I have to consult the views of the House about the next Business and so have to require a high degree of silence. There is a Motion by a Minister and a Prayer in the name of the right hon. Member for Bridlington (Mr. Wood). The Statute under which the Minister's Motion and the Prayer have been tabled tonight places the House and the Chair in some difficulty. The draft Order in Council under the Census Act, 1920, Section 1(2), has to lie on the Table for 40 days, during which either House of Parliament may resolve that the draft be not submitted to Her Majesty. The Prayer, therefore, being against the whole of the draft Order in Council, allows the House to debate the whole of the Order. On the other hand, the affirmative Resolution which the Minister seeks to move is limited to certain words in the Schedule, and for the Chair to restrict debate to those words only would impose a highly artificial restriction on Members' speeches.
I therefore propose to follow the only recent precedent—it was of 4th May, 1960—and to suggest that both Motions should be discussed together and that the scope of debate should cover the whole the draft Order in Council. If necessary, I would be prepared to allow the House to divide on each Motion separately, but if the hour of 11.30 is passed and debate is still continuing then I would put the Motion for an affirmative Resolution only and the Prayer would fall and would not be called again on any subsequent day. I hope that this course will be acceptable to the House. It is, I believe, the only one which will preserve the rights of both sides of the House in this peculiar position.
I beg to move,
That the words "or other person by whom the return is made, or position in establishment" in article 2, article 7, the words "and if there is a kitchen or scullery, whether it is used for meals" in article 9 (a), article 9 (b), (c), (d) and (e), article 10 (c) (iv) and article 11 of the Second Schedule to the draft of the Order in Council, entitled the Census Order
1965, which was laid before this House on 4th March, be approved.
The draft Order, which provides for the taking of a sample census of the population in 1966, has been laid before both Houses of Parliament, in accordance with the Census Act, 1920. The Act requires us to adopt this unusual procedure by which, although the whole Order is subject to negative Resolution, certain parts must be dealt with by affirmative Resolution of both Houses. These are the items of information which are wanted in the census returns but which are not specified in the Act. On this occasion there are nine such items. However, the Motion in the names of the right hon. Member for Bridlington (Mr. Wood) and the hon. Member for Bebington (Mr. Howe) gives the opportunity of debating the Order in full and, as it is convenient to take both Motions together, I should like to make some general remarks about the census proposals before turning to the particulars that are subject to affirmative Resolution.
Parliament has made it possible to hold a quinquennial census, but this is the first occasion on which the Government have taken advantage of this provision in the 1920 Act. Hitherto, Governments have been content with the traditional 10-yearly stock-taking. Today population changes are taking place so rapidly that the 10-year period leaves too wide a gap. About 10 per cent. of the whole population move every year to a new address, and with migration on this scale figures quickly lose their relevance to the problems of the day. There is therefore a pressing need for mid-decade statistics so that plans for social and economic developments for land use and for housing, for schools and for hospitals can be made on the basis of firm up-to-date statistics.
The strength of the case for a mid-decade census was recognised by my right hon. Friend's predecessor, and proposals for the sample census were announced at the end of 1963. We have had no hesitation in endorsing those proposals and allowing the project to go forward, for we accept that the census is vital to the formulation of sound social and economic policies.
The sample census has one enormous advantage over a full census—it is much cheaper. Although the saving cannot be proportionate to the size of the sample, because of irreducible overheads, the 10 per cent. sample proposed for 1966 will cost some £2 million as against £5 million for a full census. There is also the advantage of quicker results made possible by the reduction in the mass of data to be processed. The limits to the usefulness of a sample census are governed by the size of the sample and the consequential sampling errors, but these errors can be measured by the statisticians.
The proposed 10 per cent. sample will give reliable figures for the country as a whole and for the larger areas where the major problems of social and economic development are likely to arise before the results of the next decennial census could become available. For areas with less than 50,000 inhabitants —places smaller than, say, Guildford or Rugby or Clydebank, the figures will be of varying reliability, but they will still have value for many small towns. The census reports will explain how figures for these areas should be interpreted. It is also worth remembering that, even in the 1961 census, important questions on occupation, education, and migration were put to only 10 per cent. of the population.
While sample figures will meet the urgent planning needs which I have mentioned, there are some special areas in Scotland for which sampling would be unsatisfactory. There are six small areas with a population of less than 250,000 which are the subject of special economic and planning studies. The population in these areas will have to be counted in full. The sampling method cannot be used for them.
The sampling methods which are to be used in 1966 are designed to ensure that everyone stands an equal chance of inclusion—no more and no less. There will be no deliberate selection of particular individuals. In England and Wales, most of the sample is being taken from the 1961 census records supplemented by valuation records for houses built since the last census. In Scotland, the whole range of the sample will come from the valuation rolls.
A random starting point is taken in each block of records and from that point onwards addresses are picked on a strictly one-in-ten basis. Census enumerators will deliver forms of return to the selected addresses before census day. In private households, the head of the household will be responsible for filling in the form. In institutional premises—hotels, hospitals, barracks and so on—and in ships, the people in these institutions and the ships will themselves be responsible for filling in separate personal forms which they will receive from the manager of the institution, the skipper of the ship or whoever is in charge. Generally speaking, one tenth of the people in all such institutions will fill in a form, but in hotels, where sampling has proved troublesome in the past, everyone will have to take part.
There is one important proviso to these arrangements in England and Wales, where only one tenth of institutions expected to have less than 15 people staying in them will be visited. These small institutions will be treated like private households and the person in charge will make the return for everyone in the establishment.
The complexity of these sampling arrangements account for the fact, that while the draft Order before the House is similar in outline to the 1960 order, it is longer. It was because of the novelty and complexity of a sample census that the Registrars General found it necessary, in the spring of 1964, to hold a census test. This showed that the proposed sampling arrangements were satisfactory and that a sample census of the required accuracy was in practice feasible.
To give effect to these arrangements, the Order prescribes the persons by whom returns for the purposes of the census are to be made. The Order prescribes also the date on which the census is to be taken and the particulars to be stated in the returns. In addition, it lists the special areas in Scotland where there will be a full enumeration and, since the Order applies to Great Britain, my right hon. Friend the Secretary of State for Scotland joins with my right hon. Friend the Minister in submitting it to the House.
The date proposed for the census is Sunday, 24th April, 1966. A Sunday in April is the usual date for the population census. Sunday is the night in the week when virtually everyone is at home, and April comes just before the main holiday period. We feel that this is the correct date and we are carrying on the normal practice in the past.
The particulars which are to be asked for in tie census are listed in the Second Schedule to the draft Order. They will appear as actual questions in the forms of return which will be laid before the House by Regulation later this year. Many of the particulars are similar to those asked in 1961, but a few of the 1961 questions have been omitted, others have been amended, and some new items have been added. Generally speaking, however, the public is being asked to answer very much the same number of questions as were answered by 10 per cent. of the population in 1961.
Primarily, the questions are intended to meet the pressing need for mid-decade statistics showing the size and structure of the population by areas, the quantity and standard of housing and the composition and distribution of the working population. But they will also yield information about migration, transport and education. I should explain to the House that the list of particulars is the product of an extensive consultation with all the Government Departments involved, that is, all those which regularly use census statistics. There have also been full discussions with the local authorities, the universities and many other interested people.
So many suggestions were received that it would have been impossible to make room for all of them in the census. Nevertheless, the proposed list has gone a long way towards meeting the suggestions and it is fair to say that no field of census inquiry for which there was wide support has been neglected.
The first of the items subject to affirmative Resolution is that relating to people's position in an establishment. The Schedule to the 1920 Act includes a person's relationship to the head of the family but it is also necessary, in premises other than private households. to ask about people's relationship to the head of the establishment—whether, for example, they are staff or guests. This information is essential for the interpretation o analysis of the census returns and we think that no exception can be taken to it.
The next item deals with migration. The extent of population movement, both within the country and from without, is one of the factors that has made a quinquennial census necessary and there has never been any doubt that migration would have to be covered by the census in some form or other.
In 1961, people were asked to give their address of a year before census day. We have added a little bit here because we felt, after consultation with the various Departments, that that information would not be quite sufficient. So it is proposed to ask people not only for their address of a year before but also for their address of five years before
I make it clear that this does not mean that we want all the addresses if a person has moved five times in six years We simply want the address for one year before the date of census and for five years before. These questions will give, we believe, invaluable information about the extent and characteristics of population movement over the five years preceding the census and will help us assess, to some degree at least, any likely future trends.
The third item is about the use of kitchens. This is not included because of some curious interest of Government in the social habits of the population but because of the difficulties in deciding what should be counted as rooms in the census. Statistics about the number of rooms occupied by families are an important measure of housing standards.
As kitchens take many shapes and sizes, some rule must be evolved for their inclusion or exclusion from the count. In 1961, the criterion was whether the family used the kitchen for meals. If they did, the kitchen counted as a room. However, the information obtained was not altogether satisfactory because the decision was often governed by family habit rather than the size of the kitchen. It is proposed to overcome this problem in 1966 by counting a kitchen or scullery as a room if it is used for cooking, but comparison with the 1961 statistics will still be necessary and this can be obtained only by finding out whether kitchens are used for meals.
Does the question of whether a kitchen is used for meals include whether it is generally used for meals or only partly used for meals? Some people have breakfast in the kitchen but other meals in other rooms.
We shall use a fair amount of discretion in this. What we want to find out is whether the room or kitchen is used regularly for meals rather than for having snack meals at any given time.
The questions about tenure were asked in almost exactly the same form in 1961. They seek to establish whether people are owner-occupiers, or are renting accommodation from councils or private landlords, or are occupying it on some other basis. The questions about household amenities are also subject to affirmative Resolution. These are simply questions on the lines of those asked in 1961. In this census it is not proposed to ask about cold water taps and the questions are confined to sinks, cookers, lavatories, baths or showers and running hot water. The information, together with that about rooms and tenure, will illustrate progress in housing and supply analyses for use in formulating national housing policy.
The growing concern with traffic and transport problems and their implications for town planning explains the next two items. It is proposed to ask about ownership and use of private cars and whether they are garaged or parked at night, whether they are on or off the road, in a garage or in the open.
I come next to the questions about employment and economic activity. Only one item in this instance is subject to affirmative Resolution and that is the completely new question about means of transport on the journey to work. This question, of course, is related to other transport questions and should be particularly helpful in considering measures to deal with the ever-growing commuter traffic. The rest of the employment questions are not subject to affirmative Resolution, as I have said, and for the most part are similar to those asked in 1961.
However, they contain a novel question about people with more than one job. This is a factor of increasing importance in the country's economic life and the 1966 census for the first time will measure the extent of this factor on a national basis.
I come finally to the particulars about higher education. A question about scientific and technological qualifications was asked in 1961, but it was soon apparent that its mere repetition would not give us anything like the information now required for a study of manpower problems. There was a strong and wide demand by interested Departments and outside authorities for as much statistical information about educational qualifications as could be managed in the census. With many other urgent facts about the population, about jobs and about housing to be gathered, some limit had to be set to the questions about the numerous types of educational qualifications. What it has been found possible to include is a question on higher educational qualifications obtained since the age of 18. People will be asked to give details of their academic, professional and vocational qualifications, including teacher-training qualifications. From this mass of information statistics can be produced about people with higher educational qualifications in the field reviewed by the Robbins Committee, for scientists and technologists on a basis broadly comparable with the 1961 census, for people with higher degrees and for people with teacher-training qualifications.
I am very conscious of the need that the information yielded by these various questions should be available for use as soon as possible after the census. The sample should help to make this easier and it is the avowed aim of Her Majesty's Government to complete the publication of the main results within two years of the census day. I am glad to say that for this purpose the General Register Office will have the exclusive use of its own computer at its census office, and as this is the type of machine with which it is already familiar, it should be placed in far happier position than it was for the 1961 census. The 1966 results will, of course, be published in a series of reports, starting with county volumes and ending with national reports on the various census subjects—employment, housing migration, education, transport and so on. A short preliminary report will also be published within two to three months of the census, giving populations for the country as a whole, the regions and the largest cities.
Before I end, I should like to give the House and the public at large an assurance that the census returns are treated at all stages as absolutely and completely confidential. The names, which are essential for the interpretation of the returns, are lost in the very early stages of the processing, so that there is no question of the names going on and on and revealing themselves at the end. Information about particular individuals is not available to other Departments or, indeed, to anyone in the community.
We know that some people are worried about disclosing their personal affairs to the head of the household with whom they are staying, and because of this we provide in the Order for personal returns. This means that where people should be included on a return made by someone else, it is open to them to make a separate personal return should they so wish and the census officers and enumerators will be told how to arrange this. So that there will be no question of embarrassment to anyone who wants to make a personal return as distinct from being included in a return by the head of the household.
Similarly, in premises where people will in any case be making separate returns, we will arrange for people tow make their returns without the head of the establishment seeing them, if they are at all worried about this. I may add that never in recent censuses has it been necessary for anyone to complain of or to be prosecuted for a breach of confidence. The people can take it that any information which they give us will be treated with complete confidence, as has been the case in the past.
This sample census will be taken under the compulsory powers of the Census Act. There are penalties for people refusing to make returns or wilfully making false returns. It is not, however, my intention to lay stress on these compulsory aspects. Rather is it our policy to enlist the public's willing cooperation in this necessary and vital project.
The census figures, dry and formidable, have a vital part to play in determining the social and economic policies of the Government. This, I think, was one of the reasons why this proposal came from the party opposite as long ago as 1963. It will be a principal object of our census publicity to explain as imaginatively and helpfully as we can so that the man in the street will have an idea of the value of the census. By using all possible means of publicity, we look forward to winning the ready participation of the people who are selected in the sample. I am confident that the census plans are well founded and I trust that everyone will do his best to make this first sample census a great success.
The hon. Gentleman used the phrase "the man in the street". As a matter of interest, how is that expression interpreted? Is it the man on the Embankment, for example?
I am sorry that the hon. Member came in at a late stage of my explanation. I explained the proposal fully in detail. I said that we would take one Person in 10 of the population. It will be done solely on the basis of lists, valuation lists in particular. After the first person is chosen, every subsequent tenth person will be visited by an enumerator and handed a form. The procedure is clear and simple. There is no deliberate attempt to choose people. They will be taken purely at random. I hope that in the light of what I have said, the House will accept the Motion.
I should like, first, to convey to the House the apologies of my right hon. Friend the Member for Bridlington (Mr. Wood) for his inability to be here tonight, and perhaps my own apology for my inadequate presence in his place.
The House is grateful to the Parliamentary Secretary for the way in which he has explained in such detail the provisions of the Order which we are debating tonight. One cannot help observing that the representatives of the Liberal Party, who have seen fit to table critical Motions about the lack of attention paid to them on previous days, are not gracing us with their attendance to pay attention to the debate on this matter which affects the whole population.
As my right hon. Friend points out, it is not possible to take a 10 per cent. sample of that party, and we do not have even an 11 per cent. sample of the party in the House tonight.
We on this side of the House are glad that the Government are implementing the plans laid before the House in December, 1963, by my right hon. Friend the Member for Altrincham and Sale (Mr. Barber). It was acknowledged then, on both sides of the House, that there was a real need for greater frequency of information about the population, of the kind which the census is designed to obtain.
The House will be glad also that the Parliamentary Secretary has been able to give an assurance that the information from this census will be available within two years, as was promised in 1963, because among those who have to deal professionally and day by day with this information there is a great deal of anxiety at the delay, which has not yet ended, in the publication of the 10 per cent. sample results of the 1961 Census.
It may be that the hon. Lady, the Under-Secretary of State for Scotland, who I gather is to reply to the debate, will be able to tell us—and I am sure that we shall be glad if she can—when the rest of the 10 per cent. sample results will be published, because there is no doubt that in the whole of this statistical field speed is of the essence. It was Harold Macmillan who talked about the difficulties of working with last year's Bradshaw, and it is even more difficult if one is dealing with last decade's Bradshaw, which is what sometimes tends to happen.
We were grateful for the attention which the hon. Gentleman paid to the part that will be played in giving publicity to this census by all possible means available. I feel that it would be easy for this census to underestimate the misunderstanding, and even unpopularity, to which it might give rise. There is a certain biblical simplicity about a 100 per cent. census, when people understand that occasionally they have to stand up and be counted in the aggregate, but this census makes two changes.
It speeds up the frequency, so that it comes every five years, and it introduces an element of sampling, and when a man is asked "Do you eat your breakfast in the kitchen, or in the scullery", he is inclined to wonder why the other nine chaps next door are not being asked the same question. I hope that this matter will be borne in mind when the publicity is considered.
I think it is only fair and right that the hon. Gentleman should make it clear from his side of the House that he supports the need for the 5-year census. I know that he does support it, but he ought to make it clear.
I had hoped that 1 had made that clear, but I do so now quite gladly. After all, it was initially announced by this side when we were the Government, and we recognise the need for it entirely.
The point I am trying to make is that the public at large may not recognise the need for it as readily as they recognise the need for a decennial census. At the time of the 1961 census a pamphlet was produced, at a rather extravagant price, I thought, of 1s., called "Why a census?" This time another pamphlet, possibly two pamphlets, will probably have to be produced in rather more cogent terms answering the questions, "Why another census?" and "Why me?" to justify the sampling element and the increased frequency of it.
I was glad, too, that the hon. Gentleman emphasised, because it is important to the work of the census, the complete confidentiality of the information once it is received, but I wonder whether the whole exercise of Governmental census-taking need be, or need appear to be, so far as its documents are concerned, quite as fierce and forbidding as it is. The document itself with which one was confronted at the last census—I do not know what changes will be proposed this time—at first sight resembled a tax document or a summons. It had got in prominent type the fact that a £10 penalty was attached to failure to answer. Hon. Members will probably recollect the exercise carried out by my right hon. Friend the Member for Wallasey (Mr. Marples) when he was at the Post Office, when, I believe, he examined every document produced by the Post Office to try to reduce it into simple, attractive, persuasive English.
This is not in any sense a false point, or a point which is critical for the sake of being critical. I believe it may be of some importance that a market research questionnaire is couched in rather more attractive language and is generally more welcome to the average householder. Indeed, it is ironical, if we make this contrast, that those door-to-door salesmen hon. Members on both sides sometimes condemn find market research such a useful and apparently acceptable way of getting their feet in the doorway. It is a useful introduction to say that one is engaged in market research.
Certainly. I do not challenge that at all, but the fact remains that the housewife does not run away from a man who turns up on the doorstep and says, "I am doing some market research." He is welcomed, for the most part. There is this contrast. Why is it that market research in that sense is acceptable while the census is regarded as being to some extent an alarming manifestation of officialdom?
What I am suggesting is that it may be possible, though it is probably too late for this census, to diminish the official tone of it. If one looks at the language used on the form, one sees it is headed "Notice" in this rather fierce way. Even the paragraph about confidentiality is rather forbidding:
The person who is obliged in the prescribed manner to make a separate confidential return will not be liable to give information to the head …
a rather schoolmasterly touch about that—
… and in such case the person responsible for making the return will only be liable Lo interview in this respect
and so on. I should have thought one could look at the style of it to see if it could be put in a rather different way.
There is a great difference between a questionnaire which is designed to be presented by a market research interviewer who can make sure that the questions get clear answers, because the is himself able to present the questions in a clear way, and the kind of form to be filled in by a person by himself. In that form every word must be precise and mean precisely what it says and with no possibility of its meaning anything else. That is the essential difference between the technique of the interviewer and the technique of these forms.
I quite accept what the hon. Lady says, but it still seems to me to be no justification for not making the wording as intelligible and attractive to the householder as it is to the interviewer. If we compare the layout and the wording of the census form with that adopted, say, by the Social Survey, we find a great difference. The questionnaire in the Milner Holland Report was an interviewer questionnaire. It seems odd that the interviewer should be able to ask questions of the householder in this form:
Do you have your own fixed bath or shower? Have you hot water at your sink?
which are direct, crisp and intelligible questions, while the census form uses the words:
Has this household the use of the following in the building?
and so on. It seems to me that this could be looked at.
In the same connection, I wonder whether the samples which I gather have already been carried out, and are yet to be carried out in the future, extend to pilot tests of the wording and layout of the census form. I say, with the greatest respect for the expertise of the Registrar-General's Office, that something which has been running for a long time—like the forms used in the Courts of Justice—may be regarded as good and as not requiring a pilot sample survey simply because they have worked in the past. I know that the Registrar-General acknowledges the difficulties involved in some of the questions asked in the past. Would it not be worth considering—if it has not already been proposed—pilot testing the wording and the layout in a rather more intelligible form?
I want to say a word about the scope of the census. As the Parliamentary Secretary said, it contains very much the same questions as those that were asked in 1961. It is obviously desirable—and a welcome innovation—that we are to have questions about transport, in the light of the Buchanan Report on the growth of traffic, but I wonder whether a question confined to the main means of transport to one's place of work might not produce some misleading answers. In my constituency many people travel by bus to a ferry and then go by ferry across the Mersey. I would hesitate to say which was the main means of transport, because the bus may go longer in distance, but the Mersey Ferry may take a longer time. I would not like either the ferry or the bus to be overlooked because everyone opted for one rather than the other. I know that any change in this question would make it less simple to work.
It is obviously desirable to embark upon a survey of vocational qualifications in the general sense, because one of the things that we want to know is the precise number of teachers, nurses and other qualified people who are on the market, and about whom we have no certain knowledge. Since this is an extra census, and since the 10 per cent. sampling method makes it easier to get the results quickly, I wonder whether we might not be hanged for a sheep as well as a lamb and he willing to ask more questions, possibly over limited and selected areas.
The experiment is being made of 100 per cent. census in certain areas of Scotland. Would it not be possible or desiable, so long as it did not affect the object of tabulation at the end, to conduct a rather more exhaustive survey, in areas with large numbers of immigrants, as to language as well as country of birth which we shall get in the census?
We will get to know about this question of language. Where there are large concentrations of immigrants we intend to have forms printed in the languages peculiar to the countries of origin of the immigrants and to have interpreters. We will even let our Welsh people have a form in the Welsh language. We are dealing specifically with the question of immigration and are ensuring that we get the fullest information. We will employ interpreters in these areas and get full particulars of the country of origin and the language, and language difficulties as well.
I am grateful to the Parliamentary Secretary for that intervene- tion I became a little sensitive when he said that we shall deal "even" with the Welshmen. As an English-speaking Welshman, I may say that we have had our language inquiry in Wales, just as the Scots have had one in Scotland.
In the 1951 census, but not in 1961, I believe.
At all events, one can see that there is scope for being more adventurous in the number of questions which can be asked. Once this, to some extent, experimental quinquennial 10 per cent. census has been done, we can learn from it. I am sure that it will be borne in mind as a possibility for the future.
There is one last point which I should like to make. I know that, in considering the shape of the census, the Registrar General's office take advice not only from all the Departments of State, but from private interests outside—bodies such as the Market Research Society and the Institute of Practitioners in Advertising. One wonders whether, at some point in the near future, it might not be appropriate—because there are differences between the classification of social and economic classes used in the census and those used by private bodies—to set up a wide-ranging inquiry like that taken by the Government's inter-departmental committee on Government statistics, bringing in representatives of private market research, commercial and sociological, to survey the question of whether, as a nation, we are getting the right statistics at the right time, found in the right way and presented in the right way.
Certainly, to a layman, the book produced by the United States Bureau of Census for their 1960 census—a 700-page volume giving virtually a snapshot of every community in the United States —is an achievement which we have not yet matched up to. Nothing which 1 say should detract from the House's respect for the efficiency and integrity with which the census has been conducted in the past, and with which this one will be conducted by the Registrar General's office. We welcome this census as an essential and modernised tool of modern Government. We are grateful for the assurance that the results will be forthcoming with speed.
I hope that the Government will be able to take some notice of the points which 1 have tried to put about the way in which it should be presented to the public. On that basis, we welcome this Order.
The census is certainly a fine and vital tool for social policy, as my hon. Friend has pointed out. The proposed form of the new census, I believe, represents an improvement over its predecessors. The 10 per cent. sample is certainly very welcome. It is, as has been pointed out, cheaper and quicker, and we can get extra information because of this. We are already getting information through the census on cars, methods of transport and migration, but I feel that there are regions where we are possibly not getting enough information. We could get more if the census sample were smaller still. 1 believe that, on scientific grounds, we have good reason to believe that it can be considerably smaller than 10 per cent. and, therefore, even quicker and cheaper. If it were reduced to, say, 1 per cent., we could use professional interviewers and get, I believe, more accurate and more detailed information than we get by the present do-it-yourself forms.
I do not want to detain the House with a dissertation on the various sociological tools used by the demographers, such as the net reproduction rates and specific fertilities, but if we are to have really accurate projections of our future population, I think that we need much more accurate information about specifically how many children women are having, and the ages of the mothers when those children were born.
There was a great increase in the number of births in 1956–57, and I think that we tended to assume that this was due to a reduction in the age at which people were getting married. This may partly have been true, but one would have expected the increase in births at about that time to be levelling off, which does not appear to be happening. We seem to be getting an increase in the number of children per family, and it is generally reckoned that the average family now has 2·5 children compared with 2·3. This will have a tremendous impact on the number of school places, hospitals, jobs needed and so on in, say, 20 years' time.
I therefore appeal to my hon. Friend to consider this point carefully and to bear in mind that in America evidence has been forthcoming showing a close correlation between the intentions of married couples regarding the size of their families and the actual final size. If this is so, it provides us with means of projecting even further ahead our population estimates, and this would be invaluable.
If we adopted a much smaller census and had professional interviewers we would be able to investigate the intentions of married couples, obtain a correlation between their intentions and achievements and, if we may tread into this delicate matter, obtain a correlation between the methods used to achieve those ends and the efficiency of the results. Plainly, people cannot be made to answer such questions compulsorily, but in the vast majority of cases people are only too willing to answer if they know that their answers will be kept secret, as they always are.
I realise that these modifications cannot be introduced into the present census, but I urge my hon. Friend to bear in mind the importance of them and I hope that she will undertake to consider them and consult with the Registrar General on the possibility of such modifications in future censuses.
I must say, first, how pleased I was to hear the Parliamentary Secretary moving the Order because I seem to recall that the first occasion on which I met him was on his return from a visit to Scotland, having much more colour in his face than have those who live in the southern part of this Island.
Perhaps my only complaint with his remarks is that I doubted if he paid quite enough attention to the census being taken in Scotland, considering the 100 per cent. nature of it in certain districts there. Perhaps he laboured the claims of the 10 per cent. census in England a little, although I am sure that the hon. Lady the Under-Secretary will, when she replies, redress the balance.
It would be churlish of me not lo welcome the 100 per cent. census being taken in these districts because it is within my recollection—in what I might describe as those dear, dim days beyond recall when I was the predecessor of the hon. Lady—that certain papers to do with this matter came before me. Perhaps the hon. Lady will explain, for my benefit, if not for the benefit of all hon. Members, some of the reasons for the selective nature of the districts involved.
It seems perfectly clear that the list of towns within the County of West Midlothian will be involved in the development of Livingston. When mentioning Fort William the pulp mill immediately springs to mind. Then there is the development of the tourist industry in Kingussie and Badenoch. I hope that the hon. Lady will recognise that this all stems from the actions of the previous Government in securing development in these areas. No doubt that is the reason why she is anxious to have censuses taken in them.
Something which the Parliamentary Secretary said gave me a slight start and I hope that the hon. Lady will side with me on this issue, if on none other. I was surprised to hear him say that people living in Wales will have census forms printed in Welsh. Now, if this is so, and remembering the fact that Scotland suffered defeat at Murrayfield only three weeks ago solely because of the fact that the Welsh were allowed to sing "Land of My Fathers" before the match started, I hope that the hon. Lady will insist that in such places as Armadale and North Lorn, mentioned in Schedule 3, as well as the County District of Lewes, where there are numerous speakers of the Gaelic, we are on level terms with the Welsh.
It is all very well for the hon. Gentleman to come from the West Country of England and to put in this claim for the Welsh, to try to slip this one in, but I do hope the hon. Lady will make it quite clear that there will be an opportunity for those who prefer the ancient tongue of their ancestors to be able to have these forms in their own language.
I can assure the hon. Gentleman that the ancient tongue of my ancestors, who were around these shores 200 years B.C., is very much the same tongue as those of the hon. Gentleman sitting in that corner.
I am not absolutely certain what that means, but I will do my best to surmise it. My recollection is, and perhaps the hon. Lady will remind me, that when it comes to the borderline area the original idea was that there should be a 100 per cent. census taken only around the area of St. Boswells, where there has been some speculation that there might be a new town. I now notice under Schedule 3 that it is to apply to the whole of the County of Roxburgh. If there has been a change, it leads me to suggest that if depopulation is the reason rather than development in the other areas, why not go for the whole County of Peebles next door, where depopulation is a very much greater problem than in Roxburgh.
Are the premises in which workers, let us say, in a hydro-electric scheme in Scotland stay covered by Schedule 1? I imagine that the lodgings provided for them by the Board would not be covered by Group 1, which is:
Any dwelling (including a caravan)… separately occupied by a private household.
If they were to lodge altogether they presumably would not qualify as a private household. Are they covered by Group 2:
Any hotel, boarding-house or common or other lodging house."?
Where premises are provided by employers, do these come into the category of a lodging house? In Schedule 2, paragraph 9, it seems to me that there is a tremendous complication for the occupier here when he is asked to make a return, when sub-paragraph (i) says:
occupied by virtue of employment
and sub-paragraph (ii) says:
rented as part of, or together with, other premises…
Sub-paragraph (iii) says:
rented from a local authority or… the Scottish Special Housing Association otherwise than as described in sub-paragraph (i) or (ii).
I can see people scratching their heads and trying to work out whether they are "otherwise than" in (i) or (ii). Finally, in paragraph 9(e) there is a question about how many vehicles are owned and where such vehicles are usually kept at night. I suspect that the hon. Lady will get in ninety-nine answers out of a hundred the reply, "On the road", because as long as it is her policy and
that of her hon. and right hon. Friends to have the rents of garages in Scotland about three times those of the houses it is almost certain that the answer will be not the garage but on the road.
I congratulate the hon. Member for Edinburgh, West (Mr. Stodart) on bringing so many political points into discussion of a census Order. I am sorry that I shall not be able to satisfy him by following him into the by-ways and into the traps which he has laid for me in some of the answers which no doubt he would like me to give him. No doubt I shall disappoint him.
We are very glad that the hon. Member for Bebington (Mr. Howe) so warmly welcomed the Order and are grateful to him for his constructive comments on what my hon. Friend the Parliamentary Secretary to the Ministry of Health had to say about its contents. I shall try to deal with all of the great many and rather detailed points raised. I think that the best thing for me to do is to take them almost chronologically.
The hon. Member for Bebington asked when the rest of the reports were likely to come forward to us from the 1961 census. He will know of the ones which have been issued and, therefore, there is no need for me to go over them. Before the middle of this year we hope that reports will be published concerning usual residence, age, marital condition and general tables, and on housing, and before the end of the year the rest, including a miscellany of highly interesting ones. We are all looking forward to receiving reports on migration and on fertility, mentioned by my hon. Friend the Member for King's Lynn (Mr. Derek Page) before the end of 1965.
The hon. Member for Bebington also asked about public reaction to the one-in-ten sample idea. He asked whether the public would take happily to the idea that "I, Mr. Brown, am selected for this inquiry. Why is not Mr. Smith next door also selected?" The hon. Member has his psychology rather wrong. In my experience of observing the way in which social research and market research are conducted it seems to me that the public reaction is much more that of slight annoyance not to be the one in ten selected.
Will the hon. Lady recognise that she fortifies my point nine times? I visualised only one disgruntled customer. She visualises nine. Does not that increase the need for explaining the census to the public?
From a slightly different point of view, one is not trying to justify something one is imposing upon them, so one is encouraged in the whole process of educating the public in the use of inquiries of this kind and if one has an unsatisfied demand, one can use the emotion that unsatisfied demand has created quite confidently to go forward with further inquiries, if one wishes to do so, which is a very satisfactory position to be in.
The hon. Member, quite rightly, called attention to the nature of the document and the possible improvements that could be made in it. He quoted, for example, the phrase "head of household" and said that it was rather forbidding and cold. But it is very difficult to think of an alternative which would mean precisely what "head of the household" means. One cannot say parent, father, mother, grandparent, or substitute parent. None of these has quite the same meaning.
I do not criticise that term. I criticise the other phrases around it. I recognise the need for precision, but it is the somewhat legalistic style of the phraseology surrounding it that I am not happy about.
I recognise that, but the point is that, whatever term one examines, one usually finds that it is very difficult to find an equivalent which is not exposed to possible alternative interpretations. I can only assure the hon. Gentleman that there has been an effort made on the actual wording of the form. A pilot inquiry into the wording has been conducted.
On the question of the layout, I assure the hon. Gentleman that what he said will be borne carefully in mind. It would be excellent if we could move, even in a census form, towards a more attractive layout. If it can be done, it will be done.
The hon. Gentleman asked whether it was possible to ask even more questions, putting more questions, perhaps, over limited areas. The cost of a more extensive inquiry even in other limited areas tends to be very high. There is the cost of processing and dealing with the results one collects, apart from anything else. I suggest that, if one were wishing to go further into some of the specialised kinds of question he suggests, probably the better way of doing it would be by special inquiry among a special sample in a particular area or a selective sample of some kind rather than by seeking to obtain the information in respect of everybody. There is a strong argument for moving in the direction of more specialised inquiries rather than trying to contain everything in the census.
I am glad that the hon. Gentleman asked about the general use of statistics. It is, as he knows, a matter which my party gave a good deal of thought to over the last year or two from the standpoint of the need to collect information effectively so that it could be used effectively in the whole process of government. He will remember that my right hon. Friend the First Secretary of State, in reply to a Question on 8th December, said:
that is, the Departments—
are examining statistics to identify any important gaps in the data available and considering means by which they might be filled".—[OFFICIAL REPORT. 8th December, 1964; Vol. 703, c. 1319.]
This review is covering both economic and social statistics. I agree with the hon. Gentleman that it is an important prerequisite of good planning and decision making to know what is relevant, to know in what form one wants the facts, and just as important to know what facts one does not want, and to provide for an even flow of information. This is essential in any field where rapid social change means that new situations and new needs may be emerging almost without one recognising them. The work depends also on having enough professional "know-how" for guidance and advice. We are very much looking forward to the results of the review now going on. Clearly, it is a most valuable aspect of government.
My hon. Friend the Member for King's Lynn asked about smaller samples, possibly with the use of interviewers, which he thought might be quicker and cheaper. He spoke of the possibility of 1 per cent. samples, using interviewers. One has to be very careful here about the avoidance of bias, as my hon. Friend will recognise. The moment one lets any substantial bias creep in, one might as well not do it at all. Generally speaking,, the bigger the number of observations in the sample, the less chance there is of the grossed-up figures being subject to a large margin of error. If one were to use only a 1 per cent. sample in limited areas, there would be a great danger that the bias involved would be so considerable as to distort the results completely.
If my hon. Friend will consult the statisticians, she will find a difference of opinion on that. Indeed, the balance of opinion is to the effect that, if professional interviewers are used, there is great advantage in accuracy.
I agree that there is a difference of opinion. I know of no subject which absorbs the statisticians more than the study of bias, exactly what constitutes it and how to avoid it. I was about to say that, rather than the 1 per cent. inquiry of the census type, my hon. Friend is thinking more, I suggest, of special inquiries, using the interviewer. He is not really talking of modifications of this or any census as such. I agree entirely on that, and I think that for certain special purposes it would be very good to have a special inquiry using interviewers, in which case one could have much more elaborate information collected because it could be asked for and one could be sure that the interviewers were highly trained and knew what they were about.
My hon. Friend asked particularly for information about fertility, and I think that something ought to be said about this. As he knows, there are the annual statistics in the Registrar-General's Report, and the Annual Statistics of the Registrar-General for Scotland, and other statistics, but a closer analysis is only possible of attainment by way of the kind of questions—the block of questions—such as those included in the 1961 census; that is, fertility according to educational attainments, or social class, which are subjects of the greatest interest to social scientists, and there must be more information.
For example, there are such questions as the age at which the woman married; the age at which she bore a child, and the birth intervals for women of particular ages, and particular marriage durations. This information, as I have said, is at present based on the 1961 census, and the 1961 fertility report is due before the end of this year, and it is intended to consider the repeating of these questions in the 1971 census so that there will be a 10-year gap. But these questions take up a tremendous amount of space in the census form; large families have to be catered for, and something might have to be sacrificed in order not to make the form too unwieldy.
Obtaining this kind of information regularly is very necessary, particularly at a time of rapid social change, and one looks at the fertility statistics with great interest in order to see, for example, if there are differences in the new housing areas from other areas, irrespective of social class. Incidentally, I would say in passing that I know of no more fascinating source-book for the social scientist than the Report on the Family Census of 1946 on "The Trend and Pattern of Fertility in Great Britain".
The hon. Member for Edinburgh, West (Mr. Stodart) asked some provocative questions, including the production of a Gaelic form, and whether one was not entitled to presume that, since the Welsh people had a Welsh form, there should be a Gaelic form in Scotland. The answer to that, probably, is that of the 80,000 Gaelic speaking people, only 1,000 in 1961 were not bilingual. Where they are not bilingual, there is the need for a translated form, but the total of 1,000 is not sufficiently substantial to justify a Gaelic form. Nor has there been one for many, many years. It would be turning back the pages of history if we were to issue a Gaelic form. I understand that the reason why there is a Welsh form in Wales is because a higher proportion of Welsh people is not bilingual.
The hon. Gentleman also asked about the 100 per cent. districts in Scotland, as they are termed, and the basis on which they were considered. The answer is, not because of the introduction of more people because of the arrival of, for example, new industry, but because they are areas of depopulation. It is not growth, but depopulation, and we want to know about depopulation, and what it can show about internal migration within Britain. It is one of the aspects of statistical information about which least knowledge is established.
The Livingston area is an exception to these areas of depopulation. The Livingston area is selected because there is a new town development and because around it is a very sparsely populated area.
The hon. Gentleman asked about the particular part of the Borders which has been selected. Roxburgh County was selected as representing a sufficiently large border area for special study. But, broadly, the answer is that all these areas are areas of which special studies are being carried out and we could not, unless we had 100 per cent. enumeration, get reliable enough information. The population would be so small that there would be too much bias arising from 10 per cent. sampling. Workers living in lodgings as described will be included in Group 2.
I do not want to seem condescending, but I think that the enlightened interest shown tonight shows that the need for information has been accepted in these mid-1960s. Hon. Members have sought simply to improve the methods by which we gather this information. Questions have not concerned the impropriety of asking people for details that are necessary, but whether we should not ask for more information and improve the methods of gathering it.
This is an excellent indication of the trend of the times, when more and more people are recognising that a factual basis for social action is absolutely necessary if wisdom is to prevail. I am certain that the population as a whole will respond to the census in as enlightened a way as has the House. I think that the public likes learning about itself and enjoys the opportunity to give information which will allow the country to know more about its own people. That is the whole point of the census and it is on this basis that we make our decisions in this country. It is the correct basis and I am grateful to the House for the interest and support it has shown.
That the words "or other person by whom the return is made, or position in establishment" in article 2, article 7, the words "and it there is a kitchen or scullery, whether it is used for meals" in article 9 (a), article 9 (b), (c), (d) and (e), article 10 (c) (iv) and article 11 of the Second Schedule to the draft of the Order in Council, entitled the Census Order 1965, which was laid before this House on 4th March, be approved.
May I consult the situation? It would appear, as a corollary of the Resolution, that the hon. Member for Bebington (Mr. Howe) does not wish to move the Prayer.