directs the Registrar-General to make arrangements so that, as nearly as possible, each of the 10 different types of form is used for one-tenth of the returns. By spreading some of the Census questions over a number of different samples in this way, we have reduced the number of questions which each person will be asked and at the same time will obtain the information that we need in order to provide the tables of statistics, including various cross-analyses of the items.

My Lords, to come to the position so far as Scotland is concerned, in Scotland, with only about one-tenth of the population of England and Wales, the size of the Census operation, and the small populations of some local areas about which detailed information is needed, make a sampling system unsuitable. The statistics derived from a 10 per cent. sample would contain too great an element of uncertainty to be acceptable. But it would clearly be unreasonable to expect every householder in Scotland to provide the whole range of information on the topics which in England and Wales will be spread over 10 sample forms. So, if we are to reduce the burden of form-filling to an acceptable level—as we think we must—it is necessary to leave out in Scotland some of the topics on which questions will be asked in England and Wales.

This, of course, involves making a judgment—in the interest of reducing the length of the Scottish Census form—as to which topics we can reasonably dispense with in Scottish circumstances. Given the general desire and need for more statistical information, this is not an easy judgment to make, as your Lordships will appreciate. The proposals for Scotland, listed in Schedule 4 to the draft Order, represent the items to which the Scottish Departments, as users of Census data, give highest priority, bearing in mind that the demands on the public in the different countries should not be seriously out of scale. As a result, the range of topics in the Scottish form will be rather smaller than the range spread over the 10 sample forms in England and Wales, but the Scottish form will cover rather more topics than any one of these ten forms. In terms of the actual detailed questions, it will be about the same length as the longest of those South of the Border, and will be shorter than its counterpart in 1971.

My Lords, there is one entirely new item which relates to England and Wales —No. 21—which is one of the items in the Motion. This relates to the journeys of school children and students to their place of education. Such journeys form a substantial proportion of the morning peak travel in many areas in England and Wales, and information about the journeys is needed to supplement information, as collected in previous Censuses, on travel to work. Apart from these items, all other particulars listed in the Schedules to the draft Order relate to matters which were included in the 1971 Census. These items are all printed in bold in the Schedules to the Order. I will take them in sequence from Schedule 2 for England and Wales, mentioning where they also appear in Schedule 4 for Scotland.

My Lords, the first item is No. 7 in Schedule 2, and is the country of birth of father and of mother; this applies only to England and Wales and does not appear in the Scottish list. The information which we shall get on parents' countries of birth, taken together with the person's own country of birth, will help us to estimate the number of people of overseas origin. We continue to need information about the areas in which the immigrant communities live, their housing and their jobs, and to compare these with the rest of the population, which the Census will enable us to do. To obtain a full picture in many areas, it is important to have parents' countries of birth, because some members of the immigrant communities were themselves born over here. The Community Relations Commission has supported this question.

Item 8, which is item 6 in the Scottish schedule, relates to the presence or absence of the person in the dwelling on Census night.

When we come to compile statistics on the composition of households, which, as your Lordships will understand, is important for social policy, we need information about absent members of the household and these items in the Schedules are designed to enable us to distinguish these people. The particulars of people who are not in employment which are to be found in item 9 (item 10 in Schedule 4 for Scotland) are printed in bold type. We need these details as a source of information on the number of people available for employment and— by cross-tabulation with other items—on how many of them are married women, for example, or highly qualified people.

My Lords, item 10 (item 13 in Schedule 4 for Scotland) relates to the position in premises of vessels, and enables us to distinguish staff from guests or patients, or the crew of a ship from the passengers, not only in the statistics but also when seeing that all returns are collected. The next group of items apply only to private households and relate to the accommodation occupied. Items 11 and 14, which are 14 and 16 in Schedule 4 for Scotland, require approval by Resolution. We need the information about owner-occupied and rented accommodation, whether private, council or tied property, in order to estimate the quality of housing separately for each category, and the information on the availability of various house-hold amenities will show where needs have still to be met.

Item 15–17 in Schedule 4 for Scotland —relates to private households and concerns the number of cars and vans normally available for private use. This will indicate the relative demand for road space in different areas and the degree of dependency upon public transport. The items which follow are the ones which in England and Wales are only to be included in some of the forms. Item 17— 8 in Schedule 4 for Scotland—concerns the numbers of people who have moved house in the one-year and five-year periods preceding census day and the areas they have moved from. Informa- tion on migration is a major ingredient in making population projections for local areas. Item 18–11 in Schedule 4 for Scotland—relates to people's employment. Among the various particulars listed there is one which is subject to approval by Resolution; namely, the main means of transport to work. At the end of Schedule 2 your Lordships will see that in the case of journeys to school and college we need to seek your approval not only for asking for the means of transport but also for the school or college address and the address from which the student or school child usually starts the journey, the latter because so many students live away from home in term-time.

The two remaining items in bold-type relate to educational qualifications. Item 19, mainly school-leaving qualifications, appears only in Schedule 2 for England and Wales; item 20 relating to higher qualifications obtained after reaching the age of 18 has a counterpart in the Scottish list at No. 12. We need the information which the Census statistics can provide on qualified people, and their distribution among industries and occupations in particular, for use in formulating policies on education and industrial recruitment and deployment. In the case of the higher qualifications, we need the details such as the names of the awarding institutions in order correctly to classify the multifarious qualifications.

I come to the question of the inclusion of these items requiring Affirmative Resolution in a compulsory national Census, since some of your Lordships may be wondering why we should not shorten the Census itself by using voluntary surveys to cover these topics. First, the Schedule to the Census Act which defines which matters may be included without approval by Resolution and which may not, was drawn up in 1920 to meet the needs of that time. Governments' requirements for information have changed considerably since then and the lists in the draft Order have been drawn up to meet the demands of today. Our predecessors of 1920 wisely provided for the inclusion of additional matters subject to Affirmative Resolution. It will be clear to your Lordships that many of the items in the Motion are closely related to those which do not need approval by Resolution, and to exclude some items in our list would severely curtail the useful-ness of this Census.

If we were to employ voluntary surveys for the collection of the information with which the Motion is concerned, we should either have to link the survey responses directly to the Census responses or else repeat many of the Census questions in the survey questionnaire so that the required cross-analyses of the various topics could be made. One way of linking the survey to the Census might be to have a voluntary section of the Census form, but this could cause great confusion and probably lead to a poor response to both the voluntary and compulsory parts of the form. Indeed, the disadvantage of a voluntary survey is that some people opt out and there can then be no assurance that those who take part are representative of those who opt out. This is why a response rate even up to 80 per cent. in voluntary surveys on subjects such as those for which we are seeking approval in the draft Order would be inadequate.

But there are subjects for which voluntary surveys are more appropriate than compulsory censuses, either because they are sensitive and people should be able to refuse to answer or because they apply only to a minority of the population and it would be wasteful to ask everyone about them. We plan to carry out several sample surveys on a voluntary basis following the Census, using the Census to provide a sampling frame. They are of two kinds : the first group is designed to extend the information provided by the Census and the second are conducted for the purpose of validating the Census itself. The first group covers subjects that are considered unsuitable for a compulsory Census or that involve only a minority of the populations; there are; four such voluntary surveys planned at the moment and details of these are given in the White Paper. If any further voluntary surveys linked to the Census are planned, they will be announced to the House in good time. The voluntary follow-up surveys to validate the Census would continue the recent practice of making sample checks on the quality of the information collected in the Census and on the completeness of coverage of households and dwellings.

My Lords, I wish to emphasise that strict confidentiality will apply to the voluntary sample surveys as it does to the Census, and your Lordships may wish to explain the safeguards that apply to the information given in the Census. First and foremost, the House has our firm assurance that no information about identifiable individuals or households will be disclosed by the Census organisation to any other Department or anyone else outside the organisation. I am using the terms "Census organisation" to mean the two Government offices conducting the Census; namely, the Office of Population, Censuses and Surveys and the General Register Office for Scotland and the people whom they employ specially for the Census such as the local enumerators. The Census Act, 1920, provides for penalties, including imprisonment for up to two years, for people employed in taking a Census who divulge information from the Census unlawfuly, and the staff of the Census offices will be subject to the provisions of the Official Secrets Act 1911. So far as possible, enumerators will not be allocated to enumeration districts in which they live or work, in order to reduce the chance of their knowing, or being known by, someone whom they have to enumerate.

The normal method of returning the Census forms will be direct to the enumerator—this has the advantages that the enumerator can help the public with any queries and can also check the forms for completeness on the spot—but envelopes will be available in which households who wish to bypass the enumerator can seal their completed returns so that they will not be seen by him. In exceptional circumstances it will also be possible for people to return their forms to a Census office by post.

I can also assure the House that every care will be taken to ensure that there is no disclosure of information about identifiable people indirectly in the statistics which are produced from the Census. These various arrangements to safeguard the confidentiality of the Census information are described in more detail in the recent White Paper. The security of the Census was the subject of two indepen-dent reports made after the 1971 Census at the invitation of the previous Govern-ment and published in a White Paper in 1973. We have carefully and sympathetically considered the recommendations in those Reports and where appropriate have implemented them. I hope, therefore, that I am able to satisfy your Lordships that the confidentiality of the Census will continue to be maintained in the tradition of the Census offices.

The main results from the proposed Census will, we expect, be produced with-in a period of three years of Census day —I am referring here to the main programme of tables, because the Census offices will continue to meet requests for additional analyses after that time. This will represent an improvement by comparison with the last Census which has been under some criticism in your Lordships' House and elsewhere because of the time taken to produce some of the results. On that point, I must say that a far greater volume of material has already been produced from the 1971 Census than from any of its predecessors. Your Lordships will wish to know that we are doing all we can to have the remaining Reports for publication ready for laying before the House by the middle of the present year. We are taking a number of steps to speed the production of results from the next Census. In particular, we shall be recruiting a larger clerical staff to code the information for input to the computer. In England and Wales we shall decentralise this operation so that the necessary numbers of people can be recruited; and we shall be enhancing the computer capacity to deal with the Census data, by installing a larger computer in the Office' of Population Censuses and Surveys, and by carrying out some of the work on the Scottish data on a separate installation in Scotland.

I should like to say something about the cost. Despite these increases in resources, we do not expect the cost in real terms of the 1976 Census to be significantly different from the last one. A total of £21 million at 1974 prices is estimated for the whole of Great Britain —£17½ million for England and Wales, and £3½ million for Scotland.

My Lords, that brings me to the end of my description of the proposals for the next Census. It is an operation which needs to command the confidence and inspire the co-operation of the public. The need for the information must be clearly demonstrated and the arrangements to protect its confidentiality must be effective. I hope that what I have said will have reassured your Lordships on these important points and I commend the draft Order to the House—

Lord LLOYD of KILGERRAN

My Lords, before the noble Lord sits down, may I ask whether the penalties in the new Order are the same as in the Census for 1971? I may have missed what the noble Lord said about penalties, because I was a few minutes late in coming in, when the noble Lord was speaking.

Lord WELLS-PESTELL

My Lords, my recollection is that the penalties are the same. I think I said that the penalties could incur a term of up to two years imprisonment.

Lord LLOYD of KILGERRAN

I did not quite understand whether that was a penalty for disclosing confidential information, or a penalty for refusing to answer a specific question on the Census form. That was the aspect of the penalties in which I was interested; that is, the penal-ties for non-compliance with a request for information.

Lord WELLS-PESTELL

My Lords, the noble Lord is quite right. I was referring to penalties for disclosure. I understand that the penalties in respect of people not completing the form are as before. My Lords, I beg to move.

Moved, That Articles 7, 8, 9(d) to (j), 10, 11, 14 and 15 of Part I, and Articles 17, 18(a)(vi), 19, 20 and 21 of Part III, of Schedule 2, and Articles 6, 8, 10(c) to (i), l1(a)(vi), 12, 13, 14, 16 and 17 of Schedule 3 of the draft Order in Council entitled the Census Order 1975, be approved.—[Lord Wells-Pestell.]

5.55 p.m.

Baroness YOUNG

My Lords, I should like to thank the noble Lord, Lord Wells-Pestell, for his explanation of this Census order. This is, indeed, a technical and complex subject; not only the subject matter in itself, but the Parliamentary procedures which go with it, and we are very grateful for the clarity of the noble Lord's explanation. We on this side of the House are glad to welcome the Order in general terms. It is a proposal to have a mid-term Census which has the support of all the local authority associations; and it is not difficult to see why. The Census has been taken every 10 years since 1801 —with the exception of 1941—and it started with a simple head count, but; now, of course, it is an essential tool in the economic, social and educational planning processes of this country. Because; of the complexity of our society, it is essential to review the purpose of the Census from time to time. This Census is welcome, in that it is to be, as I under-stand it, a full Census and not a sample Census, such as that taken in 1966.

In that Census, I understand, only a 10 per cent. sample was taken and, owing to faults in the sampling framework, there were a number of errors which soon became apparent, particularly in urban areas. The Census proposed is therefore a vital tool in regional and local planning, particularly in the new rolling structure plans now well under way in all first-tier authorities. Equally, the Census is needed because the results of the 1971 Census, to which the noble Lord, Lord Wells-Pestell, referred, are coming out so slowly, so that some local authorities, considering their structure plans, must still use data derived from the 1961 Cen-sus. The difficulties and the inaccuracies derived from using data now at least 14 years out of date must be evident to any-one, and I believe that the Greater London Council has already experienced difficulty over its planning because it has not had the up-to-date information. I am therefore pleased, as I asked a Question on this matter in December, to hear that the noble Lord, Lord Wells-Pestell, agrees with me on how important it is for local authorities to have this up-to-date information readily available.

My Lords, the need is vital, but the proposed Census will be of real use only if the results are produced quickly. What local authorities, for example, will want to be assured of is that the undertakings given in paragraphs 31 and 32 of the White Paper are really watertight. Para-graph 31 of the White Paper says that the aim of the Government is to pro-duce the results within three years, and this has been repeated this afternoon by the noble Lord, Lord Wells-Pestell. This is a long time, because fewer questions are being asked. It is all the more disappointing as both the White Paper, and the noble Lord, referred to the new processing offices and the new and more powerful computer. But can we be sure of the effectiveness of the new computer? If I may quote from Hansard and the Question that I asked on this subject, I was told at column 4 of 2nd December: There have been computer problems, in regard to which the Government are not alone. My Lords, I am not a computer expert, but this statement has an ominously familiar ring about it, and what we need to know is whether the computer now installed will be subject to lengthy problems of what I think is called " running in"? I should like to ask specifically: when is the computer to be installed? I should also like to ask whether the noble Lord, Lord Wells-Pestell, can give an undertaking that it will be working effectively as soon as the Census results are available. I am glad to hear the assurances that he has given; but I am certain that he will agree with me that the whole value of the Census is in get-ting out the results, and it is very un-fortunate that some of the results of the 1971 Census will be available only in mid-1975.

I should like to know also what the Government are to do about all the organisations which were asked to have last-minute information extracted from the Census. This, too, I believe was a cause of delay in the publication of the 1971 results. So that although we on this side of the House welcome the pro-posed Census, we do so provided that the results can be produced to the time-table proposed; otherwise, I believe that the whole Census will be to a large extent wasted and, at the same time, will be a terrible waste of public money. I am glad to note what the noble Lord said about the cost of the Census. It is approximately what was forecast in the article in The Times of 2nd December on the need for a five-year Census. I suppose one must be gratified that the cost will be no greater than the 1971 Census, even at 1974 prices and even taking into account that fewer questions are to be asked.

My Lords, turning now to more detailed points, there are a number of questions I should like to ask. I found the accompanying White Paper to the Order very helpful, particularly about the reasons for the questions to be asked. No explanation seems to have been given for leaving out a number of questions and, in particular, the questions listed in paragraph 8 of the White Paper. The noble Lord, Lord Wells-Pestell, hinted that these were left out because they were the most controversial. If this is the reason, although it is not stated in the White Paper, it is a very real reason for omitting them. The fact of the questions being left out will affect the comparability of the 1976 Census with the 1971 Census; and this in turn will affect basic planning decisions. Planning is both a short-term and a long-term process, and long-term trends are important. It seemed to me unfortunate that the question listed as Occupation one year ago "has been dropped in view of the current employment situation. Turning to questions on housing, I wonder why any question of leaseholds has been omitted. This was a subject debated in this House as recently as last July, and a question on this subject would monitor this declining sector of the market. It would be interesting if, as is suggested, the Community Land Bill included provisions about selling land on long leases for house building.

It is with hesitation that I raise any question about the Welsh and Gaelic languages. I was surprised to see that on the questions on Scotland concerning the Gaelic language the speakers were not to be asked if they can read or write the language, but simply if they can speak it. On the other hand, the Welsh are to be asked if they read or write Welsh, as they were asked in 1971. I should have thought that the whole issue of languages was to what extent they were living and used languages, and therefore it is essential to know how many people not only speak it but read and write it and what development there is in the language itself.

My Lords, there is also the question of a voluntary Census. I was interested to hear the explanation for the various questions that are going to be posed in the voluntary Census. This seemed to work on a very good principle of collecting a great deal of information which would take a great deal of time if it were included in the whole of the Census form. A question of enormous importance is that of migration within Great Britain. As I understand it, the voluntary Census is to supplement questions being asked in the main Census as to why people move. What is equally important is why they do not move. This gives you the balance of population between town and country, a subject of great interest not only to geographers, social scientists and to demographers, but one that will give a great deal of information about the problems of inner cities and of development and developing areas. I should be interested to know why it has been decided not to ask the question as to why people do not move from places as well as why they do. It would be interesting to know whether the population is becoming more static, having moved about a great deal and, if so, why it is. On the other hand, paragraph 22 of the White Paper specifically states that further voluntary surveys could be carried out. It would be interesting to know whether, as a result of this, the Government might consider a voluntary survey simply on this subject.

In such a detailed matter as this there are many questions that one could ask. The final question is on the employment sector. I wonder why the Government had not considered a question about the employment of children. This is an important social issue and one which affects social and educational policy. I should like to conclude with two general observations. I think that there must be good and effective publicity about this Census and its purpose. There are to be ten different types of Census form to be completed by a different 10 per cent. of the population. Although there will be what is called a hard core of questions, people will inevitably wonder why some are being asked one lot of questions and some are being asked another, and whether there is any significance in the difference in the questions being asked. I trust that when the leaflets are delivered to households setting this out, they will set out the reasons for it; other-wise people will be justifiably concerned as to what they are being asked to fill in.

The second general question concerns the issue of confidentiality. I do not intend to pursue this at any length this afternoon, not because I think it is not important; on the contrary, I think it is exceedingly important. I think it is almost a subject for debate in itself. Over these matters I was glad to have the assurance of the noble Lord, Lord Wells-Pestell, that the Government have considered the results of the complaints of the 1971 Census and have done their best to act upon them. I appreciate very much that in these matters we must weigh in the balance the value of the Census and the data it can give us against the inevitable intrusion into the privacy of the individual.

I am aware of the criticisms arising from the last Census, and I hope that the assurances we have heard from the Government this afternoon will be fulfilled as far as possible, and that the Government will tell the House before the Census takes place if they are going to act on the suggestion of the Royal Statistical Society that there should be a standing review body to handle complaints about the Census. It seems to me that this is a very good suggestion and one which I think everybody would be glad to follow. It would mean satisfaction to those who felt that the questions were perhaps insensitively put or that perhaps there was too great an intrusion, and that perhaps the kind of criticisms made last time could be brought before an impartial body. I hope that the Government will consider this matter seriously before the Census takes place. It seems to me that they have time to do so.

My Lords, I end as I began, by saying that we on this side of the House welcome the 1975 Census Order and hope that the information gained will be speedily processed so that the vital decisions necessary in our complex society can be made on the most up-to-date of information.

6.10 p.m.

Lord AVEBURY

My Lords, I cannot altogether share the enthusiasm of the noble Baroness for the Census. I cannot altogether agree with her that the need is vital and that this Census is an essential tool until that has been demonstrated. I will agree with her that the value of the Census has to be weighed against the inevitable intrusion into privacy which it involves. I think that the fundamental objection to any Census is as was stated by my honourable friend Mr. Alan Beith, in a letter to The Times on 25th February. He said: It constitutes an extensive and detailed social survey but it retains the legal sanctions against non-completion that are appropriate only in a population count. Nobody, so far as I am aware, objects to filling in Form A of the Electoral Register, which asks only for the names and addresses of the adult members of a household and, in the case of those reaching their majority during the period for which the Electoral Register will be in force, for the dates of birth of those individuals. But it is when the questionnaire is expanded to include material about the personal backgrounds of house-hold members that resistance begins. And the more information is demanded the more resistance there is, together with the higher number of refusals and, of course, false answers. One could take the purist line and say that no compulsion should be exercised beyond the basic details of name, sex, marital status and age, and that all the remaining questions should be answered at the voluntary discretion of the householder. That suggestion has already been mentioned.

That was evidently the view of my honourable friend in his letter to The Times, and it is perfectly logical; but I suggest there is an intermediate stage between that position and the complete acceptance of the proposals in this Order that I detected in the speech of the noble Baroness. One could say that all the questions about premises—which are obviously not of a personal nature—are clearly harmless and that the questions specifically laid down by the provisions of the 1920 Census Act have been sanctioned by long custom and practice, but that the remaining information could be ascertained by other and voluntary means. With respect to my honourable friend, I think it would be likely to give rise to confusion if some questions were voluntary and others compulsory, and the extent of non-completion of the voluntary questions would probably make the statistics resulting from the Census highly inaccurate. Neither those who opted out nor those who agreed to answer all the questions would necessarily be a true cross-section of the public as a whole.

Mr. Joseph Hanlon, writing in New Scientist on 27th February, proposed a two-part questionnaire, the second part of which would not contain the name and address of the respondent. If the two parts were to remain permanently separated, it would mean that statistics were no longer available for particular areas, because answers in the second half of the form would never be correlated to names and addresses in the first part. For example, it would not be possible to say how many dwellings in a given local authority area did not have a fixed bath or shower. A compromise might be to give each part of the questionnaire a code number and allow the householder to return each part separately so that they would be "married up" only after being input to the computer. At that stage the data is filed by reference to the Ordnance Survey grid within 100 metre squares, and not by reference to individual names. So the problem of confidentiality would be greatly reduced, though not altogether eliminated.

The two halves of the form, having been processed, would then be filed separately if it was still thought desirable to retain them for historical purposes for 100 years, as has been done so far. It would be interesting to know how much use is made by historians of the material from the early nineteenth century Censuses which is now available to them. If the two halves of the form are processed and then stored in different places, neither the enumerators nor the clerks in charge of data entry would see both of them, and it would double the hurdle a snooper must surmount before gaining access to information about a particular individual. There is always the possibility that an extremist Government itself might wish to use such material; and this has been canvassed. In my view, even if the previous records had been destroyed such a regime would have the power to reconstruct them so that destruction of the records would not be any adequate safe-guard against that eventuality.

What is the purpose of a Census? According to the White Paper: Census statistics are needed by Government departments and local authorities for planning and developing their policies and for monitoring the effects of those policies. They are also needed to inform Parliament and the general public and for industry and research. Those objects, of course, are entirely legitimate in themselves, but we have to ask three questions: The first is: is the information used to an extent that justifies the cost of collecting it and the obligations imposed on citizens to provide it? Secondly, could the same information have been gathered by other means? Thirdly, what are the safeguards against unauthorised disclosure?

On the first point it will be obvious, as the noble Baroness mentioned, that the longer it takes to produce the information the less useful it will be. There are still quite a few analyses from the 1971 census remaining to be published. The main reason given by the noble Lord, Lord Wells-Pestell, for the delay is the vast in-crease in the quantity and detail of the results which are demanded. It is per-haps in the light of these circumstances that five subjects on which questions were asked in 1971 are being omitted altogether in 1976. But the noble Baroness will remember how, when we were dealing with rents legislation last summer, we had no information about the number of privately rented furnished properties, and we were arguing about the problem in a vacuum. This was a very good illustration of the uselessness of taking censuses and then not providing the information to Parliament when required. It is far too late for the 10-year tables to have been published after that legislation had gone through your Lordships' House and another place. No doubt many years will pass before we have another opportunity of amending that legislation, if it were thought desirable to do so in the light of figures ultimately produced.

I want now to consider in particular the use that might be made of the information provided in connection with Questions 7 and 17, dealing with parents' country of birth and usual residence one year and five years previously. Professor Sir Claus Moser, writing in Social Trends 1972, discussed the collection of statistics about "immigrants" in some detail. He distinguished between the strict use of the word in its meaning of a person who has moved into this country from abroad during the period under consideration, and its extension to groups whose family origins are overseas. He says: Where discussion is of a group which, in principle, it is required to characterise in terms of race or colour, descriptions such as 'New Commonwealth ethnic origin' are used and statistically defined … I believe the noble Lord, Lord Wells-Pestell, when he remarked that we continued to need the information about immigrant communities, really meant communities which included people of "new Commonwealth ethnic origin", because he was talking about the parental bithplace question and was including in his remarks persons born in this country to parents from overseas.

From the Census we know, for instance, how many persons resident in the London Borough of Haringey in 1971 has a father, a mother or both parents who were born in Cyprus. But where does that lead? It means we are supposed to view these people as "problems", creating burdens in the area because of their ancestry, as Sir Claus makes clear later in his article, because he writes about: … selective social expenditure policies, like the Urban Programme and grants under Section 11 of the Local Government Act 1966 (both of which were influenced by immigrant problems) … ". The incessant coupling with the word "immigrant" of the word "problem" fosters and encourages the racism which these well-meaning policies are intended to counter. The more one accepts this principle of channelling money into areas with high concentrations of people of "new Commonwealth ethnic origin" the more difficult it will become to contain racism. The anti-immigrant natives will surely point out that even fairly liberal Governments and local authorities view immigrants as problems. But it does not satisfy me when the noble Lord, Lord Wells-Pestell, tells us that the Community Relations Commission has agreed to Question 7. I have had representations from the Standing Council of Asian Organisations in the United Kingdom, from the leaders of the Cypriot community, from the Joint Council of Welfare of Immigrants and from the National Council for Civil Liberties. With great respect to the CRC, I believe the organisations I have mentioned are more closely in touch with grass-roots opinion among the immigrant communities.

If the Census statistics were used as a yardstick for distribution of age to local authorities, I suggest they would be quite inadequate. In the 1971 Census persons not born in the United Kingdom were asked to give the year of first entry into the country. Without approving of that question, one could see the point of it. A person from a non-English speaking country who had been here for several years would be less likely to have language difficulties than someone only recently arrived. But that question has been dropped in the 1976 Census, not-withstanding the assertion of the Minister who replied to the debate on the 1970 Census Order in another place, that the information was indispensable. The reason why this "indispensable" question has been omitted this time may well be that given in relation to Question 7 by Mr. John Blake, writing in New Society of 16th March 1972. He said: The possibility of error is, of course, likely to be much greater for some questions than others. For example, there can be few people who will now rely on the information given about parents' place of birth, the question being so closely related to recent legislation establishing patriality as a basis for immigration control. The sensitivity of information about date of entry was obviously even greater than that of parents' birth place, since the former is omitted while the latter is now retained. But this is retained for only England and Wales, and if it is indispensable to know about parental birth place of the residents of Liverpool, why is it not also indispensable to know the parental birth place of residents of Glasgow? It is also of great interest to note that in the test Census conducted by the Office of Population Censuses and Surveys in 1974 to evaluate the 1 in 10 technique which is now being used for Part II questions, parental birth place was omitted altogether. I would further observe that not one of our partners in the EEC has a parental birth place question in its Census, and it must be for consideration at some point in the future that we should harmonise the questions asked in our Census with those of our partners in the EEC.

Even if one could be certain that correct answers were given in the overwhelming majority of cases, the statistics would, I suggest, be an unsatisfactory yardstick for any purpose one can imagine. The population of newly arrived immigrants in a given area can change very considerably between one Census and the next. For example, since 1971 this country has received refugees from Uganda and more recently from Cyprus, and quite naturally these immigrants tend to settle in areas where they find others of their former compatriots. It may well be desirable to have prompt information about children arriving from overseas so that educational provision can be made for them by the local education authorities well in advance—as did not happen in the case of the Ugandans and Cypriots. The Joint Council for the Welfare of Immigrants have suggested that when entry certificates are applied for on behalf of dependants overseas, details of the ages and numbers of the children should be taken and passed on to the local education authorities concerned. No one is likely to object to this procedure if it is clearly explained to them, since it is obviously for the benefit of the children. Once the dependants have arrived and are placed in schools, then it is for the head teacher concerned to evaluate their needs, and this can be done only with real children and not with statistics.

To take an example from another field, consider the Runnymede Trust Report on Race and Council Housing in London. That report shows that at the time of the 1971 Census persons with both parents born in the new Commonwealth (abbreviated to BPBNC) were less likely to obtain a council house and, when they did so, it was likely to be in older properties on high density estates. These conclusions could have been arrived at from casual observation, although of course we should not have had the exact percent-ages. But as the report itself admits, since 1971 about another 200,000 households have been rehoused in London. So if the argument is that accuracy is essential, it would be much better to conduct a special survey for this purpose alone.

However, that raises the question of what purpose one is to put the information to once it has been collected. The Census gives no information about the ethnic origin of persons who are on the waiting lists of local authorities, so it is impossible to say whether any bias exists in selection procedures. As the report puts it: Without proper monitoring of allocation practices it is not possible to say how far uneven distributions are the consequence of discrimination of of other factors. In other words, the need is for local authorities to examine their own pro-cedures rather than for analyses to be made of the situation as it was four years ago. Until 1971 Islington's policy was not to rehouse furnished tenants from clearance areas, for instance. This, the Report says, must have limited quite severely the access of BPBNC to council housing during the 1960s It would be useful to know of any other rules operated by the Greater London Council or the boroughs which might indirectly discriminate in this way. I hope that the Runnymede Trust might look into this question. The presentation of statistics from a four year old Census is not likely to result in the amendment of any such discriminatory rules that exist.

When the last Census Order was debated in another place, I argued that neither the parental birth place nor date of first entry to the United Kingdom were relevant to the social or civil condition of the population, and that those questions—only one of which now remains—were therefore ultra vires, falling outside the limitations which were specified in the 1920 Act. The Minister who replied to the debate said that because the Order had gone to the Select Committee on Statutory Instruments, and they had had nothing to say on the matter, I must be wrong. That was as far as he went, so my argument has never been refuted in public. It is necessary, I suggest, to consider what limits, if any, are placed on the information which may be legitimately collected by (he phrase in the 1920 Act, Any other matters with respect to which it is desirable to obtain statistical information with a view to ascertaining the social or civil condition of the population. The assertion of the Government appears to be that under this provision the State is given a blanket power to ask any question it pleases so long as the Select Committee on Statutory Instruments does not kick up a fuss.

The trouble is that, with the vast increase in delegated legislation over recent years, the Select Committee may not have time to scrutinise every Order coming before them with the care that is needed, and inevitably some things that ought to have been stopped will slip through without being noticed. I think that in the case of a Census the issues involved are so delicate that we should not treat them just in the same way as any other Order which is not concerned with matters that touch the liberty of the individual. In my view, there should be a special Select Committee to consider the desirability of all questions proposed under the discretionary provisions of the 1920 Act, and the task would be not only to make certain that the questions are within the scope of the Act but to receive evidence from Government Departments, local authorities, universities and any other users, who are always demanding more information, to see whether compliance with their demand is really in the interests of the people.

Apart from this, we must consider what additional safeguards are needed to make sure that whatever information is collected is not misused. I must emphasise here that it is not a question of the extent and magnitude of breaches of security that actually occur, but of the feeling of respondents about the likelihood that they may occur. It is not only Ministers in the Home Office or even the experts in the Royal Statistical Society who need to be satisfied with the arrangements for confidentiality, but the millions of house-holders who are going to be made to complete the form. In spite of all the publicity that was given to the 1971 Census, according to surveys which were described in the White Paper, Security of the Census of Population, 27 per cent. of those who were interviewed believed that anyone working for the Government was allowed to see the forms, and 45 per cent. thought that there is a risk that information could fall into the hands of the wrong people". Partly, these fears can be dealt with by improved public relations, as the Government themselves have recognised and as the noble Baroness has mentioned this afternoon. Having at first rejected the RSS recommendation that a White Paper should be published in advance of the Order, very sensibly the Government changed their mind and have now done so. This may at least help to remove some of the fundamental misunderstandings that evidently still existed after 1971.

Having cleared those matters out of the way, as one may hope, there are still some important grounds for anxiety to which the Government appear to be in-sensitive, and I shall mention just three. The official attitude on the related proposals made by the RSS and the British Computer Society for independent monitoring of security arrangements and the handling of complaints on a continuous basis instead of ex post facto, as was done last time, is still that it would be premature to make any decisions on these matters independently from the general conclusions which the Government will reach on the recommendations of the Younger Committee on Privacy. As the memorandum published this morning by the National Council for Civil Liberties rightly says, it is disgraceful to use that excuse when the Younger Report was published as long ago as July 1972 and Ministers are regularly fobbing off questioners with the assurance that a Statement will be forthcoming in the near future, the most recent occasion being when the noble and learned Lord, Lord Gardiner, put a question to the Government on 27th January and received an apology from the noble Lord, Lord Harris, who admitted that a Statement on the Younger Report had been half promised before Christmas and that now he was only hoping to produce one before Easter.

It is very difficult for us to argue this case this afternoon when we still have no knowledge of the Government's official attitude to the Younger Report. In the meanwhile, the Government pointed out in the Security White Paper that because the Registrars General in England, Wales and Scotland, carry out their duties under the directions of Secretaries of State, com-plaints can be dealt with already through Parliamentary Questions and, ultimately, through the Ombudsman. It will be appreciated, however, that Members of Parliament have to deal with Census com-plaints among literally hundreds of other matters which are brought to them by their constituents and would never be able to see the patterns that might emerge if all complaints were dealt with by a single body which also had the resources to monitor the effectiveness of the detailed procedures which have been mentioned by the noble Lord for ensuring confidentiality.

Secondly, there is the use made of Census data by the Social Survey Division of OPCS and also of ad hoc tabulations by outside customers. The RSS say that it is doubtful whether follow-up services of the kind which have been mentioned are permitted. They are not alluded to in the 1920 Act; it is only since 1970 that the social survey has been combined with the Registrar General's Department to form the OPCS. On that principle, there is nothing to prevent the OPCS from being still further expanded by merging it with some other office of Government and then transmitting information to that office without breaching the letter of the law. It is unsatisfactory that other subjects which are not covered in the White Paper may be dealt with, using the Census as a sampling frame in the way that the noble Lord has described, without any Statement being made to Parliament, while we still have an opportunity of questioning it, as we do this afternoon. If the noble Lord says later that additional voluntary surveys will be conducted using the Census as a sampling frame, we shall have no opportunity of voting against that, because the Affirmative Resolution procedure is the last instance that we have of dealing with the matter.

On the sale of tabulations to third parties, the assurance that all information will be arranged in a way that avoids disclosing anything about individuals, whether directly or by comparison with other information, is of course welcomed. Yet according to Mr. Hanlon in the New Scientist article which I have quoted already, he says that, it is easy to buy data which tells a surprising amount about an individual. This point has already been raised by the Royal Statistical Society which suggested that it could be checked by the review body. In reply, the Government said that, first, the general guidelines on confidentiality apply to unpublished equally as they do to published tabulations, and the noble Lord has repeated that this afternoon; secondly, where a tabulation refers to small numbers, random adjustments are deliberately made, virtually eliminating the possibility of indirect disclosure; thirdly, a record of all the ad hoc tabulations and the names and addresses of their recipients would be made available for public inspection. But who is to verify the effectiveness of these safeguards? Obviously, they ought to be audited by professional statisticians who are com- pletely independent of the Government, as the RSS has proposed.

Lastly, there is the matter of the choice of enumerators. In the book The Invasion of Privacy by Tony Smythe and Donald Madgwick, it is shown that enumerators sometimes cover the street that they themselves live in and that among the types of official employed as enumerators were income tax inspectors and rating officers. Indeed, the Security White Paper says that as many as 29 per cent. of the enumerators were local government officers of one kind or another. The RSS proposed that no enumerator should operate within five miles of his home or place of work. However, the Government said that this would affect recruitment, add £1½ million to the cost and decrease efficiency, because the enumerators would be operating in unfamiliar surroundings. While I accept those arguments, may I suggest as a compromise that no enumerators should cover either the enumeration district in which he lives or works—the noble Lord has already given an assurance on that point—or the districts immediately adjoining it. That would give some necessary precision to the undertaking, which is also contained in the White Paper, that extra care will … be taken to reduce the chance of overlap between an enumerator's personal or business contacts and his enumeration area. Turning to the conditions of eligibility to be an enumerator, the National Council for Civil Liberties rightly demands that tax inspectors, policemen, immigration officers, salesmen and some kind of local government officials should be ruled out. I think the sensible approach would be to give priority to the unemployed and, after that, to housewives, and only then to consider people who already have jobs of their own—but not the kind of sensitive jobs which the NCC mentioned.

My Lords, I apologise for having taken up so much of your Lordships' time, but the value of the Census depends on the extent of the co-operation it secures from the public at large. If satisfactory assurances cannot be given upon the matters which I have raised, then the quality of the information will be downgraded, and to that extent the planning which is based upon it will be less effective. In 1970, the Government ignored the friendly advice which I ventured to offer. The result was that in that Census more people refused than in any previous Census, and wide-spread resentment and hostility was caused. This evening it would be only a gesture if I were to divide the House against this Order. However, in the absence of any signs of fresh thinking on the part of the Government, I shall be tempted to do so.

6.39 p.m.

Lord WELLS-PESTELL

My Lords, the noble Baroness has had to leave the Chamber for a moment or two. In her absence, perhaps I may attempt to deal with the points raised by the noble Lord, Lord Avebury, but I cannot give him a comprehensive reply to all the points he has raised this evening. I am aware of his attitude with regard to a number of them, because he has made his opinions clear in the past. I can say only that I prefer to read at leisure what he has said, and I am sure that the Department will do so, also. I do not think that I can go much further than that tonight.

The noble Lord appears to have made a valid criticism of some of the workings of the Census. I am not suggesting for one moment that they are new criticisms. However, they are matters which we need to look at in detail and I do not presume to give an off-the-cuff reply this evening. In the last analysis, whatever we may feel about a Census of the kind we have in this country every 10 years, the purpose of a comprehensive Census is ultimately to serve the community in the best possible way, to find out how it is living, and the various things which impinge upon the community as a whole, which we hope will result in some kind of good Government which will give them a better and a fuller life.

To do this, as I see the situation, we obviously have to get as much information in those fields as the Government, rightly or wrongly, feel is essential in order to meet the needs of the people and I think every possible precaution is taken to safeguard that information. So far as I know, there is no evidence at all that there has been any serious breach of confidentiality—or any serious leak if a breach has taken place—or even that there has been a disclosure that should not have been made. When that kind of guarantee is given—and so far as past Censuses are concerned I think one can say there has been no evidence at all of any malpractice—we should want to look at what the noble Lord has said to see whether anything can be done with regard to a future Census. He is in a position, as noble Lords in this House always are, to seek to improve upon an Act of Parliament or to amend it, and it may well be that the time has come when we ought to do this.

The noble Lord referred to people whose parents were born abroad and whose children or dependants in this country are regarded as problems. This may be so, but the attitude of some local authorities does not stem from the Census. It stems from the fact that the community recognises them as being a different group and tends often to feel that they are responsible for certain types of behaviour or that they are inadequate. I do not see what else the Department could have done in addition to consulting the Community Relations Commission on the steps that we have taken in respect of questions in the Census. I presume that the CRC does not commend itself to the noble Lord, but I should have thought it would have been difficult to think of any other source to go to in order to see whether what the Department proposes to do is in itself correct.

The noble Lord raised the question of the parental background of those filling in the forms as not applying to Scotland. That was something I had noticed myself and wondered why it applied to England and Wales and not to Scotland, and the reply I was given (and it may not meet the noble Lord's query) is that it was not considered necessary to apply it to Scot-land because the numbers of people involved are of no great significance, and since therefore one wants to keep the Census down to a minimum number of questions, unless it is a matter of great significance there is really no point in including it.

Referring now to the enumerators, I live in an area where, if one did not have somebody who was fairly remote from that area all sorts of things could happen, and I can only say that very considerable care is exercised in the choice of enumerators and where they go. Again I know of no evidence where there has been any abuse by them; they are under the very severe penalty of up to two years' imprisonment. Of course some others who are engaged in the Census, as I said in my earlier speech to your Lordships, come under the Official Secrets Act.

The noble Lord asked whether historians are making use of the Census records from the earliest times. My understanding of the situation is that the Public Record Office report that Census forms of the late 19th century are the most consulted type of record open to public inspection. If there is any information about the type of user—and I cannot answer that now—perhaps the noble Lord will allow me to write to him on that particular matter.

The noble Lord also suggested a two-part Census form, one part being without the name and address. I should have thought that the name and address were necessary to check that each person had completed the Census form, and I would not think that the form could have any validity at all without that kind of information.

Lord AVEBURY

My Lords, if the noble Lord will allow me to interrupt him, what I suggested was that each half of the form should have a code number, which might even be machine readable, which would enable the two halves to be married up once they got into the computer. If the person had simply sent in the top part which contained a name and address and marital status and the ages of the occupants of the house and he had not filled in the second part, that would be discovered once it got into the computer and a reminder could be sent to him. It would be impossible for the second half not to be returned without that omission becoming evident once the two halves of the form were correlated with each other in the computer.

Lord WELLS-PESTELL

My Lords, I think the Government would have to examine whether this is going to make the task of compiling an adequate Census easier or more difficult. I do not know the answer to that, but again I would remind the noble Lord that in the absence of any evidence that there has been a breakdown in the confidentiality I could not at this stage be convinced that such a move was absolutely necessary. The proposed procedure seems likely to cause confusion, and it does not seem that it would inspire confidentiality if we were to adopt the procedure that the noble Lord has in mind. This is something that I should like to look at again, to see whether it would serve any useful purpose.

The noble Lord referred to what I said some months ago about the remaining parts of the previous Census. I should like to take this opportunity of repeating what I said then; namely, that the most essential matter is available and has been available for some considerable time. With respect, I would suggest that the more important things have gone out and they did at least go out much earlier than on the occasion of any previous Census. I think one must also bear in mind that the demand for the material which has been compiled is three times greater than in any previous Census. I accept that this does not completely excuse the fact that there are still parts of that Census which are not yet avail-able, but we hope they will be available in the fairly near future.

The noble Lord raised the question of the parents' countries of birth being in the Census Act. The view of the Government is that this information is an indicator of local immigrant communities for which it is desirable to have statistics on the social conditions. I would not have thought that was in any way detrimental to them. The questions on the local placement of the enumerator I have already dealt with, and I think the only points I have yet to deal with are—

Lord AVEBURY

My Lords, I am sorry to interrupt the noble Lord again, but on the subject of the local placement of the enumerator what the noble Lord said in his speech was that as far as possible an attempt would be made to ensure that the local enumerator did not operate within the district where he either lived or worked. I was suggesting that that should be extended so that he would not work in any district contiguous to that in which he either lived or worked. The noble Lord might care to think of removing from his undertaking the phrase "as far as possible", and make it an absolute condition.

Lord WELLS-PESTELL

My Lords, I was attempting to say to the noble Lord, Lord Avebury, that the difficulties of this situation are not altogether un-known. What I am saying is that so far as is humanly possible, every step is taken to see that an enumerator is not likely to be known in the area or district in which he is going to work. It may well be that one could lay down that he must not live within a certain distance of the area, but in these days, when people travel so much, he might go into an area where, although he lives a certain number of miles away from it, he could meet people he knows. The only way to deal with this is in the way it is being done at present; that is, as far as is humanly possible to see that people who are given a particular district have no connection in any way with that district. I should like to read in detail and more carefully, and in an entirely different atmosphere, what the noble Lord has said, to see whether any of his suggestions can be taken into account, not perhaps in this Census, but in the future.

My Lords, I should like now to deal with the points raised by the noble Baroness, Lady Young. I am sorry I have to do so in her absence, but I am sure she will understand. The noble Baroness asked whether we can be sure that the returns will be available at the time we say they will be. I suppose if I am honest about it I cannot give that assurance, but can only say this. It is the intention of the Government that the results of the Census and the information stemming from it will be available with-in three years. This depends on a number of factors. I explained that we are adding to the computerisation of these matters. The noble Baroness asked me when the computer to which I made reference would be installed. The answer is, in the next few months. I cannot give any undertaking on whether it will work efficiently. It seems to me that the more computers are brought in, the greater the risk sometimes of breakdown, but so far as is humanly possible, it is our intention that the results shall be known within a period of three years.

My Lords, the noble Baroness asked me to give the reasons why the matter relating to leaseholds was omitted from the form. My understanding is that leaseholds were not included because, rightly or wrongly, they were not thought to be of sufficient importance. I am sure it will not satisfy the noble Baroness, but I hope she will accept my assurance that the matter was considered. The noble Baroness also asked why we are not concerned with the question of why people did not move. There are two factors here. The first one is that one wants to keep the questions down to a minimum. We want to get information that can be of considerable importance. It is much more important to know when people move, and why they move. Moving within one year and moving within five years is provided for in the Census. It is more important to have the information for the two periods, and to know the reasons why people move rather than perhaps to have to burden the majority of the community—and I think it is the majority of the community —who are remaining static. I use the word "majority" in a percentage way.

My Lords, I think it was the noble Baroness, Lady Young, who raised the question of the standing review body. This is a matter to which I understand the Government are giving active consideration. My honourable friend the Minister of State for Health has asked for information on this matter, but that is all I know at this stage. The noble Baroness also asked whether there was any significance in the differences between the questions that people will be asked. The answer is "No". The different types of form will be issued randomly by the enumerators. There are ten forms. This will mean that we shall have ten dif-ferent forms, and they will be repeated. But in that sense the forms will be given out randomly, not to selected people.

Another question from the noble Baroness was why the question "Occupation one year ago" has been dropped. Again, there has been no demand from the Departments concerned, and because of this it has not been included in this particular Census. The Social Survey on Migration may investigate the reasons. I said earlier that we were not concerned to find out why people did not move; but on the other hand, it may emerge, although not specifically provided for. The noble Baroness also asked, "Why not ask Gaelic-speakers if they can read and write Gaelic?". The question was extended to reading and writing in 1971. There was some lack of compatibility in the results compared with 1961, due in part possibly to the extended question. The Government hope to resolve this by reverting in 1976 to the 1961 question. In the mid-term Census, also, we wished to restrict the number of questions. Mainly for that reason the matter has not been included this time.

I cannot think of anything else that the noble Baroness asked. I dealt with one or two matters before she was able to return to the Chamber. Perhaps one of these days she will see these in Hansard. If for any reason I have missed any of her points—and I say this also to the noble Lord, Lord Avebury —I hope that either or both will not hesitate to write and tell me so.

On Question, Motion agreed to.