 |  |
| www.dg.dial.pipex.com | 765 readers since 19 Aug 2007 |
| |
| www.dg.dial.pipex.com | 765 readers since 19 Aug 2007 |
Plowden (1967) Notes on the text Volume 2 Preliminary pages Foreword and Contents
The 1964 National Survey: Appendix 3 1964 National Survey
Appendix 8 Social services and primary education
Report (full text) about Plowden |
The Plowden Report (1967)
A Report of the Central Advisory Council for Education (England) London: Her Majesty's Stationery Office 1967
Volume 2 Appendix 10
1. In this Appendix we reproduce an abbreviated version of the First Report of the National Child Development Study (1958 Cohort). This multidisciplinary study of children born in 1958 has been financed, at the Council's suggestion, by the Department of Education and Science and other interested government departments in the belief that the data in the 1958 Perinatal Mortality Survey provided an opportunity for follow-up which should not be missed. It was hoped that a preliminary report would throw light on several of the major issues on which the Central Advisory Council were asked to report to the Secretary of State. 2. By any standards, the NCDS research team, directed by Dr M Kellmer Pringle and Professor Neville Butler, and led by Mr Ronald Davie, have succeeded well in providing us with an all-round picture of children at seven. Their success is all the more remarkable in view of the shortness of time in which they have produced their study. Unfortunately, there have been some disappointments - in the time available it was not possible, for example, to assess the effects of nursery education on children's subsequent performance at school. Other difficulties, particularly in the study of children settling in at school for the first time, have arisen from the fact that the study lay fallow for the seven years after the children were born - a reflection, perhaps, on the difficulty of getting funds for research. The risk of using data necessarily derived from parents' and teachers' recollection was bound to be present and was fully appreciated from the outset. This has been taken into account in interpreting the relevant results which, despite this inevitable shortcoming are, nevertheless, significant and valuable. 3. In the selection from the whole Report which is published here, two major excisions have been made. The full Report contains many tables of medical data. Some in the main text and most in the Appendix have been omitted for reasons of space. Secondly, the correlations between the data reported here and those in the 1958 Perinatal Mortality Survey had not been worked out at the time of publication and we are, therefore, unable to publish an interim report of them. We have also omitted the copies of the Questionnaires used by the Study. Our reasons for these omissions, apart from those of space, are that the whole of this First Report is to be published by Messrs. Longmans, Green & Company in January 1967, and those who wish to follow up medical data will find full details in that publication. 4. It will be clear to readers of this Appendix that the National Child Development Study promises to produce information of major importance to all who are interested in the educational, health and welfare services for children. We have already made use of their findings in our own Report. The NCDS research team will be the first to point out that these findings are preliminary, but they give a powerful indication at least of what more is to follow. The sponsoring bodies, the University of London Institute of Child Health, the National Birthday Trust Fund, the National Bureau for Cooperation in Child Care, the National Foundation for Educational Research in England and Wales, are to be congratulated on encouraging the production of what promises to be a valuable contribution to our knowledge.
(1958 Cohort) submitted to THE CENTRAL ADVISORY COUNCIL FOR EDUCATION (ENGLAND) APRIL 1966 Sponsored by: Institute of Child Health, University of London
in collaboration with: Association of Chief Education Officers (in England and Wales)
(including those sections omitted in this version. References to sections and tables will follow the full report) Research team
SECTION I: INTRODUCTION A. BACKGROUND TO THE STUDY 1. The 1958 Perinatal Mortality Survey
B. THE PRESENT REPORT SECTION II: DESCRIPTION OF THE SAMPLE A. THE POPULATION B. THE PRESENT SAMPLE C. INFERENCES MADE FROM THE SAMPLE D. SOURCES OF BIAS IN THE PRESENT SAMPLE DUE TO NON-RESPONSE SECTION III: METHODOLOGY OF DATA COLLECTION A. DESIGN OF THE STRUCTURED QUESTIONNAIRES B. THE EDUCATIONAL ASSESSMENT 1. The 'Educational Assessment' booklet
C. THE PARENTAL QUESTIONNAIRE D. THE MEDICAL QUESTIONNAIRE E. THE COMPLETION AND RETURN OF THE MATERIAL F. DATA PROCESSING SECTION IV: STATISTICAL ANALYSIS SECTION V: DESCRIPTIVE STATISTICS A. EDUCATIONAL FACTORS 1. Introduction 2. Reading ability
B. SCHOOL VARIABLES 1. Introduction
C. MEDICAL AND PHYSICAL FACTORS 1. Introduction
D. MEDICAL AND DEVELOPMENTAL SEX COMPARISONS 1. Introduction
E. ENVIRONMENTAL FACTORS 1. Introduction
F. BEHAVIOUR AND ADJUSTMENT 1. Introduction
SECTION VI: INTERRELATIONSHIP OF THE VARIABLES A. EDUCATIONAL FACTORS 1. Reading and arithmetic in relation to occupational group (a) Introduction (b) Occupational group and attainment in reading (c) Occupational group and attainment in arithmetic (d) Summary B. SCHOOL VARIABLES 1. Length of schooling, educational attainment and adjustment in school
C. ENVIRONMENTAL FACTORS 1. Parental situation
D. BEHAVIOUR AND ADJUSTMENT Adjustment in school in relation to occupational groups SECTION VII: INTERIM FINDINGS - SUMMARY AND DISCUSSION OF MAIN FINDINGS A. EDUCATIONAL, BEHAVIOURAL AND SOCIAL ASPECTS 1. Introduction
B. PHYSICAL AND MEDICAL ASPECTS 1. Scope and value of the medical history and examination
APPENDIX 1 List of Appendix tables
APPENDIX 2 The questionnaires
Research team Co-Directors: Professor NR Butler, MD, FRCP, DCH. (1) Mrs ML Kellmer Pringle, BA, PhD, Dip Ed Psych. (2) Principal Research Officer: R. Davie, BA, Dip Ed Psych. Senior Medical Research Officer: MJ Ball, BSc, MB, BS, DPH. Research Officers: Miss MJA Moncrieff, MA (October 1964 to November 1965); Miss M Levius, BA (from November 1965). Statistician: H Goldstein, BSc. (3) Administrative Secretary: Miss V Shenton. (1) Professor of Child Health, University of Bristol.
Acknowledgements A project, such as the National Child Development Study, would not be possible without the help and cooperation of a large number of individuals and organisations. The sponsoring bodies of the Study, the Co-Directors and the research team gratefully acknowledge and thank all those who have contributed to this enormous task. The Study has been carried out with the collaboration of the Association of Chief Education Officers (in England and Wales), the Society of Medical Officers of Health and the Association of Directors of Education (in Scotland), who have throughout given their advice and support. Acknowledgement and thanks are due to the Central Advisory Council for Education (England) and the then Welsh Department, Department of Education and Science, and to the Scottish Education Department, who wrote to all the Local Education Authorities; and to the Ministry of Health for circularising the Local NHS Executive Councils. Assistance was kindly given by the Senior Administrative Medical Officers of the Regional Hospital Boards, the staff of the Local NHS Executive Councils and also by many Children's Officers. The staffs of The Government Social Survey (Central Office of Information), the General Register Office and the MRC Medical Sociology Research Unit in Aberdeen have given valuable advice and assistance. The generous cooperation of the Chief Education Officers, Principal School Medical Officers and their staffs who were concerned with the administration of the Study in each local authority is gratefully acknowledged. They are warmly thanked for all the time and attention they have given to ensure the success of the Study. Special appreciation must also be expressed to the teachers, school medical officers and health visitors and school welfare officers in every part of England, Scotland and Wales, who undertook all the interviewing, testing and examining. Without their help, indeed, the Study could not have been carried out. The parents of the children are most warmly thanked for their cooperation and support. It is the second time that most of them have given valuable information about their children. The sponsoring bodies, the Co-Directors and the research team are glad to be able to take this opportunity of thanking the members of the Steering Committee, both collectively and individually, for their invaluable and unfailing help and support; and to acknowledge and thank the organisations and their representatives on the Consultative Committee. Acknowledgement and thanks for the interest and specialist help are due to many people, all of whom cannot be named, but among whom must be mentioned: Mr WB Barrett, Dr DE Cullington, Mrs J Davey, Dr JJB Dempster, Dr PA Gardiner, Professor M Healy, Professor R IIlsley, Miss HJ Lewin, Miss M Manning, Dr IC Monro, Dr J Morris, Dr RC Pearson, Mr TV Pretty, Mr AP Round, Professor M Quenouille, and Dr M Sheridan. Mrs V Booth and Dr DH Stott generously waived royalties on their test material, and the latter also gave further help. Special thanks are due to the staffs of the four sponsoring bodies for all they have contributed and for their patience and competence in the face of the frequent and very heavy demands made on them. Mention must also be made of all those temporary staff who for varying times have so enthusiastically helped with the vast amount of clerical and administrative work involved. The University of London Press kindly supplied the Bristol Social-Adjustment Guides and the Southgate Tests at cost; and the Ames Company gave Urine Test material for all the children concerned.
Foreword This Report, prepared for the Central Advisory Council for Education (England), is inevitably very much of an interim nature. The follow-through Study of the 1958 Cohort of children was not begun until October 1964, and the task of tracing the whereabouts of 17,000 children nearly seven years after their birth was itself immense and had to be completed substantially before the field work could begin in 1965. Not unexpectedly, the analysis of the data for this Report had to commence before information on all the children had come in. Hence, the work reported here bears upon only 11,000 of the children in England and consists of relatively straightforward analyses. It does, however, represent a considerable feat. The reader will be surprised at the width and richness of this harvest of facts - many of them barely surmised before - which have been obtained, analysed and presented in a total period of 19 months. The deadline for this Report could not have been met in this way without the fullest and most rapid cooperation of the schools, the parents, the medical and social services, without the organising skill and drive of the Co-Directors and, above all, without the devotion and competence of the very small research and clerical staff assigned to the project. Particular mention should be made of the Principal Research Officer, Mr R Davie, and the Statistician, Mr H Goldstein. The Steering Committee wishes to place on record its awareness of certain inevitable shortcomings - the absence of the more complex and powerful analyses, the fact that the references to previous literature have not all been followed up, and the marks of haste which will be apparent to the attentive reader. It also wishes to underline the achievement of the team in carrying through what to many of us experienced in these things would appear an impossible task in the time. WD WALL
I: INTRODUCTION The group of seven year old children featured in this Report are part of a larger national cohort in England, Scotland and Wales. The children were the subjects of a survey at the time of their birth; and information has now been gathered on many aspects of their growth and development. It is hoped that this study of the children can be continued at intervals throughout their childhood and, perhaps, into adult life. A. BACKGROUND TO THE STUDY 1. The 1958 Perinatal Mortality Survey In 1958 information was gathered on some 17,000 babies born during the week 3rd to 9th March in England, Scotland and Wales. This Survey (Butler and Bonham, 1963), sponsored by the National Birthday Trust Fund, reviewed the administration of British maternity services and the causes of perinatal death (still births and deaths in the first week). During the course of the investigation a vast amount of sociological and medical data were collected about the children and their families, including details of the parents' education and occupations; antenatal care; any obstetrical medical complications; the duration of the pregnancy; and the condition and weight of the babies at birth. This cohort of children can be considered unique for a number of reasons: it was a truly national series, selected only by date of birth; the very high proportion of returns (an estimated 98 per cent of all babies born during the week in question) reduced the possibility of any bias; and the comprehensive nature of the perinatal data was unparalleled in the world for any national cohort. 2. The National Child Development Study (1958 Cohort) In 1964 the opportunity arose to trace and study these children again, and the National Child Development Study (1958 Cohort) was set up for this purpose. The Study is financially supported by a grant from the Department of Education and Science, the Home Office and the Scottish Education Department, and is sponsored by the four bodies detailed at the beginning of this Report; one of these bodies, the National Bureau for Cooperation in Child Care, is responsible for its administration. The investigation is at present a three year project, but it is hoped that it will constitute the second stage of a longitudinal study of the original 1958 cohort of children throughout their childhood and into adult life. Thus, the aims of the Study can be divided into: short-term goals relating to the present three year investigation; and long-term aims which depend upon continued follow-up of the children. The former may be summarised as follows: (a) To study the educational, behavioural, emotional, social and physical development of a large and representative group of British children in order to gather normative data; to investigate the complex interrelationships between the many facets, both normal and deviant, of children's development; and to report the incidence of handicaps and the provision currently being made.The long-term aims of the Study are: (a) To explore the constancy and change in the pattern of children's development, longitudinally, and to investigate the associated educational, environmental and physical factors.These aims demand an inter-disciplinary research team and an integrated approach. The Co-Directors are an educational psychologist and a paediatrician: the former is also the Director of the National Bureau for Cooperation in Child Care, providing a direct link with the body responsible for the administration of the Study; and the latter, having directed the 1958 Perinatal Survey, establishes a link with the earlier work. An educational psychologist has overall charge of the research, and the staff further consists of a Senior Medical Research Officer - for a period of 18 months - a Sociological Research Officer and an Administrative Secretary. In addition, a statistician was seconded part-time to the Study and has been concerned with all aspects of the research. 3. The structure of the research The research team for the present three year project started their work in October 1964, with the exception of the Senior Medical Research Officer, who commenced three months later. The children in the cohort were at this time about 6½ years old and it was decided to complete as much of the field work as possible by the end of July 1965 for a number of reasons. First, the majority of the children would be leaving infant schools and classes at this date and it was obviously desirable, where a change of school was involved, to gather educational information from teachers who, in many cases, had had the opportunity of observing the children's development over a number of years. Secondly, a change of school would involve further administrative work in tracing new schools. Thirdly, the Study had undertaken to present an interim report to the Central Advisory Council by the spring of 1966, so that time was extremely short. Tracing the children was the first major exercise, and it would not have been practicable without the generous assistance of every local educational authority in the country. In response to a request from the Central Advisory Council for Education (England) and the then Welsh Department, Department of Education and Science, and from the Scottish Education Department, the authorities circularised schools, scrutinised their records and returned details of every child known to them born in the 'Survey week'. However, a sizeable minority of children were still untraced and a great deal of time and effort has been spent in reducing the size of this residual group in order to obtain a maximum possible return. Even more administrative work was involved in the task of identifying the perinatal records of the 15,300 children whose names were returned so that the data gathered in the Perinatal Survey could be 'linked' with current data. The invaluable help of the General Register Office must be acknowledged here. The Study was mounted in collaboration with the Association of Chief Education Officers (in England and Wales), the Association of Directors of Education (in Scotland) and the Society of Medical Officers of Health. With their support and advice, active cooperation was obtained from every local authority in gathering data on the children. The Chief Education Officers and Principal School Medical Officers kindly undertook to arrange for the completion and return of the questionnaires and tests used; and for the prior distribution of letters to schools and parents. The material consisted of an 'Educational Assessment' booklet and five tests (for schools); a 'Parental Questionnaire' to be completed usually by a health visitor, in an interview with the mother; and a 'Medical Questionnaire' for completion by a school medical officer on examination of the child. Further details about this material and the processing of the data is contained in Section Ill: 'Methodology of Data Collection'. Copies of the three questionnaires are in Appendix 2. The educational material was distributed to local authorities in March 1965, shortly after the children's seventh birthday, and the Parental and Medical Questionnaires were sent in the following month. B. THE PRESENT REPORT This Report, then, is of an interim nature, being produced about half-way through the span of the three year project. Further, it had in general to be confined to a consideration of the results for children in England because of the terms of reference of the Central Advisory Council. The data for the Scottish and Welsh children will, of course, feature in subsequent analyses. Most of the first 18 months was, of course, devoted to tracing the children and then identifying their perinatal records; deciding what kinds of information would be gathered and from whom; formulating hypotheses to be tested; designing the various questionnaires and selecting the most appropriate tests to be used; distributing material to all the local authorities; checking it for accuracy and completeness as it was returned; and transferring the data on to punched cards and then on to magnetic tape for computer analysis. More detailed accounts of some of these aspects of the project are given elsewhere in this Report. In order to produce a report for the Central Advisory Council to consider, it was necessary to impose a 'deadline' after which no material could be included for analysis. Thus, material received after the middle of August 1965 has not been utilised, but will subsequently be added to the earlier material which forms the basis of the present Report; all the data, including the information on children in Wales and Scotland, will then be analysed afresh for the final Report of the three-year project. The present Report, therefore, is based upon a sub-sample of the cohort of children. There is evidence of some bias in this sample of 'early returns', and this is fully discussed in Section lIB: 'Description of the Sample'. However, such bias as there is has been taken into account, where appropriate, in any interpretations made or conclusions drawn. The terms of reference of the Central Advisory Council and the great speed with which the present Report had to be prepared have both played their part in shaping it. In addition, tentative interpretations of the results have been made where more elaborate or more sensitive statistical treatment is indicated but has not yet been possible. Information obtained from this Study falls broadly into two categories. First there is a 'descriptive' element. A large amount of data has been gathered on a national sample of seven year old children. This information is of considerable interest and value in itself, providing as it does a picture of many facets of the children's education, growth, behaviour, health and environment. The results are presented and discussed in Section V: 'Descriptive Statistics'. Strictly speaking, however, this section is not of a purely 'descriptive' nature since, where appropriate, comparisons are made between the results for boys and for girls, so that hypotheses are being tested about possible sex differences. The second category of information is contained in Section VI: 'Interrelationship of the Variables', which deals with a few of the relationships between the many factors likely to affect children's development. Of course, the most interesting relationships are causal ones because these are most likely to have practical implications. If, for example, it can be shown that a particular circumstance, or combination of circumstances, is likely to lead to certain adverse effects, the way may be open for preventive measures. Even where prevention is impossible, a knowledge of causation may permit some amelioration of the effects. However, to demonstrate an association between factors is not necessarily to show a cause-effect relationship. For example, it is known that there is a relationship between children's height and their measured intelligence. But this does not mean that on the whole taller children are more intelligent than shorter ones because of their height. It seems probable that both of these factors are linked to others which form part of a complex web of predisposing conditions favouring both physical growth (in terms of height) and mental growth (in terms of intelligence). Great care must therefore be taken in interpreting demonstrable relationships. The factors involved are often complex and subtle, and it is extremely unlikely that simple answers will be found to those questions we most want answered. The analyses carried out and discussed in this Report are, then, in the nature of first steps along a difficult but challenging path towards a greater understanding of children's growth and development. Some of the results presented confirm those obtained by other workers in this field; some produce interesting new evidence; some point the way to further investigation; and some results have not been presented at all at this stage because of the danger of oversimplification or because it was considered advisable to await a more comprehensive analysis of all the material. Finally, there are, of course, many aspects of the Study which have not yet been examined for lack of time. References BUTLER NR and BONHAM DG Perinatal Mortality (E & S Livingstone Ltd) (1963)
II: DESCRIPTION OF THE SAMPLE A. THE POPULATION The population selected for the present Report consists of all children who were singleton births, living in England at the time of the survey, who were born between 3 and 9 March 1958, inclusive. (See Section III for a description of data collection.) This Study was designed so as to include the surviving children of the Perinatal Mortality Survey (Butler & Bonham, 1963) who were currently living in England. In addition, the population includes immigrants and some children who were born in Britain during 3 - 9 March but who were, for various reasons, not included in the Perinatal Survey. Where the population of interest is different from the above (e.g. children in local authority schools only), this is indicated in the body of the text. B. THE PRESENT SAMPLE The children included in the present sample are those for whom information had been received by the middle of August 1965. Since more 'Educational Assessment' booklets had been completed by this date, and since punched cards - containing most of this information - were available at an early stage, some of the analyses were completed using this data before all the information on each child was available for analysis. The total number of children for whom most of the educational data were available was 10,963; of these, 10,833 were known to be in maintained infant, junior with infants or all-age schools, or in independent schools catering wholly or mainly for children who are not handicapped; and this latter figure is the total involved in 'counter sorter' analyses using these cards. The number of completed Parental and Medical Questionnaires received by the middle of August was 7,985; in general, therefore, where tabulations were produced from these questionnaires using the counter sorter, this latter is the total involved. Where computer analyses were done, the totals for tables will vary according to the number of cases where particular information was unanswered. (See Section IIIF for the details of the data processing.) In addition to the above, 1,896 'late returns' had come in by the beginning of January 1966. Although it was not possible to include these 'late returns' in the main analyses, the Social Class distribution has been compared with that of the present sample. An overall test of significance indicates a difference in the distributions, although a test for trend of the proportions of 'late returns' in each Social Class category (Registrar General's classification) from Class I to Class V was not significant (see Table 1). Thus, although there does appear to be a difference between the two groups occurring in Social Classes II and V, there is no evidence that it involves a shift towards one end of the (nominal) scale. In most analyses, however, comparisons have been made within occupational group classifications, thus allowing for any bias arising in this way. The 1961 census figures on occupational classification are also presented in Table 1. It should be noted that the present survey is of seven year old children, whereas the census figures relate to a sample of households. Table 1 Occupational group distributions A further comparison has been made using sex, reading and adjustment scores, comparing the children for whom there are educational data but no parental and medical information with those for whom this latter information is also available. Significantly worse adjustment and reading scores were found among those children with the missing parental and medical information, but no significant difference was found in the proportion of the two sexes (see Table 2). Table 2 Southgate and Bristol Test Scores - with and without parental and medical information C. INFERENCES MADE FROM THE SAMPLE Inferences from the present sample are strictly applicable only to the (one week) population as defined above. However, it may reasonably be assumed that this population is a representative, although strictly speaking non-random, sample from a larger population consisting of children aged between 7 and 7½ years (the ages when the data were obtained) who were born during an (unspecified) period of time which includes 3 to 9 March, 1958. Thus, interest centres on inferences made about all children aged between 7 and 7½. Therefore, the assumptions have been made that secular trends have not taken place - or were negligible - during such a period and that the period is long compared to the one week of the sample; and for the purpose of testing hypotheses the sample has been treated as if it were a random sample from a population which, for practical purposes, can be considered infinite. Where it is felt that these assumptions are inadmissible, especially with regard to estimation of population means, this is indicated in the main text. It must also be emphasised that the basic sampling unit is the child. It is not possible, therefore, to use the data directly to estimate population parameters of distributions where the basic units are not individual children but, for example, households or schools. Thus, an estimate of the mean size of infant schools, calculated in the usual way from the sample, would yield an estimate too large since the larger schools contribute more children to the sample. All estimates which are given are therefore applicable only to the population of school children. D. SOURCES OF BIAS IN THE PRESENT SAMPLE DUE TO NON-RESPONSE In the present sample this matter has been investigated quantitatively in the comparison of 'early' with 'late' returns for occupational group, Reading and Social-Adjustment. Some of the possible sources of bias might, with advantage, be outlined. Four categories of children who will not be included can be distinguished. First, there were those whose parents decided not to participate. This category is small - a little more than one per cent at present. Secondly, there were the children whose home circumstances made it impossible for a parental interview or medical examination to be made; or whose parents simply failed to keep appointments. This group will contain a disproportionate number of children from large families, socially underprivileged homes; or parents who for any reason have little time to devote to their children. Thirdly, there were the children in the cohort who have not as yet been traced or, having been traced, moved to another area and have not yet been retraced. It is likely that in the majority of these cases incorrectly recorded dates of birth or other clerical error is responsible. However, this category will contain a disproportionate number of children whose families move home frequently. The fourth category includes children who have emigrated since the 1958 Perinatal Study. This will be relevant only to associations with retrospective data, including the perinatal material. References BUTLER NR and BONHAM DG Perinatal Mortality (E & S Livingstone Ltd) (1963)
III: METHODOLOGY OF DATA COLLECTION A. DESIGN OF THE STRUCTURED QUESTIONNAIRES It was decided to gather the information from three main sources: from schools, by means of a questionnaire (the 'Educational Assessment' booklet), specific tests and other assessments; from mothers, who would be interviewed by an officer of the local authority, usually a health visitor, using a structured questionnaire (the 'Parental Questionnaire'); and from School Health Services, who would undertake a medical examination, some special tests and complete a questionnaire (the 'Medical Questionnaire'). In considering the data to be collected, the following factors were particularly important: 1. The relative priorities to be given to contemporary and to retrospective data, having in mind that no information was gathered on the children during a seven-year period.It became apparent at an early stage that the material would have to be designed so that the data could be transferred to punched cards. Furthermore, it was decided to structure the questionnaires so that the answers were, wherever possible, in pre-coded form; i.e. the field workers in response to the questions were asked to ring a code rather than write in an answer. This reduced the need for coding the material when it was returned. Such a structure, of course, imposed limitations upon the kinds of questions that could be included and determined to some extent the way in which they were asked. It was further decided that the questionnaires should require no transcription sheets, so that the data could be punched direct on to cards from the forms themselves. Before the general layout was finalised, advice was sought both from the printers and from the commercial bureau who were to do the card punching. Amongst other factors which had to be decided were the size of print and paper, the spacing of the questions and hence the overall size of the documents. Important considerations here were general 'readability' and attractiveness of layout and, not least, the cost involved in terms of the weight and quality of paper. By and large, the individual experience, skills and qualifications of the members of the research team determined who did the actual drafting of particular items or questionnaires. However, since the approach throughout was inter-disciplinary, regular and detailed discussions on all the questionnaires took place, involving the Co-Directors and the whole research team. Advice and comment was sought from all members of the Steering Committee and some members of the Consultative Committee, as well as from countless other individuals - doctors, teachers, health visitors, school welfare officers, research workers - all experts or practitioners in their respective fields. It was possible to do some piloting on all the questionnaires, but pressure of time meant that we were unable to do as much of this as would have been desirable. For the most part the material took shape and was finalised as a result of exhaustive discussion and consultation. It was thought essential to ensure that anyone who was to use the material should know something about the Study. Each of the questionnaires, therefore, contained a brief summary of the background to the Study and its present aims. B. THE EDUCATIONAL ASSESSMENT A questionnaire was needed which would establish certain basic facts about the children's school environment, such as the type and size of school and size of class. It was thought desirable, too, to obtain some information about less tangible aspects of the environment, such as the contact between the school and the home, the 'social class' of the parents whose children attended the school and the basis on which children were allocated to classes. Finally, a fairly comprehensive picture was needed of the individual child: his abilities, his adjustment and behaviour and the interest and support given by his parents. Finally, six different forms of assessment were decided upon: 1. The 'Educational Assessment' booklet which established facts about the school and its organisation, the relationship between the school and parents, and which required assessments by the teacher of the child's ability and certain aspects of his behaviour.Piloting of the educational assessment indicated that completion for one child would take from one to one and a half hours. For more than one child an additional half to three quarters of an hour per child would be needed. However, most teachers would have only one child in their class who was taking part in the Study. C. THE PARENTAL QUESTIONNAIRE A questionnaire had to be designed which would be suitable for completion by an officer of the local authority, usually a health visitor, by means of an interview, where possible, with the child's mother or permanent mother substitute. Specially trained interviewers would obviously have been preferable, but this was too costly even to consider. It was soon realised that the problem of achieving a satisfactory compromise in terms of items which would be included was going to be most acute in this particular document. The chief difficulty lay in reconciling the need to obtain information about the child's present environment and development with the desirability of obtaining retrospective data over a seven-year period without over-burdening the informant and interviewer in terms of the time needed to complete the questionnaire. It thus became essential to establish some priorities in respect of items to be included. It was felt that basic contemporary data should have first priority since in general these would normally be more reliable (i.e. less subject to distortion or simple inadequacy of memory) than retrospective details. However, some kinds of contemporary data, such as the parents' religion, could equally well be ascertained in future studies of the children and was therefore given lower priority. With regard to retrospective information, a balance had to be achieved between the assumed importance of particular aspects in relation to the child's development and the extent to which any reliability could be attached to the mother's present report. Information about developmental milestones or pre-school difficulties are good illustrations of this difficulty. On the one hand, it was felt that such information would be valuable whilst on the other hand it seemed likely that mothers' memory of these events might not only be faulty - particularly if she had a large family - but, more important, might actively be influenced or distorted by the child's subsequent development. Information about the emotional and intellectual climate in the home, the parents' aspirations for the child and their general attitudes towards bringing up children would have been most valuable. Reluctantly, it was decided, however, either to omit or cover only sketchily these particular areas because of the limitations of time, the varying circumstances under which the interviews would take place and the varying experience of the interviewers. The Parental Questionnaire was considered to be the most suitable means of obtaining a full medical history from the mother. Not only would this save the doctor's time during the medical examination, but in many cases the health visitor might be acquainted with the health of the child, both in its pre-school and school days. It was advised, and proved possible in the great majority of cases, that the completed Parental Questionnaire be made available to the doctor in time for his examination of the child. For those instances in which this was not possible, part of the medical history, the systemically classified section, was reproduced in the Medical Questionnaire for completion or amendment at the time of the examination if this was thought necessary. Piloting of the questionnaires indicated that they would take an hour to an hour and a quarter to complete for an 'average' child. D. THE MEDICAL QUESTIONNAIRE A comprehensive medical examination of each child was considered to be an essential part of the follow-up of this cohort. Since the examinations were to involve large numbers of full-time and part-time medical officers of the local authorities, it was important to ensure as far as possible a uniform pattern of examination, as well as a standardised method of recording the findings. To give full clinical freedom in taking the medical history or in conducting the examination would create an impossible task in the classification and analysis of the data; at the other extreme, over-simplification leads not only to loss of information but also to a sense of frustration in professional field workers whose skills are not fully utilised. It is hoped that the method by which medical data were gathered for this Study avoided these two extremes by careful questionnaire design. The spectrum of information was wide, whilst individual bands within that spectrum were sharply defined. The pre-coded form of question was adopted to this end, and a questionnaire designed with sections classified in such a way that they matched the corresponding sections of the medical history in the Parental Questionnaire. To overcome the limitations imposed by the form of question used, the opportunity for written amplification of each answer was provided with each question. In the case of tests of function and the examination of the special senses, the conditions for the examination were set out in detail in order to ensure that the results would be as comparable as possible. The questionnaire included measurements of height, weight and head circumference: tests and assessments of vision, speech and hearing, including an audiogram; a urine test; tests of motor coordination and laterality; as well as a full clinical examination. Piloting indicated that, with the help of a school nurse, the whole examination, without the audiogram, would take 30-45 minutes. E. THE COMPLETION AND RETURN OF THE MATERIAL The material was sent in bulk to each local authority, which then made arrangements for its distribution, completion and return. In most cases, too, the authorities were able to give further help by checking the material for completeness before returning it. F. DATA PROCESSING The returned questionnaires and test forms were double checked by hand for completeness, coding errors, certain logical inconsistencies and accuracy of scoring. Where errors had occurred it was sometimes possible to rectify them by reference to other parts of the questionnaires. The data for each child were punched on to seven 80-column cards, the last of which also contained information from the 1958 Perinatal Mortality Survey for those children whose records had been linked. These latter data were reproduced from the 1958 Survey punched cards. The punching and verifying of the cards was carried out by a commercial bureau because of the size of this operation. The first card containing most of the educational data for each child was punched by the end of August 1965 and used for 'counter sorter' analysis. When all seven cards for each child were available in January 1966, they were loaded on to magnetic tapes using the IBM.1401 computer at Imperial College, London, and the data were edited on the IBM.7090 computer at Imperial College for incorrectly coded and mis-punched information. The edited magnetic tapes were then used for computer analysis on the 7090. This work was carried out using an adaptation of a program lent to the Study by the Government Social Survey (Central Office of Information). Due to technical difficulties which could not be resolved in the time available, data for some 300 children could not be loaded from the cards on to the magnetic tapes. Thus, where comparable analyses were carried out on the computer and also on the counter sorter, the totals differ somewhat. For the latter analyses, the cards (1 to 7) were divided into seven separate packs and straight counts and two-way tabulations produced from each pack. References GOODENOUGH F Measurement of Intelligence by Drawings (World Book Co.) (1926) SOUTHGATE V Southgate Group Reading Tests: Manual of Instructions (University of London Press Ltd) (1962) STOTT DH The Social-Adjustment of Children: Manual to the Bristol Social-Adjustment Guides (University of London Press Ltd) (1963)
IV: STATISTICAL ANALYSIS Due largely to the shortage of time, the statistical analysis has been restricted to providing estimates for means and testing hypotheses of a straightforward nature, using the methods of analysis detailed below. Since most of the hypotheses were concerned with associations in contingency tables, the basic method of analysis has been to apply an overall test of association. Although in some cases more detailed hypotheses have been formulated, it has not always been possible to use a more powerful procedure than the overall chi-square test. In other cases, however, particularly where sex differences are concerned and a natural ordering exists in the levels of the other classification, a test for a linear trend of the proportion in the resulting 2 x K table has been carried out, with integers I to K used as scores. This also furnishes a valid test of no difference between the mean scores of the two levels of the dichotomous classification (Armitage, 1955). This test is referred to as the Trend Test. It is also possible to obtain a test for departure from a linear trend. If no such departure exists, it may be inferred that the trend persists throughout the whole table and that the mean proportions in any two levels of the K-Ievel classification are different. In relation to sex differences, one may reasonably assume in most cases that the sexes are randomised over other factors and that any contrast between the sexes is orthogonal with respect to the other factors. One limitation of the present analyses is the absence of powerful tests for partial association in multi-dimensional tables, except in the special case of a 2 x 2 x K table (Birch, 1964). where the hypothesis concerns the partial association of two factors at fixed levels of a further factor and where the assumption of no second order interaction has been made; that is to say, the degree of association is the same in each constituent 2 x 2 table. Tests for second order interaction have been made (Plackett, 1962), and in no case has a significant interaction been found. In any set of analyses on one sample, the statistical tests of hypotheses will not generally be independent. This fact must be taken into account in the interpretation of any results. Furthermore, in a large group of independent tests, some are to be expected to show significance at the nominal level, even if the null hypotheses are true - purely by chance - and this has led to regarding the one per cent level rather than the five per cent level as the more appropriate one at which to begin rejecting null hypotheses, although significance values are shown at the five per cent as well as the one per cent and 0.1 per cent levels. Thus, the tendency has been towards a conservative interpretation of the results, which is also desirable in an interim report of this nature. The hypotheses to be tested had all been formulated in advance. This was done on the basis of past evidence with a view to confirming the results of previous research studies and experience and to answer certain basic but relatively straightforward questions. References ARMITAGE P 'Tests for linear trends in proportions and frequencies'. Biometrics II, pp. 375-386 (1955) BIRCH MW 'The Detection of Partial Association, I: The 2 x 2 case'. Journal of the Royal Statistical Society, Series B 26, pp. 313-324 (1964). PLACKETT 'A Note on Interactions in Contingency Tables'. Journal of the Royal Statistical Society, Series B 24, pp. 162-166 (1962).
V: DESCRIPTIVE STATISTICS A. EDUCATIONAL FACTORS 1. Introduction The data included and discussed in this section were gathered from schools. The topics dealt with centre on children's abilities and the question of backwardness and difficulties in school; the results for boys and girls are compared. It will be seen that the total number of children for whom results are given differs in various sub-sections. The reasons for this are outlined in Section lIB: 'Description of the Sample'. However, they might with advantage be restated in so far as they relate to the educational data presented here. Most of these data were recorded on one 80-column punched card. This card was available at an earlier date than the other six cards and it was therefore possible by sorting the cards (using a 'counter sorter') to obtain information and produce simple tables. The number of children in English schools for whom this card was available was 10,963. A breakdown of this number by the type of school attended is given in Table 3. It will be seen that 10,833 of the children were known to be in 'ordinary' maintained and independent schools, i.e. infant, junior with infants, and all-age schools and independent schools catering wholly or mainly for children who are not handicapped; 48 children were in special schools for handicapped pupils; and there were 73 children in schools which apparently did not fall into any of the categories listed. At a later stage, a re-scrutiny of the individual Educational Assessment booklets may make it possible to classify these schools. Table 3 Type of school attended In this sub-section, results are presented largely for children in 'ordinary' schools; thus, if the results were obtained by means of sorting punched cards, the total number of children involved is 10,833. At a later stage in the data processing, the information was transferred to magnetic tape for computer analysis. There were one or two minor difficulties at that stage which could not be resolved in the time available, so the data for some 300 children had to be excluded. Where the results are presented from computer analysis, therefore, the total number of children involved in 'ordinary' schools is 10,596. There is one further total which appears frequently in tables in this section where there is a comparison between the sexes. When the comparisons are one facet of more detailed computer analyses involving the occupational groups of the fathers, the figures for boys and for girls have been abstracted from larger tables. Where the occupational group of the father was not known at the time, it was not possible to separate the boys from the girls in these larger tables (not presented in this Report). Thus, it has only been possible to compare the sexes, if the figures were abstracted from larger tables, when the occupational group was known. For these tables, then, the results are presented for a sub-sample of 6,878 children in 'ordinary' schools. There is evidence that this sub-sample is biased in certain respects: better reading ability; better social-adjustment in school. However, there is no reason to believe that this bias will affect the validity of the comparison between the boys and the girls in that group. 2. Reading ability (a) Southgate Group Reading Test results This test is one which is primarily an assessment of word recognition. The results have been abstracted from a more detailed table, not presented in this Report, and are for children in 'ordinary' maintained and independent schools. They are given in Table 4 and in graph form in Figure 1. The superiority of the girls when compared with the boys is very clear. The difference in the distribution of the scores between the sexes is highly significant statistically. It will be seen that of the boys tested in this sub-sample only 1,202 (35 per cent) achieved a score of 28 or more, whereas 1,583 (approximately 47 per cent) of the girls did so. Further, the results of the statistical analysis confirm what inspection of the table suggest, namely, that there is a difference between the sexes at every level of reading ability, as assessed by this test. The overall superiority of the girls was not unexpected and confirmed the hypothesis which was postulated in advance. Table 4 Southgate Reading Test Score Table A1 of Appendix I gives a more representative picture when the results for boys and girls are combined and is based upon a larger sample of 10,596 children. One point is worthy of note in relation to the distribution of scores. The test clearly did not contain sufficient difficult items to 'extend' the more able readers. Thus, 47 per cent of the girls and about 35 per cent of the boys achieved a score of 28 or more out of a possible 30. This was expected and, as is explained elsewhere, this test was chosen partly because it was known that it would give good discrimination among the poorer readers. Figure 1 Southgate Group Reading Test scores by sex - children in ordinary maintained and independent schools (b) Stage reached in basic reading scheme In addition to the Southgate Test, information was obtained about the 'primer' or reading book in the basic reading scheme which the children were currently using. This item of information was gathered to supplement the results of the reading test and a rating of the children's reading ability by their teachers. Secondly, it was considered that the results obtained would be of particular interest to teachers, since the assessment of a child's reading ability in terms of the primer he is reading is a familiar frame of reference. Thirdly, the results would have possible implications for reading provision not only in infant but also in junior classes. Fourthly, it was felt that a comparison of the present findings with those of Morris (1959) in Kent schools would be valuable. It is recognised that there will be some variability in the level of difficulty of books in the most widely used reading schemes. Nevertheless, Morris found - on questioning experienced teachers - that, despite the variability, meaningful distinctions could be made between children who had reached different stages. Thus, 'it was customary for infant teachers to consider a first primer as dividing the poor and non-readers from the rest at the age of seven'. Children who by this age were reading Book 2 or Book 3 were at this stage where their reading was mainly mechanical; such children still needed a great deal of skilled help and encouragement for optimal progress. Pupils who had reached Book 4 or beyond were considered to have true reading ability; given continued encouragement and appropriate reading material they were, on the whole, capable of sustaining progress without the need for specific help from the teacher. (i) Comparison between the sexes The results obtained for children in 'ordinary' maintained and independent schools are given in Tables 5 and 6. The figures in the first of these two tables were abstracted from a more detailed table, not presented in this Report, in order to compare the sexes. Table 5 Stages reached in basic reading schemes The superior reading ability of the girls on this criterion is again very clear and highly significant statistically. This adds to the findings discussed in the previous sub-section in that the ability to read Book 4 or beyond will require not only word recognition but will, in the main, be accompanied by good comprehension of what is read. The statistical evidence also indicates that there is no departure from this overall tendency. Thus, the difference between the sexes is consistent at all levels of reading ability, as assessed by this 'primer criterion'. (ii) The present findings The figures for the sexes are not totalled in Table 5 to give overall results because of possible bias. However, figures based on a larger sample of 10,596 children are given in Table 6 and there is no expectation of bias here. They are presented together with results achieved by Morris in her Kent sample. Table 6 Stages reached in basic reading schemes The present findings, in so far as they represent the national situation in 1965, indicate that some 10 per cent of seven year olds in the final term of their infant schooling had still barely made a start with reading. A further 37 per cent had progressed beyond this stage but continued to need specific help. It has been acknowledged that this criterion of reading ability is relatively crude. Furthermore, the interim nature of the present report dictates a need for care in using precise figures. However, it is clear that a substantial proportion of children transferring to junior schools or classes has not reached a stage in reading where they can make optimal progress without continued teaching of the basic reading skills. Given the present age of transfer, it follows that teachers in charge of first-year junior classes should have a thorough knowledge of methods of teaching reading. Clearly this has practical implications for teacher training as well as for the staffing of junior schools and departments. (iii) Comparison with Morris's results It will be seen in Table 6 that there is virtually no difference in the proportion of children in the two samples who were on Book 4 or beyond. There may have been differences between the children of above average reading ability, but since this criterion gives no differentiation among this group, no conclusions can be drawn. The most striking difference between the samples is that whereas about 19 per cent of the Kent pupils had not progressed beyond a first primer, only 10 per cent of the children in the present sample were in this category. Interpretation of the difference between the present 1965 findings and those of Morris, obtained in 1954, in terms of a national trend, is not straightforward, although the difference between the samples is, statistically, highly significant. A number of factors must be considered. First, the average age of the Kent children when the assessment was made - at the beginning of their junior school course - would have been approximately 7½ years, slightly higher than that of the present sample. Secondly, it is very likely that the reading standard of these Kent children was above the national average, not only because of the generally higher socio-economic level in Kent, but also because Morris found on testing a sample of 10-11 year olds in that county 'that the reading attainment of Kent children at the end of their primary school course was above average for the country as a whole'. Against this, the present sample included a proportion (less than three per cent) of children in independent schools whose reading standard as a group is likely to have been higher than for the rest of the sample, whereas the Kent pupils were all in maintained schools. Thirdly, a small proportion of the present sample would no doubt have been using the 'initial teaching alphabet' (Downing, 1964) in learning to read, whilst none of Morris's sample would have done so. It is possible that this medium might alter the level of difficulty of primers and so change the nature of the criterion. Lastly, there were differences in the sampling technique used in the two studies. These complicating factors impose a need for caution in interpretation. Further, if the difference between the two samples does reflect a national trend, it provides no evidence for an increase in the proportion of good readers. This is not to say that such a change has not taken place; unfortunately, this 'primer criterion' does not discriminate amongst the children of average reading ability and above, i.e. those who are reading Book 4 or beyond. Nevertheless, the difference between the proportion of children in the two samples who had not progressed beyond Book 1 is very marked. The tentative conclusion seems warranted that in the country as a whole the number of poor and non-readers transferring to junior schools and classes has dropped in the interval from 1954 to 1965. (c) Teachers' ratings of reading ability A third assessment of the children's reading ability was obtained in the form of a rating by the teacher on a five-point scale. In an attempt to increase the comparability of ratings made by several thousand different teachers, three steps were taken. A distribution of ratings was suggested which the teachers were told might be expected in a representative cross-section of children of this age; the teachers were asked to rate the child 'in relation to all children of his age (i.e. not just his present class or, even, school)'; verbal descriptions were given for each of the five possible ratings. These verbal descriptions are detailed in Table 7, together with the results for boys and girls. Once again the figures are abstracted from a more detailed table, not presented in this Report, but the combined results for the sexes on a sample of 10,833 children in 'ordinary' maintained and independent schools are given in Appendix 1, Table A3. Table 7 Teachers' ratings of reading ability Once again the superior reading ability of the girls in this sample can be clearly seen and is statistically highly significant. There were more girls (approximately 39 per cent) than boys (about 26 per cent) rated as of above average ability; also, there were fewer girls (approximately 17 per cent) than boys (approximately 31 per cent) below average. The difference between the sexes is so marked, and confirmed by the results of the other two assessments of reading ability, that the firm conclusion is warranted that at this age girls are superior to boys in all aspects of reading ability. 3. Ability in number work (a) Problem Arithmetic Test Results There was some difficulty in deciding upon an appropriate objective assessment of the children's ability in number work. A test of mechanical arithmetic could have been devised or selected; but this, it was thought, would hardly do justice to the wider range of activities to be found in many infant classes. On the other hand, a test of mathematical concepts might have been time-consuming or difficult to administer. It was finally decided to devise a short test of problem arithmetic, thus avoiding some of the restrictions of purely mechanical calculation and assessing to some extent the ability to apply arithmetical knowledge to problems appropriate to this age group. The test was devised especially for use in this Study. There were ten problems in all and they were presented in the estimated order of difficulty. Six of the problems had been used before by the National Foundation for Educational Research on a large sample of seven year old children. Information was therefore available on the probable level of difficulty for these questions. The original intention had been to make this an oral test to avoid penalising those children who could not read fluently. However, in doing this it was possible that some children would be put at a disadvantage because of difficulty in remembering questions accurately whilst calculating the answers. Thus, it was decided to present the problems in printed form but also to ask teachers to read the questions one at a time to the children (repeating where necessary) and allowing as much time as was needed for answering. If a child because of some disability was unable to write his answers, the teachers were asked to record them for him. The results for a sub-sample of boys and girls in 'ordinary' schools are given in Table 8. They were abstracted from a larger table not presented in this Report. The combined results for the sexes taken from this same table on a sample of 10,596 children are given in Appendix I, Table A2. Table 8 Problem arithmetic results There was a highly significant difference between the distribution of scores for the sexes, strongly indicating that boys are superior to girls in the particular aspect of arithmetic ability assessed by this test. The statistical analysis provided no evidence of any departure from linear trend, indicating that this difference between the sexes persists over the whole range of test scores. (b) Teachers' ratings of number work As with reading ability, a subjective rating was obtained from the teachers of the children's ability in number work. The introduction to all the ratings was the same and has been outlined in sub-section 2(c) of this section. Once again, verbal descriptions were given for the five points on the rating scale. It will be seen from these descriptions shown, together with the results in Table 8, that stress was laid upon an assessment of the children's insight and grasp of new processes rather than their ability to calculate accurately. These results were obtained from sorting the punched cards so that the total number of children involved is 10,833 (see Introduction to present section). Table 9 Teachers' ratings of number work The statistical test used rejects the hypothesis that there is no difference between boys and girls in their number ability as rated by their teachers. There was a significant overall tendency for the boys to receive higher ratings. However, the highly significant departure from linear trend shows that this tendency was not consistent at all levels of number ability. Inspection of the table suggests that the reason for this departure from linear trend is that the proportion of boys and girls rated below average is virtually the same, approximately 36 per cent; on the other hand, about 22 per cent of the boys were rated above average, whilst only 18.5 per cent of the girls were placed in this category. The conclusion seems warranted that within the framework of ratings which lay stress upon insight and understanding in number work, teachers feel that boys of this age show more evidence of above average ability, whereas there is little or no difference between the sexes in the proportions which are below average. The difference between this finding and the analysis of the Problem Arithmetic testing, which produced no evidence of any differential results above and below average for boys and girls, may be due to the different nature of the two assessments or to different aspects of number ability being assessed. Further research should throw more light on this matter. However, the results of both assessments indicate that there are more boys than girls of above average arithmetic ability. 4. Other abilities (a) Introduction In addition to assessments of number ability and reading, it was important to obtain a more complete and rounded picture of the children's abilities for a number of reasons. Thus, it is hoped at a later date to combine the assessments of other abilities in order to obtain a measure of the children's general level of intellectual functioning. Such a measure will make it possible, for example, to contrast those who are making slower progress in school than would be predicted from a knowledge of their general level of ability with those who are making better progress. Further, it will make possible analyses of the children's attainment when some allowance or correction has been made for those abilities less directly influenced by teaching in school. An additional reason was that, although there is little published information about children's progress in reading and arithmetic in infant schools, there is even less available about other abilities. It would have been possible, by adding an intelligence test to the range of objective tests included in the assessment of the children's development, to obtain a measure which might have fulfilled some of the functions outlined. However, intelligence tests which could readily have been administered by the teachers tend to have limited reliability at this age; further, it was felt that to add another test would have been an unreasonable imposition upon the teachers' time in the context of an educational assessment which was already to take more than an hour for an individual child. It was decided, therefore, to obtain ratings by the teachers of each child's 'oral ability', 'awareness of the world around' and 'creativity'. These areas of functioning, it was felt, were very relevant to children's progress in school; they were aspects of development which schools would be attempting to foster; at the same time, they were likely to be influenced by factors outside the school to a greater extent than were reading or arithmetic. The shortcomings of ratings were fully appreciated; and the attempts made to reduce subjectivity have been briefly outlined already (see sub-section 2(c) of the present section). Time has not yet permitted any 'pooling' of the ratings to obtain some more general measure; moreover, particular care will be needed in devising a system of weighting to take account of the different distributions of the three ratings. In the following three sub-sections, the results for boys and girls are compared. The hypothesis tested in each case was that there would be no difference between the sexes. (b) Oral ability The ability to express thought and meaning orally is of obvious relevance to educational progress. It cannot be equated with verbal ability, which embraces the capacity for understanding and dealing with verbal material of all kinds; nevertheless, these two abilities will be highly correlated. Since a rating of verbal ability would be influenced to some extent by a child's performance in reading, it was decided to obtain an assessment of the children's oral ability, as it manifested itself in conversation and in the normal verbal inter-play of the classroom. Table 10 Teachers' ratings of oral ability There is a highly significant difference between the sexes, the girls receiving more favourable ratings than the boys. However, there was a highly significant departure from linear trend, indicating that the 'gap' between the sexes is not consistent at all levels of oral ability. The results suggest that there is a greater difference between the sexes below average than above. It will be seen that about 24 per cent of the boys were rated above average compared with 28 per cent of the girls; on the other hand, approximately 25 per cent of boys were rated below average, against 17 per cent of the girls. It is likely that the superiority of the girls in this context is to some extent a reflection of their tendency to be more forthcoming, orally, at this age. (c) Awareness of the world around Some may question the inclusion of this rating in a consideration of abilities on the grounds that what was being assessed was knowledge rather than ability. It could be argued that the acquisition of knowledge is dependent upon this knowledge being made available to a child and therefore it is not a 'true' ability. However, if 'awareness of the world around' is regarded as a child's ability to comprehend a complex world, the distinction is surely a fine one. It may be that this comprehension and the background of general knowledge which accompanies it is more dependent upon environmental factors than are many other facets of intellectual functioning; but this is a matter of degree rather than kind. In any event, there can be little doubt about its relevance to and association with educational progress. The results for children in 'ordinary' maintained and independent schools are given in Table 11. Table 11 Awareness of the world around The statistical test for linear trend shows a highly significant tendency for boys to be given higher ratings for this factor. Again, however, there was a highly significant departure from the linear trend. The results indicate that whereas a higher proportion of boys than girls are felt by their teachers to have above average 'awareness of the world around' - the proportions were about 26 per cent and 19 per cent respectively - there is little or no difference in the proportions below average. In fact, what small difference there was in this latter category in the present sample was in the girls' favour. (d) Creativity There has been increased interest in recent years in the assessment of creativity, particularly at the secondary school level. It has been claimed that it is possible to measure and distinguish creative thought processes from the kinds of ability assessed by conventional tests of intelligence. No attempt is made in the present Report to produce any fresh evidence on this topic; indeed, it is open to question whether such relatively crude measures as ratings could do so. Nevertheless, within the context of the assessments made, it was felt that such a rating might prove a useful and interesting addition. One of the principal difficulties in assessing creativity is its multi-dimensional character, and it can operate in many different spheres. One can opt for a clearly defined area of creativity as, for example, in free written expression. However, to do so is to penalise those children with reading difficulties as well as to exclude those whose creativity manifests itself in other activities. To ask for a more global assessment carries the risk of reduced reliability and validity, since the verbal descriptions of each point in the rating scale must be rather more general in nature. One way to escape from this dilemma is to ask for ratings of creativity in a number of well-defined fields. However, this would have been a further imposition upon the teachers' time, which was not felt to be justified. It was finally decided to obtain a global assessment, but to stress the generality of the rating by giving examples of the activities in which it was anticipated that creativity would be shown. In the Educational Assessment booklet, therefore, the heading 'Creativity' was followed by: '(e.g. in free writing, telling a story, handwork, painting, drawing, dramatic work)'. The results are shown in Table 12. The difference between the distribution of rating was statistically highly significant. The result indicates that, as assessed by this rating scale, girls of this age are felt by their teachers to show more evidence of creativity in school. It will be noted that there was no statistically significant departure from this tendency, suggesting that the difference between the sexes is consistent at all levels of 'creative' ability. Table 12 Teachers' ratings of creativity 5. Backwardness and difficulties in school (a) Children receiving special educational help in ordinary schools Within the framework of the 'ordinary' school, the term 'special educational help' is here preferred to the more widely used 'special educational treatment' . Children needing such help come within the broad definition of handicapped pupils categorised as 'educationally subnormal' (HMSO, 1953), namely: 'pupils who, by reason of limited ability or other conditions resulting in educational retardation, require some specialised form of education, wholly or partly in substitution for the education normally given in ordinary schools'. This definition and its implications were further elaborated in 'Special Educational Treatment' and in 'Slow Learners at School' (HMSO, 1946 and 1964). It was made clear that the definition is interpreted broadly and covers every kind of educational provision for children who 'need special help', even if this is only 'in certain parts of their work'. It was estimated in the former publication that about 10 per cent of the school population would fall into this category; of these, some eight or nine per cent would be catered for in ordinary schools. These estimates related to 'registered pupils over the age of seven' but, in fact, excluded children of this age who were in infant classes; indeed, in a footnote it was indicated 'that special educational treatment is not normally required for infants unless they are so seriously retarded that they should attend a special school'. The statement is not in line with some more recent thought on this question; thus, 'a crucial time for getting to grips with backwardness is the last year of the infant school and the beginning of the junior school' (Tansley and Gulliford, 1960). In order to identify children with learning difficulties and to throw some light on the current situation in infant classes, it was decided to ask the head teachers of the children in the present cohort whether the children were currently receiving any special educational help in school; and, if they were not receiving it, whether they would benefit from such help. The first question was phrased: 'Apart from anything which the class teacher may be able to do in the normal way, is the child receiving any help within the school because of educational or mental backwardness?' Of the 10,833 children in ordinary schools for whom the information is available, approximately seven per cent of the boys (379) and four per cent of the girls (205) were receiving this help. The total proportion of children was thus over five per cent (584). The difference between the sexes in this respect was highly significant (p<.001). Of those who were not receiving any help, the head teachers were asked if they considered that the children 'would benefit from such help within the school at the present time'. A further 879 children (eight per cent) the head teachers considered would benefit. Of these, 541 (10 per cent) were boys and 338 (six per cent) were girls. The difference between the sexes was again highly significant (p<.001). It is particularly important that the figures relating to the second question be viewed in relation to its precise wording. The term 'would benefit' (i.e. from special help) was used; the corresponding term in the definition of educational subnormality, quoted above, was 'require'. The former term was considered to be more acceptable to head teachers in the context of the present study: it would minimise their natural reluctance to 'label' a child prematurely; and it would afford them greater freedom to express an opinion which had regard above all else for the needs of the individual child. Substitution of the term 'require' would, it is felt, have reduced the number of children thus classified by schools. In summary, more than five per cent of the children were already receiving special educational help in infant classes and there was a further eight per cent who, it was considered, would benefit from such help. The fact that more than 13 per cent of this sample could, with advantage, have been given such help stands in such marked contrast to the opinion expressed in the publication mentioned above (HMSO, 1946) that there is clearly an urgent need to re-examine this issue. The question of provision at the junior stage is discussed in sub-section (c) of the present section. (b) Children in need of special schooling The figures given at the beginning of this main section show that 48 (0.4 per cent) of the children were in special schools. It is likely that some of the 73 children in 'unclassified' schools were also receiving special schooling, but it is not possible to determine this at present since time does not permit individual re-scrutiny of the Educational Assessment booklets. Of the children in ordinary schools, the head teachers were asked: 'Do you consider, irrespective of the facilities in your area, that the child would benefit now from attendance at a special school?' In their view, two per cent (219) of the children would have benefited; 2.6 per cent of the boys (144) and 1.4 per cent of the girls (75) were included in this category. The difference between the sexes was highly significant (p<.001). In terms of the total number of children who might have benefited from special schooling at the infant stage, the above must be considered as minimum figures; for a further 190 children (1.8 per cent) the head teachers felt unable to give a definite answer. This latter group would obviously have contained a proportion of 'borderline' cases. In interpreting the situation,. one must again weigh the effect of the term 'would benefit' used in this question. Nevertheless, it is felt that an affirmative answer about the need for transfer to a special school would not have been made by head teachers in respect of children still in infant classes without considerable forethought. The results would appear to reflect a need felt by the head teachers for earlier transfer to special schools than is currently the practice. (c) Children likely to need special educational treatment in future The term 'special educational treatment' is here used to denote special schooling as well as special educational help within an ordinary school. As a separate question, the head teachers of all the children currently in ordinary schools were asked: 'Do you consider, irrespective of the facilities in your area, that the child is likely to need some form of special schooling or other special educational help within the next two years?' The number of children thus classified was 530 (approximately five per cent); 348 (6.3 per cent) of the boys were included in this total and 182 (3.4 per cent) of the girls. Once again, the difference between the sexes was highly significant (p<.001). There were a further 381 children (3.5 per cent) whose head teachers were not able to express a definite opinion as to their future needs. Again there were more boys than girls: 243 (4.4 per cent) and 138 (2.6 per cent) respectively. This difference, too, was highly significant (p<.001). It seems likely that the majority of these children would be 'borderline' cases. The overall position is thus as follows: (i) 48 children (0.4 per cent of the sample) were already in special schools.If one totals the proportions of children in items (i) and (iii), and makes the assumption that the majority of the children in item (iv) were 'borderline' cases and approximately half of them would in fact need help, then one would conclude that about seven per cent of the present sample were likely to need some form of 'special educational treatment' between the ages of seven and nine years. The assumption that the 3.5 per cent in item (iv) were 'borderline' cases is supported by the fact that three quarters of these children were rated as poor or non-readers by their teachers. A detailed scrutiny of the questionnaires at a later stage should enable a more precise estimate to be made. However, at present there are three factors which prompt one to feel that the above estimate of seven per cent should be regarded as a minimum figure. First, the fact that head teachers were asked in this question to predict the children's likely needs over a period of two years may have led to some reservations and a cautious approach which gave the child the benefit of any doubt. Secondly, the possibility must be considered that the answers may have been influenced by the presentation of the question. The head teachers were asked: 'Do you consider, irrespective of the facilities in your area, that the child (a) would benefit now from attendance at a special school' (already discussed in sub-section (b)); and '(b) is likely to need some form of special schooling or other special educational help within the next two years?' It will be seen that the juxtaposition of the two questions, the order in which they were asked and the order of the two alternatives in the second question may have predisposed head teachers to have the possibility of special schooling uppermost in their minds. Further, the nature of the 'other special educational help' was not made explicit, which may have led some head teachers to have in mind a narrower concept of possible provision than is now embraced by the term 'special educational treatment'. Thirdly, the evidence from the head teachers' replies to the question relating to the children's current educational needs - discussed in sub-section (a) - suggests that the above estimate of seven per cent should be seen as a minimum figure. It will be recalled that some 13 per cent of the children in the head teachers' opinions could, with advantage, have been given special educational help in ordinary schools. Although it is no doubt true that some of these children's need for help might have diminished or even disappeared by the time they transferred to junior classes, it is hardly credible that the proportion would have dropped so markedly - from 13 per cent to seven per cent. One reason for this changed picture is likely to be the use of the term 'benefit from' (special help) in the earlier question and 'need' (special educational treatment) in the later one. It is arguable which of these terms best reflects the actual needs of children. To summarise, then, it is estimated that at least seven per cent of the children in the present sample were likely to 'need' some form of special educational treatment whilst they were from seven to nine years old. It is difficult to compare this figure with the estimates of the incidence of children requiring 'special educational treatment', made by the Ministry of Education (HMSO, 1946), since these estimates give no indication of the incidence of multiple handicap. In particular, the definition of educationally subnormal pupils, quoted earlier, if strictly interpreted could include many of the children with other handicaps. In view of these complicating factors, all that can be said at the present stage is that our findings do not provide evidence on whether the figures given by the Ministry under- or over-estimate the position. There is strong evidence from the present Study to suggest that the phraseology used in defining children in need of special educational treatment is more crucial than may be realised. (d) Children referred to outside agencies The original reasons for including a question about referral to outside agencies were fourfold: to discover which children had been referred so that a more detailed study of the causes of referral could be made; to study the extent to which various agencies had been used; to discover the overall proportion of children who had been referred; and to test the hypothesis that more boys than girls would have been referred. Although identification of the children concerned is straightforward, a detailed study of the causes of referral and the use made of various agencies necessitates perusal and analysis of the individual 'Educational Assessment' booklets, and there has not yet been time for this. However, results are presented on the second two points. The question was put to head teachers as follows: 'Has the child, because of difficulties which have affected his progress or behaviour in school, been referred to your knowledge to any agency? (e.g. School Health Service, Child Guidance Clinic, School Psychological Service, Education Welfare Service or School Attendance Officer, Children's Department, General Practitioner, Private Specialist.) (Include referrals made at a routine medical examination, and any made by another school or by the parents, if known.)'Where the answer was 'Yes', the head teachers were asked to state the agency/agencies involved and, briefly, the reasons for referral. The number of children in the sample reported to have been referred was 1,127 (9.5 per cent). It will be evident that this should be seen as a minimum figure since there would have been a number of children falling into this category where the information was not available to the head teachers for various reasons. Approximately 11 per cent of the boys (604) and eight per cent of the girls (423) had been referred. The difference between the sexes in this respect was highly significant (p<.001). This result, then, confirmed the hypothesis previously postulated. 6. Summary (a) Three criteria were used in assessing the children's reading ability: performance on a standardised test of word recognition (the Southgate Group Reading Test); the stage reached in the reading scheme used by the school; and the teacher's rating of reading ability on a five point scale. Judged by all three criteria, the girls were better readers than the boys. The differences were so marked and consistent that the generalisation is warranted that at this stage in their schooling, shortly before transferring to junior schools and departments, girls are superior to boys in all respects of reading ability. A comparison was made between the results of the present Study and a study conducted in 1954 (Morris, 1959). The tentative conclusion was reached that in the country as a whole, the number of poor and non-readers transferring to junior classes has dropped in the interval from 1954 to 1965. Approximately 47 per cent of the present sample had in the final stage of infant schooling not reached a stage where they could make optimal progress without further specific help in the acquisition of basic reading skills. About a fifth of these children (10 per cent of the present sample) had barely made a start with reading. It was clear that given the present age of transfer, junior schools and departments have to be prepared and equipped to continue the specific teaching of reading skills to a substantial proportion of their first year children. (b) Two assessments were made of ability in number work: performance in a Problem Arithmetic Test designed for the present Study; and a teacher's rating of number work on a five-point scale which stressed insight and understanding rather than mechanical or rote ability. There was evidence that in problem arithmetic ability, as assessed by this test, boys of this age are superior to girls. There was evidence from the analysis of the teachers' ratings, too, of superior ability amongst boys; but here the indications were that there are more boys of above average ability and no difference between the sexes in the proportions below average. This variation in the results obtained may have been due to the nature of the two assessments or to the different aspects of arithmetical ability assessed. (c) Ratings were also obtained from the teachers of the children's 'oral ability', 'awareness of the world around' and 'creativity'. As rated by teachers, girls manifest better 'oral ability' than boys, and there were indications that this tendency is more marked in the children of below average 'oral ability' than those above average. Boys are felt by their teachers to have more 'awareness of the world around' than girls, but this difference between the sexes appears to be confined to the children rated above average. Compared with boys, girls show evidence at this age of superior 'creativity' as rated by their teachers. (d) More than five per cent of the present sample in ordinary schools were receiving special educational help because of educational or mental backwardness. The head teachers considered that a further eight per cent would have benefited from such help. In view of the statement (HMSO, 1946) that 'special educational treatment is not normally required for infants unless they are so seriously retarded that they should attend a special school', the present findings indicate a need to re-examine this matter. (e) Approximately 0.4 per cent of the sample were known to be in special schools. Of the children in ordinary schools, head teachers considered that about two per cent would currently have benefited from special schooling. For a further 1.8 per cent the head teachers did not feel able to express a definite opinion. These results appeared to reflect a need felt by the head teachers for earlier transfer to special schools than is the practice at the present time. (f) It was estimated on the basis of head teachers' opinions that at least seven per cent of the present sample were likely to need special educational treatment whilst they were between the ages of seven and nine years. However, there were a number of complicating factors and it was not possible to draw any conclusions about the accuracy of estimates previously made (HMSO, 1946). There was evidence that when head teachers' opinions are sought on the question of special educational treatment, the precise terminology used is of more importance than may be realised. If it is asked whether children 'would benefit' from such provision, the proportion of children included is likely to be higher than if the term 'need' (and, probably, 'require') is used. It is arguable which of these terms is most appropriate. References DEPARTMENT OF EDUCATION AND SCIENCE Slow Learners at School Pamphlet No. 46 (HMSO) (1964). MINISTRY OF EDUCATION School Health Service and Handicapped Pupils Regulations (HMSO) (1953). MINISTRY OF EDUCATION Special Educational Treatment Pamphlet No. 5 (HMSO) (1946). DOWNING J The i.t.a. Reading Experiment (Evans Bros.) (1964). MORRIS JM Reading in the Primary School (National Foundation for Educational Research) (1959). TANSLEY AE and GULLIFORD R The Education of Slow Learning Children (Routledge and Kegan Paul) (1960).
B. SCHOOL VARIABLES I: Introduction All the information gathered in this project was judged to have possible relevance to the development of individual children. Some of the data describes characteristics of the schools, and it is intended at a later stage to test certain hypotheses concerning associations between these characteristics and the development of the children. In this section descriptive statistics are presented for those school variables upon which little or no information is available from other sources. It must be emphasised here that the Study is concerned essentially with a representative sample of children and not of schools. Thus, the data cannot be used to answer this kind of question: 'What proportion of schools catering for seven year olds have parent/teacher associations?' However, the data can answer the question: 'What proportion of seven year olds are in schools which have a parent/teacher association?' There may, in fact, be little or no difference between the answers to the two kinds of question; but where the characteristic being considered is related to the size of the school there will be a difference, since larger schools contribute more children to the cohort than smaller ones. The 10,833 children for whom information is given in this section were all in maintained infant, junior with infants or all-age schools, or in independent schools catering wholly or mainly for children who are not handicapped. The actual numbers in each type of school are detailed in the Introduction to Section VA: 'Educational Factors'. In one sub-section (5) the results for a smaller sample of children had to be used for reasons which are mentioned in this sub-section. 2. Contact between schools and parents Of course, the most important contacts made are those in which parents discuss their children with the teacher or head teacher. This topic is specifically dealt with in Section VF: 'Environmental Factors'. In this sub-section the emphasis is upon organised and relatively more formal contacts between the school and the parents. Nevertheless, these also create opportunities for informal discussion. There is a wide range of school activities and occasions in which parents may participate. An attempt was made to seek information about those contacts which might best reflect the general tenor of the relationship between the school and the home; also those which at a later stage of analysis might afford some measure of discrimination between schools which chose - or were able - to foster actively the interest and involvement of the parents in the school's work. Even within this general area a rigorous selection of questions had to be made to achieve a reasonable balance between the various educational aspects to be studied. Four questions were put to schools and the results are detailed in Table 13. Table 13 Contact between schools and parents Although these results are given in one table, it will be appreciated that the questions were not mutually exclusive. The numbers in each category are expressed as a percentage of 10,833. It is clear that the majority of the schools in this sample preferred to establish contacts with parents without the more formal framework of a parent/teacher association. 3. Introduction to school It is the practice of some head teachers to allow pre-school children to spend some time in the school before they actually start. Many people consider that this has much to commend it. It goes some way towards lessening any anxiety felt by children as they approach what is for them a new world. Whether the practice is adopted will depend upon a number of factors, amongst which the staff/pupil ratio is likely to be very important. In this sample, the schools of 3,443 children (approximately 32 per cent) were in fact, using some form of 'introductory attendance'. 4. Allocation of children to classes The question of the allocation of children to classes is more complex than would at first appear. It might seem that children are either 'streamed' by ability - however this is assessed - or they are not. Instead, one is not faced with two possibilities but with a continuum. At one end of this continuum would lie those schools in which the head teacher feels that the best class grouping is one which achieves the greatest heterogeneity in terms of the children's ages and abilities. Infant school head teachers who adopt this approach place children from five to seven years old in the same class, deliberately setting out to achieve a wide ability and age range. Such a practice is sometimes referred to as 'family grouping'. At the other end of the continuum are head teachers who feel that the more homogeneous the class, the better is the teacher able to meet the needs of the children. Such a head teacher would 'stream' by ability, where possible. In between these two positions are a large variety of situations: for example, all the infants may be in one class, so that a heterogeneous group is inevitable; or children may be allocated to classes by age, in which case a measure of homogeneity is sought. Even between classes formed in the same overt way there will be differences in approach on the part of the teachers which will reflect in some measure their attitudes - or those of the head teachers - towards this question. Enough has been said to outline some of the difficulties in assessing what is basically an educational approach or an attitude. All that was possible in the present Study was to obtain relatively crude information on the method of allocation to classes. It was felt, nevertheless, that this data, despite their limitations, would be valuable since no other information appears to have been published in this country on pupils' allocation to classes at the infant stage. Table 14 Formation of class It will be seen that the majority of the children (about 73 per cent) were allocated to classes by age: in year groups, or selected by age within the year group. Approximately seven per cent were overtly 'streamed' by ability or attainment. However, there was a further 10 per cent of children in classes formed by 'other arrangements'. It seems likely that in the majority of these classes an attempt was being made to achieve some kind of homogeneous grouping; for example, by promoting brighter children to a higher class or holding back the less able. The nature of these 'other arrangements' has been detailed by the schools, but at this stage time does not permit an analysis of the replies. 5. Age of starting full-time infant schooling There has been increasing interest of late in the association between date of birth and educational performance. Younger children in any school year group appear to be at a disadvantage compared with the older ones (Pidgeon, 1965). There is clearly more than one factor operating here, but an important one would seem to be the length of schooling obtained. The age at which children are admitted to infant schools in England varies between education authorities and also within authorities from year to year depending upon the provision available in relation to the numbers of children approaching the age of five. Two years is the minimum period normally spent in an infant school or department, but children are in some areas admitted at four years of age, provided that they will become five before the start of the next term, whilst in others an earlier start than this is possible. In some cases, therefore, a much longer period than two years is spent in school before transfer to a junior school or department. Thus, although the children in the present sample are virtually all of the same age, it was known that there would be differences in the length of schooling received prior to the present educational assessment. In Section VIB the relationship between the age of starting full-time infant schooling, educational performance and social adjustment at seven years of age is examined. Here, the numbers of children starting full-time infant schooling at different ages is presented. It will be noted that the total number of children (5,805) is much smaller than elsewhere in this section. There are a number of reasons for this. First, it was only possible to include children for whom a Parental Questionnaire had been received when the data processing was commenced, because the information about the age of starting school was obtained from the mothers. Secondly, children are included only where it was known that they had not attended a nursery school or class, since it was felt that children who had had nursery schooling might as a group be atypical in this context; furthermore, it would in some cases be difficult to determine when nursery schooling ceased and infant schooling commenced. Another fact to be noted is that the figures detailed below are abstracted from a larger table, not included in this Report, which includes a breakdown of the figures in relation to the occupational group of the father. Thus, children for whom this information was not available, including those in families where there was no male head of the household, are grouped together with those who started school outside the limits of the three age groupings for which numbers are detailed. Table 15 Age of starting full-time infant schooling The results indicate that nearly half of this sub-sample started school when they were between 4 years 6 months to 4 years 11 months old. The great majority of these would have started school in January 1963, since they were five years old in March of that year. Over 43 per cent would have started school at the beginning of the following term. In assessing the reliability of these figures and the extent to which they may be representative of the situation in the country as a whole, a number of factors should be borne in mind. First, the fact that those children who could not be classified by the occupational group of their fathers are included with those who started school outside the three specified age groupings will depress the percentage figures for these groupings. However, the effect of this has been checked; it is minimal and does not alter the overall position. Secondly, the information was obtained from the mothers, and some may have had difficulty in recalling an event which occurred two years earlier in terms of the age categories detailed above. As against this, the question was put by health visitors - or another officer of the local authority - who would have helped mothers, where necessary, with this 'calculation'. Of course, health visitors are likely to be familiar with local practice. Thirdly, the school year 1962-63 was a difficult one for schools and authorities because of the effect of increasing the training college courses for teachers from two to three years at this time. The consequent acute shortage of staff may have forced some schools to take children later than was their usual practice. For these reasons, then, the figures should be viewed with some caution. However, despite these reservations, the suggestion by Pidgeon and Dodds (1961) that 'There is a general, but not universal, rule in this country that children start formal schooling at the beginning of the term in which they will become five years old' clearly needs re-examination in the light of our findings. 6. Age of commencing phonics in reading There is a wide variety of possible approaches to the introduction and subsequent teaching of reading. In this Study one particular aspect was selected for closer investigation; namely the stage at which a systematic attempt is made to introduce children to the sounds of individual letters or phonemes. The introduction of 'phonics' - as this is customarily termed - is made much earlier in the reading programme in some schools than in others. There has been - and still is - considerable debate amongst schools, and in educational circles generally, about the age at which phonics should be systematically taught. To say that the right time is when the individual child is 'ready' is to avoid the issue in the absence of any universally agreed or well validated criteria as to what constitutes readiness. This Study cannot claim to throw any new light of an experimental nature upon this question. However, it can give information, hitherto not available, about the age at which this systematic introduction is in fact commenced in schools. Although this Report is concerned with children in England, it was felt that comparative figures for Wales and Scotland would be of interest in this particular context. The results are given in Table 16 and reproduced in graph form in Figure 2. It will be seen that there are a substantial number of 'Don't knows' in this table. This was expected, since the data for most of the children were retrospective and staff changes, as well as changes of school by the children, will have meant that information on this question was sometimes unobtainable. However, there is no reason to believe that if the information for these children had been available it would have differed in overall pattern from that obtained for the rest of the children. Thus, it was felt appropriate to exclude 'Don't know' replies from the percentaging. The differences between the three countries are marked and interesting. The English pupils were introduced to phonics at a later age: the peak age for its introduction in the English schools was from 5 years 6 months to 5 years 11 months, whilst in Wales and Scotland the corresponding age was from 5 years to 5 years 5 months. The differences in the distribution, both between England and Wales and between England and Scotland, are highly significant statistically. Phonics had been taught to about 54 per cent of the children for whom this information was available in Scottish schools before the age of 5½ and to 46 per cent of the children in Welsh schools; the corresponding percentage for the children in England was about 29 per cent. Since no teacher would commence teaching phonics before she considered a child was ready for this step, it would appear that the teachers in Wales and Scotland considered their pupils to be ready at an earlier age than their colleagues in England. It is interesting to speculate whether there are any real differences in the children's degree of readiness or whether tradition or teacher training courses are responsible. Or perhaps a combination of these? Additionally, some of the Welsh children were Welsh-speaking and would have been taught to read in that language, which may lend itself to a more phonic approach than English. Apart from the difference between the children in English schools and those in Wales and Scotland in relation to the age at which phonics were introduced, there is also evidence of more variability in the English and Welsh samples than in the Scottish. Both of these findings may be a reflection of more uniformity in Scotland, if this is the case, in the age of starting school. Table 16 Age at which the systematic teaching of phonics (i.e. letter sounds) was commenced in school Figure 2 Age at which the systematic teaching of phonics (i.e. letter sounds) was commenced in school 7. Age of commencing 'formal' written arithmetic There have been considerable changes in many infant schools during recent years in the approach to number work. An increasing use of the term 'number work' or even 'mathematics' in place of 'arithmetic' or 'sums' is symptomatic of this change and the wider approach adopted in some schools. One accompaniment of the change has often been the introduction of 'formal' written arithmetic at a later stage. Again, a comparison of the practice in England, Scotland and Wales is made. The results are presented in Table 17 and in graph form in Figure 3. Table 17 Age at which 'sums' (i.e. 'formal' written arithmetic) was introduced in school As in Table 16, there are a substantial number of 'Don't knows' and, for the same reason, discussed in the previous sub-section, they have been excluded from the percentaging. In English schools pupils were introduced to 'sums' at a later age than their peers in Wales and in Scotland. The difference between the practice in the English and Welsh schools seems to centre on the fact that more of the children (approximately 32 per cent) in the latter schools had reached this stage before 5½ years of age than in England (approximately 17 per cent). Figure 3 Age at which children were introduced to 'sums' (i.e. 'formal' written arithmetic) in school One apparent difference in practice among the Scottish schools as reflected in this sample of children is in their greater uniformity. Thus, although the peak period for the commencement of 'formal' written arithmetic was between 5 years 6 months and 5 years 11 months in all three countries, well over half (about 62 per cent) of the Scottish sample reached this stage during this six-month period, whereas in England and Wales the corresponding percentages were about 43 per cent and 36 per cent. This may be a reflection of more uniformity in Scotland, if this is the case, in the age of starting school. 8. Summary (a) Whilst the majority of the children were at schools in which there was some form of organised contact between the school and the parents, most schools established these contacts without the framework of a parent/teacher association. (b) About one third of the children attended schools in which the head teachers adopted the practice of allowing pre-school children to spend some time in the school before actually starting. (c) The allocation of children to classes was done mainly on an age basis. It appeared that about half of the children were in classes where an attempt had been made to achieve a degree of homogeneity by some form of selective grouping, based on age within the year group, on ability or by some other arrangement. Only seven per cent of the children were in 'streamed' classes. (d) The results indicated that of those children who had had no nursery schooling nearly half started school when they were aged 4 years 6 months to 4 years 11 months; the great majority of these would have commenced at the beginning of the term in which they attained the age of five. The mothers' reports showed that over 43 per cent of the children commenced school at the beginning of the following term. The school year 1962-63 may have been untypical because of the particularly acute shortage of teachers due to the lengthening of the teacher training college courses at that time. (e) A comparison was made between the practice in England and that in Wales and in Scotland in relation to the age at which the systematic teaching of 'phonics' was commenced with these children and also the age at which 'sums' or 'formal' written arithmetic was introduced. The evidence was that pupils in English schools are introduced to both these aspects of their school work at a later age than those in Wales and Scotland. In all three countries a majority of the children had commenced 'formal' written arithmetic and were receiving some systematic teaching of 'phonics' before the age of six. There was also evidence of more uniformity of practice in Scotland than in Wales or England in both these spheres. (f) In varying degrees all the questions here discussed have relevance to the educational development of children; it is therefore somewhat surprising that there is an almost complete lack of comparable information from other sources, particularly on a national basis. Such information is relatively straightforward to collect; moreover, if gathered at regular intervals it would be valuable to individual schools and teachers, to administrators and to those with responsibility for the training of teachers; also it would reflect changing practice and act as a stimulus to further change. References PIDGEON DA Date of Birth and Scholastic Performance, Educational Research, Vol. VIII, No.1 (National Foundation for Educational Research) (1965). PIDGEON DA and DODDS EM Length of Schooling and its Effect on Performance in the Junior School, Educational Research, Vol. III, No. 3 (1961).
C. MEDICAL AND PHYSICAL FACTORS Note Most of the tables from which the summary tables in this and the following sections are derived are excluded from this version. They are given in full in Appendix I to the full version. 1. Introduction This part of the Study is concerned with some aspects of the physical and medical status of the sample children. The information contained in this and in the following section (VD) is based upon data from approximately 8,000 Parental and Medical Questionnaires received by the middle of August 1965. The total number of children for whom information is available differs slightly in different tables. This is due to certain minor technical considerations, including the fact that the data in this and the following two sections were divided among four different punched cards. The data were derived from a general and systemic medical history gathered from the mother and from the results of examinations by school medical officers. (a) Medical history The design of this part of the enquiry is discussed in Section IIID. Information was sought on a wide variety of conditions and recorded on a pre-coded form. A positive reply to a question on whether the child suffered from a given condition led the interviewer to ask about age of onset, recent history or other specified details. These data are sometimes retrospective and with attendant possibility of bias. These considerations are discussed in more detail in the Introduction to Section VE: 'Medical and Developmental Sex Comparisons' . (b) Medical examination A special problem was posed in the design of the Medical Questionnaire on which the results of the medical examination were to be recorded. It is known that analysis of school medical records yields gross fluctuations in incidence, suggesting very wide variations in interpretation. The pattern of the questionnaire, therefore, differed from that used for routine school records, though every effort was made to avoid complicated format. The medical practitioners who conducted the examinations consisted almost entirely of school medical officers, who are probably more experienced in technique of routine examination of children than any other section of the medical profession. With the exception of certain tests of special senses, it was therefore unnecessary to specify the manner in which the clinical examination should be carried out. The specially framed questions provided an objective determination of the current physical and medical status of the children. Any enquiry asking only for defects or abnormal findings to be noted would have allowed preconceived notions of normality to govern the decision whether or not an observation should be recorded. Therefore, within each system the medical examiner was asked to indicate a positive or negative response to the presence of specified conditions. Care was taken as far as possible to avoid questions inviting varying interpretations or subjective responses. For many specified conditions further details were enquired for special analysis. The doctor was also asked to comment upon any residual abnormality in each system examined. In the main, therefore, facts rather than opinions were sought. However, clinical assessments were also included in certain fields, notably the special senses, to allow subsequent comparison with test results. Specific tests of special senses, laterality or coordination were mostly derived from existing tests of proven value in clinical practice. In order to maximise comparability of results among a large number of examiners it was necessary to define the conditions of special tests in detail on the questionnaire. (c) Interpretations The medical and physical data on this incomplete sub-sample justify only tentative conclusions. For instance, it is known that the children involved in these 'early returns' are biased in family occupational group compared with 'late returns'. Moreover, those 'early returns' with Parental and Medical assessments show better reading and social adjustment than children for whom only educational information was available (see Section II: 'Description of the Sample '). Although the overall bias is not marked, it may become relevant when individual incidences or distributions in this sub-sample of 'early returns' are considered. Bias would be particularly important if a given physical factor is not associated with normal social circumstances, educational progress or behaviour and adjustment. Time has not permitted a comparison of the sub-sample of 'early returns' with the 'late returns' for any medical or physical factors. The possibility exists, for example, that the sub-sample does not contain a representative proportion of physically handicapped children. The present Report, therefore, deals with tests of special senses and with incidences of minor physical abnormality rather than with severe handicaps or major conditions requiring special treatment or education. A later report will present physical data, including height and weight, correlations between current physical status and perinatal information obtained on the children at birth from the Perinatal Survey of the National Birthday Trust Fund, and also relationships between current educational and medical status of the cohort. 2. Speech (a) Introduction Information was sought from the mother on a number of speech difficulties to which the child might have been subject; and the medical examiner used a speech test, noted any stammer and assessed the intelligibility of the child's speech. (b) History and examination The results of the history and examination are summarised in Table 18. Table 18 Speech: Summary of history and examination The superiority of girls over boys in speech function is amply demonstrated in the above table. Three of the sex differences are highly significant and the fourth is in the same direction, although it fails to reach the one per cent level of significance. Fewer girls than boys were reported to have attended Speech Therapy Clinics (Table A33: Appendix I), again suggesting superior speech function. However, both this sex difference and that in 'history of other speech difficulty' (Table 18) may simply have resulted from later speech development in boys, since more girls were reported by their mothers to have been talking (i.e. 'joining two words') by the age of two years (Boys 92.4 per cent; Girls 95.2 per cent). The overall incidence as opposed to sex comparisons of speech difficulty in Table 18 should be regarded with particular caution in view of the retrospective nature of the information and any bias in this sub-sample. It should also be noted that in the great majority of children in whom some stammer was observed, the handicap was slight; and likewise for any lack of intelligibility on testing of speech. (c) Speech test The sentences used in this speech test were composed of words chosen to demonstrate the commonest defects of speech in children (Sheridan, 1945, 1965). When applied by a large number of medical examiners the test can be expected to produce somewhat less reliable results than it would in the hands of those with specialised knowledge of speech development. Table 19 Speech test results Six sentences were used in the test. The child was asked to repeat each one after the examiner, who noted any mispronounced words and recorded the total number of errors. Cumulative percentages are shown in Table 19 and presented in graph form in Figure 4. Both the table and the figure are based upon the children (N= 7,876) for whom this information was available when the data were processed. A trend test for sex difference in the proportion of mispronounced words in Table 19 confirms a superior performance in girls. Figure 4 Speech test 3. Vision (a) Introduction Information presented here on the children's vision includes the results of visual acuity testing, of a reported history of squint, of evidence of squint on examination and of a clinical assessment by the examining doctor of the severity and educational implications of any observed visual defects. One objective of the present Study was to obtain representative national figures using defined procedures, although it was an inevitable drawback that many examiners had to be used. Later it is hoped to present fresh methods of classification, as published figures suggest gross variability in the way visual defect is assessed and reported. Thus, the incidences quoted by individual local authorities for visual defects in school children of all ages ranged in 1963 from 1.38 per cent to 25.2 per cent (HMSO, 1964). In 1961 (HMSO, 1962) the reported incidence of squint varied from 0.045 per cent to 10.6 per cent. (b) Visual acuity A standard Snellen test chart of block capitals was specified for the vision test and instructions were given in the Questionnaire on the conditions of the test as well as for the procedure to be adopted when a child did not know the letters in the chart. Each eye was tested separately without glasses and also with glasses if these were worn. Table 20 shows the visual acuity in each eye of all the children in this sample for whom this information was available when tested without glasses. The number of children for whom results are presented differs very slightly for the right eye and the left in Table 20 and in the following two tables due to difficulties in processing the data. A sex difference was tested for each eye separately and the results were as follows: Left eye
Right eye
As will be seen, no significant sex differences were found. Table 20 Uncorrected vision (i.e. without glasses) Table 21 Vision of children who never wear glasses Table 20 shows that 17 per cent of all the children tested had sub-optimal visual acuity in the right eye, taking 6/6 as perfect vision. A similar proportion had sub-optimal vision in the left eye, but the degree of overlap is not yet known. However, Table 20 does not distinguish between those children who already had glasses prescribed and those who had never worn glasses. The test results of the children who had never worn glasses are therefore shown in Table 21. Table 21 shows that over 13 per cent of children who did not wear glasses had at least one eye with visual acuity below the optimum, taking 6/6 as perfect vision. If 6/9 vision is considered acceptable, only three per cent of these children are shown to have imperfect vision in at least one eye, but some authorities would not be prepared to accept 6/9 vision as normal in a child until an examination by an ophthalmologist or ophthalmic optician had shown that an eye defect had been excluded. Table 22 Corrected vision of children wearing glasses In those children tested wearing their glasses, the testing of the right eye showed sub-optimal vision in just over 50 per cent, and results for the left eye were similar. It may be that some children were tested wearing glasses which were intended for close work or reading, but the figures suggest that it would be a wise precaution for teachers to consider any child wearing glasses in a primary school to have defective vision and to place that child near the front of the class. (c) Squint A history of squint or of suspected squint obtained from the mother was recorded in the Parental Questionnaire. A manifest squint observed by the examining doctor was recorded and two tests were specified to enable him to detect the presence of a latent squint, i.e. a tendency to squint which is, in normal circumstances, kept under control. The type of squint (divergent or convergent) and the eye affected were also recorded for later analysis. The teacher and school medical officer are favourably placed to pick out a child with this condition at school, but squint may derive from a number of causes dating from before birth to any time thereafter. Early detection of squint and other visual defects in the pre-school child is dependent upon the powers of observation of the mother and the use she might make of her general practitioner and the child welfare facilities. Neglect of a severe squint, or other visual defect, may result in the sight of an affected eye becoming permanently impaired. The results of the history and examination are shown in Table 23 in summary. Table 23 Squint It will be seen from Table 23 that approximately six per cent of the children had either a latent or manifest squint on examination. The significantly greater percentage of latent squints in girls is an unexpected finding. There was no difference between the sexes for manifest squint or for a history of squint or suspected squint. (d) Visual assessment On completing the examination of the eyes and after the eye test, the medical officers were asked to rate the educational implications of any visual defect. This was done on a five-point scale, but in Table 24 all children who received anyone of the lowest three ratings are aggregated. The numbers with these ratings were very small and it was felt unwise to consider them as separate groups until time permits individual scrutiny of the questionnaires. This crude breakdown was also considered advisable because many of the medical examinations were undertaken in clinics and the medical officers may have had no opportunity to discuss detailed educational difficulties arising out of any visual defect with school staffs. Table 24 Visual assessment It will be seen that there is no significant sex difference. This assessment serves to identify the 14 per cent of children judged to have some visual handicap. Thus, it is an assessment of function and it cannot fully take into account all the degrees of visual acuity shown in Tables 20, 21 and 22. 4. Hearing (a) Introduction The testing and assessment of the children's hearing included a pure-tone audiogram; a functional assessment of hearing by the examining doctor; a clinical hearing test; and any reported history of hearing difficulty or attendance at an audiology clinic. Time has not yet permitted any analysis of the audiograms. (b) Assessment of hearing A functional assessment of hearing was carried out by the school doctor after he had completed the examination and hearing test. The assessment was made on a four-point scale and the definitions at each point on the scale were framed to correspond approximately to the statutory definitions of deaf and partially hearing children. In view of the small number of children who were assessed as having some degree of auditory handicap and the impossibility of scrutinising the questionnaires individually to confirm the appropriate category, it was decided to group together all such children. The purpose was to compare them with those judged to have 'normal hearing'. The results are shown in Table 25, in which children with 'some degree of hearing impairment' include those with hearing loss which had been corrected by a hearing aid; those whose understanding of speech was impaired (even with a hearing aid); and those whose hearing disability was so severe that they could not understand speech at all. The proportion of children in the present sub-sample who fell into the latter two categories was 0.1 per cent. Table 25 Assessment of hearing The difference between the sexes on this assessment is not significant. A little under five per cent of the children in this sub-sample were judged by the medical officers to have some degree of hearing impairment. (c) Hearing test The hearing test consisted of twelve test words which the child was required to repeat one by one after the examiner. The test conditions were specified in detail in the questionnaire. The twelve words were selected to test the auditory acuity over the speech frequency range (Sheridan, 1958 and 1965). A preliminary analysis of the results is presented in the following table. Table 26 Hearing test There were no significant sex differences on the results of this hearing test for each ear separately. It will be seen that some four to five per cent of the children, for each ear separately, failed to repeat three or more words of the twelve used. A functional assessment by the examining doctor (Table 25) indicated also that five per cent of the children had some degree of hearing impairment. Some further analysis will be necessary to determine any correlation between the two, test and assessment, and their relationship to audiometry results. In the hearing test a poor understanding of the meaning of words or an inability to concentrate may adversely affect the score of a child with a minor degree of hearing loss. Conversely, a bright child will be at an advantage in coping with a loss of auditory acuity. The audiogram and the clinical hearing test are therefore complementary in the assessment of hearing. The test was developed from one which was originally designed for individual specialist use. Its application to the children in this Study may, after further analysis, establish its value as an additional screening test for use in conjunction with existing pure-tone audiometric screening methods. (d) History of hearing difficulty and clinic attendance Table 27 History of hearing difficulty and clinic attendance The results of the analysis showed no significant sex differences. Some 10 per cent of the children in this sub-sample were reported by their mothers to have had a suspected or confirmed hearing difficulty at any time in the first seven years of life. Numbered amongst these will be children who had a temporary catarrhal deafness following a cold; those whose lack of attention or responsiveness had led to a suspicion of hearing difficulty; and those whose hearing might have been shown with certainty to be impaired. Approximately eight per cent of the children were reported to have attended hearing or audiology clinics. These children would include a considerable number who had been sent for routine audiometry as a result of doubtful screening tests of hearing at school. On retesting, the majority of the children are found to be normal. The figure of eight per cent does, however, give some indication of the amount of work undertaken by audiology clinics. 5. Laterality (a) Introduction The school child who is left-handed differs from the majority of his fellows. In the past, attempts were made by parents and teachers to correct left-handed tendencies in what was then thought to be the best interests of the child. The modern attitude is more permissive. The large number of theories on handedness merely reflect the uncertainty about its origins and its implications. Hereafter in this section, the term 'handedness' is confined to the mother's opinion on the child, and 'laterality' is used to describe the results of the tests. Results on handedness and tests of laterality will be analysed in the Study for three main purposes: (i) to determine the pattern of right or left dominance and mixed laterality in this cohort;Preliminary results are presented here on the first of these items. (b) Handedness - mothers' information The mothers were asked whether their child was right-handed, left-handed or mixed right-handed and left-handed. The results are shown in Table 28. Table 28 Handedness (mothers' information) The evidence from the analysis is that, as reported by mothers, more boys than girls are left-handed and also more boys are mixed-handed. (c) Tests of laterality The pattern of limb-dominance or eye-dominance is not necessarily invariable for any one person but may depend upon the activity or task which is undertaken. Thus, 'mixed-handedness' reported by the mother (Table 28) will in most cases refer to the child's ability or preference to use different hands for different purposes. During the medical examination tests were used to determine the hand, foot and eye which was used by each child in specific tasks. The examiner was asked to record for the two tasks involving the use of the hand and arm whether only the right hand was used; only the left hand; or, both right and left hands. Similarly, a record was made of the preferred foot and eye. In Table 29 the results of these tests are presented separately for hand, foot and eye. On the basis of their performance in the tests, the children were allocated to one of three groups: 'Right' (when only the right hand, foot or eye was used); 'Left'; or 'Mixed' laterality. The categories to which the children have been allocated should be seen in terms of the type and number of tasks set. For example, had more tests been used, the number of 'mixed-laterals' may have been somewhat higher. In testing for sex differences, therefore, 'mixed-laterals' were excluded. Table 29 Laterality tests The results show that more boys than girls show a preference for using the left hand, left foot or left eye in terms of the tasks specified. The percentage figures relate only to these particular tests and, in any case, should be regarded as provisional at the present stage. Approximately one third of the children showed no clear foot preference and 58 per cent were right-footed. One third showed left eye dominance. Although the incidence of left-hand laterality shown by testing was not as high as left-handedness reported by mothers, it remains higher than that shown by many other workers. 6. Dental examination (a) Introduction The medical examination of the 1958 cohort provided the opportunity to assess the state of children's teeth at the stage of transition from infant school to junior school on a national scale, though it was realised that such an inspection carried out by doctors could not achieve the same degree of accuracy as that undertaken by dental officers. The index of dental caries used - the total number of decayed, missing and filled teeth - provides a basis for comparison with existing figures and allows possible associations with other factors to be sought at a future date. A record of the number of decayed, missing or filled teeth has the added advantage that it allows some assessment of the amount of dental decay more or less independent of the dental treatment. However, this index does not make separate allowance for naturally shed teeth which, at the age of seven years, is a relevant factor. For the purpose of comparison, figures for Nottinghamshire in 1963 (HMSO, 1964) are reproduced in Table 30, since they show the sexes separately. Table 30 Dental decay (Nottinghamshire, 1963) (b) Dental decay The results of the dental examination of the present sub-sample are shown in Table 31 for those children for whom this information is available. Table 31 Dental decay It will be seen that the percentage of children with no decayed, missing or filled teeth in the present sub-sample lies between the figures quoted for five year olds and 12 year olds in Nottinghamshire (Table 30). Inspection of Table 31 suggests no sex difference, and there is further evidence of this when analysis is carried out on the full distribution of DMF teeth in the sample (Table 32). These results are shown in graph form in Figure 5. The provisional figures shown in Appendix 1, Table A33, show that some three quarters of the sub-sample were reported to have attended, at some time in the past, a dental clinic, dental surgeon or orthodontist. A more significant point, perhaps, might be that nearly a quarter of the children were reported as not having made use of these facilities in spite of the availability of a school dental service and a free personal dental service. Both these services are already overworked, and if the needs indicated by the present findings are to be met, clearly a reappraisal of dental establishment is required. Table 32 Dental decay Figure 5 Dental decay 7. Micturition and bowel control (a) Introduction In obtaining information about micturition and bowel control, as for other developmental milestones, retrospective inquiry from each mother as to the age at which these milestones were reached would have very limited value. Instead, an age was chosen at which it was known that the great majority of children would have achieved bladder or bowel control and the mothers were asked whether their child had achieved control by this time. Though answers obtained would still not be without a measure of unreliability, the major aim was not to establish norms for these facets of development. It was rather to identify those children who had reached these milestones later than the majority of their peers so that these facets could be related to other aspects of health and development; and also so that a comparison could be made between the sexes in these respects. (b) Micturition and bowel control by day Two questions were put to mothers about micturition and bowel control by day. The results are summarised in Table 33. Table 33 Micturition and bowel control by day The percentage of boys and girls who were reported to have wet themselves in the daytime over the age of three years is similar; approximately 4.4 per cent. The response to the second question suggested that 1.8 per cent of boys and 0.6 per cent of girls had not achieved bowel control by the age of four. The sex difference is highly significant. In replying to the above questions, the mother was told to ignore the occasional mishap. However, the problems of wetting and soiling are clearly not likely to be great, even in nursery schools or classes. (c) Bedwetting The response in Table 34 suggests that bedwetting over the age of five years, even when occasional mishaps have been excluded, must be considered quite a common phenomenon, affecting some 11 per cent of children in this sample. As this question related to events occurring within two years of the Study, it is less likely to evoke unreliable replies than those concerning the earlier milestones. More boys were reported to be wet at night after five years of age than were girls (12.1 per cent boys and 9.7 per cent girls). The difference between the sexes is highly significant. Table 34 Bedwetting The figure of 11 per cent is high enough to merit further investigation. It may be that the age at which bladder control is normally attained extends over a greater range than is generally accepted. References DEPARTMENT OF EDUCATION AND SCIENCE The Health of the School Child 1962 and 1963 (HMSO) (1964). MINISTRY OF EDUCATION The Health of the School Child 1960 and 1961 (HMSO) (1962). SHERIDAN MD The Child's Acquisition of Speech, Brit. Med. J. i, p. 707 (1945). SHERIDAN MD Simple Clinical Hearing Tests for Very Young or Mentally Retarded Children, Brit. Med. J. ii, pp. 999-1,004 (1958). SHERIDAN MD Personal Communication, Paediatric Research Unit, Guy's Hospital, London (1965).
D. MEDICAL AND DEVELOPMENTAL SEX COMPARISONS 1. Introduction In this section comparisons are made between boys and girls for incidence of some past illnesses, accidents, hospital admissions, attendance at specialist clinics; and for some data from the medical examination of the children at seven years of age. The totals from which these comparisons are made should not be seen as more than estimates of incidence of illnesses or other factors in the whole cohort for a number of reasons. First, the cohort was not studied between birth and the present time; the historical data are thus retrospective. Secondly, it was gathered from parents and not from hospital or other records, although it is intended to obtain corroborative data from hospitals or clinics. The possibility exists of differential parental recall on some items favouring one or other sex, but this is not likely to be an important source of bias. In this section the results of sex comparisons are summarised from tables which are contained occasionally in Section VC or in Appendix 1, Table A33. However, the majority of the tables on which sex comparisons are based have been omitted. These are published elsewhere. (Pringle MLK, Butler NR and Davie R 11,000 Seven Year Olds Longmans, Green & Co. Ltd, 1966). Where comment is made on a sex difference in the text this is significant at a level of at least one per cent unless otherwise specified. Mortality rates are known to be higher in boys than in girls; a greater male perinatal death rate was also found for the present cohort in the Perinatal Mortality Survey. It is therefore reasonable to expect that there should be greater childhood morbidity in males than in females, and these hypotheses can be tested for many illnesses and childhood conditions. 2. Accidents The results on this sub-sample of children confirm the view that, in general, boys are more accident-prone than girls. Thus, a higher proportion of boys had had one or more hospital admissions for road accidents (Boys 3.0 per cent; Girls 1.7 per cent) and for all other accidents or injuries which had not occurred within the home (Boys 9.7 per cent; Girls 6.9 per cent). More boys, too, were reported as having had in the past a head injury with loss of consciousness (Boys 3.5 per cent; Girls 2.6 per cent). For home accidents, however, there was no significant difference between the sexes (Boys 9.9 per cent; Girls 8.8 per cent). From these provisional figures it would appear that about three per cent of the boys had been admitted to hospital on one or more occasions for a road accident; as many as about 10 per cent had had an accident in the home severe enough to require hospital admission; and a similar proportion were reported to have been admitted to hospital for other accidents or injuries sustained outside the home. The corresponding figures for girls were still disconcertingly high, emphasising the need for increased concentration upon accident prevention in childhood. 3. Upper respiratory infections Past upper respiratory infections showed no sex difference, as judged by a history of three or more ear infections during the year preceding the medical examination (Boys 12.8 per cent; Girls 13.9 per cent) or by a history of admission to hospital for tonsils and/or adenoids (Boys 15.3 per cent; Girls 15.7 per cent). On the other hand, current upper respiratory conditions did show a male preponderance in that the medical examinations revealed a higher proportion of boys with nasal obstruction (Boys 10.2 per cent; Girls 7.9 per cent) and some evidence of (0.05>p>0.01) more nasal or postnasal discharge (Boys 10.7 per cent; Girls 9.0 per cent) and enlarged glands in the neck (Boys 25.9 per cent; Girls 22.6 per cent); the mothers also reported that boys were more often habitual snorers and mouth breathers (Boys 26.0 per cent; Girls 22.0 per cent). 4. Psychosomatic and behaviour problems A history of travel sickness was reported more frequently in girls than boys (Boys 18.3 per cent; Girls 29.2 per cent). There was also some evidence (0.05>p>0.01) of more frequent periodic abdominal pain in girls (Boys 14.0 per cent; Girls 15.7 per cent). Boys more frequently showed a history of tics or habit spasms (Boys 5.9 per cent; Girls 4.4 per cent); breath holding, head banging or 'rocking' (Boys 9.7 per cent; Girls 7.3 per cent); and were more frequently reported to have attended child guidance clinics (Table A33). No sex difference emerged in the history of frequent headaches or migraine (Boys 8.2 per cent; Girls 7.9 per cent), or in that of periodic vomiting or bilious attacks (Boys 17.0 per cent; Girls 17.5 per cent). 5. Convulsions There was some evidence (0.05>p>0.01) that boys had had more fits in the first year of life (Boys 2.2 per cent; Girls 1.6 per cent), but there was not an excess of males with fits after one year (Boys 2.9 per cent; Girls 2.2 per cent) or with petit mal (Boys 0.5 per cent; Girls 0.8 per cent). 6. Allergic states The objectivity of a history of asthma or eczema given by the mother is difficult to assess without confirmation. More boys were reported to have had asthma (Boys 3.6 per cent; Girls 2.3 per cent) or bronchitis with wheezing (Boys 19.4 per cent; Girls 15.9 per cent). There was no sex difference in history of eczema after one year of age (Boys 5.5 per cent; Girls 5.7 per cent), or of eczema on examination (Boys 2.8 per cent; Girls 2.7 per cent), or of a history of hay fever (Boys 5.9 per cent; Girls 5.2 per cent). 7. Speech A history of stammer and of other speech difficulty was reported more often in boys; on examination, more boys were assessed as being not fully intelligible during speech testing and there was also some evidence (0.05>p>0.01) of a higher incidence of observed stammer (see VC, Table 18). Fewer boys were reported to have been 'talking' (i.e. joining two words) by the age of two years (Boys 92.4 per cent; Girls 95.2 per cent). It is not therefore surprising that more boys of seven years had already attended for speech therapy (Table A33). 8. Ophthalmological conditions There were no significant sex differences in a history of squint or suspected squint, nor of manifest squint on examination (Section VC, Table 23); further, no sex difference was found in those assessed by the medical officer to have a visual defect (Table 24). However, more girls than boys were found to have latent squint on examination (Table 23). 9. Auditory conditions A comparison of the sexes revealed no difference in the results of a clinical hearing test, in an assessment of hearing impairment by the medical examiners or in the proportion of boys and girls who had attended hearing or audiology clinics (Section VC, Tables 25, 26, 27). On examination there was no difference between boys and girls in the frequency with which signs of past or present otitis media were present. A higher proportion of girls were reported by their mothers as having had earache in the first seven years of life (Boys 29.7 per cent; Girls 33.4 per cent). 10. Dental status At seven years of age, as judged by the number of decayed, missing or filled teeth, there was no sex difference (Section VC, Table 32) and no sex differential in past utilisation of dental services (Table A33). A minor finding was that a history of recurrent mouth ulcers was more frequent in girls (Boys 7.5 per cent; Girls 11.6 per cent). 11. Congenital malformations Sex differences were tested on a few congenital anomalies in this sub-sample. A history of 'port-wine stains' of the skin was commoner in girls (Boys 4.4 per cent; Girls 6.4 per cent) and there was some evidence of a similar sex difference (0.05>p>0.01) in the history of 'strawberry naevi' (Boys 4.4 per cent; Girls 5.4 per cent). On examination, 'birth marks' were reported more frequently in girls (Boys 11.2 per cent; Girls 13.8 per cent). There was also a preponderance of girls (Boys 0.05 per cent; Girls 0.3 per cent) with a history of congenital dislocation of the hip - although actual numbers were small - and some evidence (0.05>p>0.01) of a similar difference in the history of talipes (Boys 0.6 per cent; Girls 1.1 per cent). More boys were reported to show deformities of chest (Boys 3.3 per cent; Girls 1.7 per cent) and external ear (Boys 2.3 per cent; Girls 0.7 per cent). 12. Hernia More boys than girls had a reported history of this condition and more had been admitted to hospital for hernia repair (Boys 2.7 per cent; Girls 1.0 per cent). This difference between the sexes was apparent, too, at the medical examinations, where more boys were reported by the medical officers to have inguinal hernia (Boys 0.8 per cent; Girls 0.1 per cent).
E. ENVIRONMENTAL FACTORS 1. Introduction A comprehensive study of children's growth and development must include a study of home background. There is mounting evidence from investigations into the complex processes involved in a child's response to school (for example, Fraser, 1959; Floud, Halsey and Martin, 1957; Douglas, 1964; Wiseman, 1964) to indicate the vital importance of social factors in educational achievement and progress. In this section some of the data which were collected about the home background of the children in the present sample are presented and discussed. Most of the information was obtained in interviews with mothers, but questions relating to school attendance and parents' interest in the child's education were answered by head teachers. Only a fraction of the relevant information could be included in this section. A choice had to be made among the original factors selected for study, and it does not necessarily follow that because a topic is included here there are not others which may prove to be of equal, or possibly greater, importance that have been omitted. Within the limits of availability, items were chosen according to two criteria: first, their assumed importance and relevance to children's development; and, secondly, the likelihood of their being of special interest to the Central Advisory Council. The material presented in this section is essentially of a descriptive nature. Time has not permitted many comparisons with other studies, but not infrequently other data were not available or were not strictly comparable. Within the sample, sex differences have been considered where relevant. The numbers of children included in the different tables will vary as follows: 10,833 - All children in 'ordinary' maintained and independent schools for whom data from the Educational Assessment booklet were available for sorting on punched cards. (See Introduction to Section V A: 'Educational Factors'.)In addition, some of the figures were abstracted from computer tables and totals will vary according to the number of cases where particular information was available. It should be noted that all totals from computer tables in this section include children known to be either in 'ordinary' or 'special' schools. These children whose type of school was not known, and the very few resident in hospital, are not included. Much of the information contained in this section was derived from Parental Questionnaires; that is, for a sub-sample of 7,985 children. The question of the representative nature of this sub-sample has been discussed elsewhere (Section II: 'Description of the Sample') and may be summarised as follows: first, there were no very marked differences between the social class distribution of the sub-sample and that of the national sample; such differences, as there were, are difficult to interpret. Secondly, a comparison of the social class distribution of the sub-sample with that of 'late returns' showed an overall significant difference, but did not indicate a consistent change in the social class distribution. Thirdly, the reading ability and social adjustment at school of the sub-sample was compared with that of the children for whom a Parental Questionnaire had not been received. Here there was statistically significant evidence that the children in the sub-sample were better readers and better adjusted than the children whose Parental Questionnaires had not been received. The implications of this statistically significant but not marked bias in the sub-sample are that when information is derived from the Parental Questionnaires, any findings relating to adverse conditions are likely to underestimate the incidence of such circumstances in the total cohort. 2. Socio-economic status Information was available about the occupations of the children's fathers or the male heads of households, both from the Perinatal Survey and from the present Study. This allowed the possibility of a choice between using current or past information for classification into socio-economic groups. The value of using current information about the father's occupation lay in the nature of the material to be presented in the Report. Where perinatal data were to be linked with current data about the children, the arguments for and against using present or past information seemed fairly evenly balanced. But for much of the Report the educational, medical and social information presented and discussed was concerned directly with the children's development at the age of seven. For this, the use of father's present occupation was considered most relevant. It was finally decided, therefore, to use the father's present occupation in all analyses of this Report which use the socio-economic status of the family as a variable. Information about the occupation of each child's father, or the male head of household, was obtained from 7,723 of the 7,985 Parental Questionnaires which were completed and returned in time for analysis. Of the remaining 262 questionnaires, information about the father's occupation was incomplete or not available in 60 cases, and in 202 cases the children were living in households without a father or male head. Each occupation was grouped initially in the five Social Class categories of the General Register Office's 'Classification of Occupations 1960'. The proportions in each Social Class for the present sub-sample are shown in Table 38. Table 38 Social class distribution A comparison has been made (see Section IIB: 'Description of the Sample') between the above distribution and that of a 10 per cent sample of married women in England and Wales, classified by their husbands' occupations (HMSO, 1966). The differences between the two distributions are not marked and may be due to the following factors: the present sample is of children and not of households; the data were gathered four years later; the present sample is only part of the 1958 cohort and does not include the data from Welsh children. In order to check further on the representative nature of this sub-sample of 'early returns', a comparison was made between the Social Class distribution of this group and that of the 'late returns'. These two groups were also compared for reading ability and social adjustment in school (see Section lIB: 'Description of the Sample'). Some modification of the General Register Office's five Social Class categories was considered advisable for the purpose of socio-economic classification in this Report. Social Class I and II are grouped together, as it was considered that the distinctions between the two categories are minimal in this kind of analysis. On the other hand, in relation to many factors in social analysis including children's development, it has been found that many of the characteristics of workers in clerical and other non-manual occupations are not similar to skilled manual workers, but lie between professional workers and skilled manual workers. As the Registrar General's Social Class III includes both non-manual and the skilled manual occupations, it was decided to separate the skilled manual workers from the non-manual workers in Social Class III. It was not considered meaningful to separate the small proportion of non-manual from manual workers in Social Class IV. The resulting regrouping and nomenclature are shown in Table 68. Table 39 Occupational groupings used in this study The above five occupational groups are the basis of the main socio-economic classifications used in this interim report. Although the father's occupation on its own must be considered a relatively crude index of a family's social, economic and cultural environment, nevertheless, there are meaningful distinctions in this kind of grouping which are highly relevant to children's development. Further information from the questionnaires is available on the social and economic circumstances of the children's families, both at the time of birth and at the age of seven. It is intended at a later stage to combine some or all of this data to provide a basis for a more sophisticated socio-economic classification. The data from the questionnaires containing information about children with no male head of household are included in all tables in the Descriptive Statistics Section and in other sections of the Report where there is no occupational group breakdown. When there is an occupational group classification, this data had to be excluded and is shown on tables under the 'No data' columns; thus, it does not appear in the percentaged totals. 3. Family structure (a) Parental situation Table 40 Parental situation The second category in the above table covers all known situations where children were not being cared for by their own mothers and fathers. It includes those living with only one parent because of illegitimacy, desertion, divorce, separation or death; those with one step-parent as well as children who were adopted, fostered or in care. The very small number who live more or less permanently in hospitals, or similar institutions, are excluded. At the time when the information was obtained, six per cent of the present sub-sample was living in a family or group situation which was different in one or more respects from the majority. It is, of course, difficult to assess the influence and weight of abnormal factors in a child's family situation. For example, a child living with his own parents, one of whom may be mentally or physically seriously ill, may well be growing up in an atmosphere of much greater stress than another child who has been successfully adopted or fostered. Nevertheless, a child not living with both his natural parents will, in many cases, have experienced a period of separation which may have been prolonged; in others, for example, one-parent families, the child may be permanently deprived of a normal home life. Thus, at some time or another he will probably have undergone an upsetting, if not seriously traumatic, experience. There is considerable evidence from research findings (Lewis, 1954; Fraser, 1959; Wynn, 1964; Pringle, 1965; Roe, 1965) to show that in our society the lives of children from incomplete or broken families are adversely affected by a combination of circumstances; these extend not only to emotional but also to social and educational development. Findings relating to the reading ability of this group of children compared with the total sample are discussed in Section VIC of this Report. (b) Number of children in the household The relationship between family size and measures of ability and attainment has been shown in a number of studies. Douglas (1964), for example, found that children of families of four or more in each of the socio-economic groups he examined were at a disadvantage educationally when compared with their peers in smaller families; the differences were most marked for children of manual workers. Of course, there are large families at all economic levels; however, there are proportionately more among unskilled workers, so that being a member of a large family may sometimes be another factor in an interacting set of unfavourable circumstances which have an adverse effect on a child's development. Family size can be defined in various ways. The number of children in the 'normal' family will usually be the same under any meaningful definition, but for some family situations different definitions will affect the number of children to be included. Thus, different indices may be relevant in considering different aspects of children's development. Here it was considered that, as an index of the social climate of the child's home life, the total number of children in the household under the age of 21 would be most meaningful. All children under 21 years belonging to one household were included, irrespective of whether or not they were related to the sample child (e.g. foster children, cousins, etc.). Also included were those living at home only during school holidays or for other short periods. The results presented in Table 41 show the distribution of children by the size of household. Table 41 Number of children in household under the age of 21 years (including sample child) In this sample of seven year olds, nine per cent were only children and seven per cent came from households of six or more. Since the sample was of children and not of households, the figures cannot, of course, be taken as an estimate of the distribution in the general population of numbers of children in households. It is hoped at a later stage to compare these figures with other classifications of family size which can be obtained from the data. 4. Overcrowding In accordance with the definition of maximum occupation density used by the 1961 census, the children in the sample were considered to be living in overcrowded homes whenever there were more than 1½ persons to a room. By this criterion, a family of three living in two rooms is not considered overcrowded, whereas a family of four people or more in the same accommodation is judged to be so. This official definition employs quite a severe standard (HMSO, 1965) and one that many people - even in average income sections of the community - would not regard as acceptable. Data are available from the questionnaires on other main indices of poor housing conditions, such as multiple occupation and lack of basic domestic facilities. However, as this information could not be included at this stage, the present results cover only one aspect of unsatisfactory housing. Table 42 Proportion of children living in overcrowded conditions It can be seen from Table 42 that 11.2 per cent of this sub-sample were living in overcrowded conditions, as defined above. It has been suggested that there is an association between various aspects of children's development and housing conditions (Douglas, 1964; Fraser, 1959), with poor housing being adversely related to children's general progress and school performance. This association, however, is an extremely complex one. Other factors must be taken into consideration, even when family difficulties appear at first to be primarily related to unsatisfactory housing. Conditions of housing vary, of course, in different regions and neighbourhoods, with respect to factors such as age, size and state of repair, as well as sheer availability. Poor accommodation in a slum area may be differently associated with a child's progress than overcrowded accommodation in a better neighbourhood; similarly, much will depend on whether overcrowding is a temporary or permanent situation. The size and composition of the household, as well as the ages of its members, are further important considerations, and the personalities of the parents - particularly the mother - can greatly lessen or increase the strain of bad housing. Studies of problems associated with the establishment of new housing estates have shown the relevance of some of these factors, together with the pitfalls of making generalisations based solely or mainly on housing conditions (Taylor and Chave, 1964). 5. Mobility (a) Moving home Data on this question were obtained from mothers of children who had lived with their parents since birth. The moves reported could be of any distance. Table 43 shows how often the family had moved since the child in our sample was born. Table 43 Number of times family has moved since child's birth About a third of the children have not moved at all in comparison with 64 per cent who have moved once or more. Thirteen per cent have moved twice and 13 per cent of the children have done so three or more times. These totals may well be an underestimate of the amount of moving in the total cohort, since the untraced children who are not, of course, included, are more likely to belong to families who move frequently. Not a great deal is known about the effects of frequent moves on the growth and development of young children. Available data will make it possible to analyse at a later stage some of the factors that may be related to what seems to be a not uncommon experience for about a quarter of the present sample. (b) Changing school The number of schools which the mothers reported the child to have attended since the age of five is shown in Table 44. Table 44 Number of schools attended since the age of five By the time the children were at the end of their infant schooling, 15.4 per cent had changed school once and 2.5 per cent two or more times. This presents a somewhat different picture from the previous table and may indicate that more frequent moves of home take place before the child goes to school, or that the moves are in the same locality so that no school change is involved. Reasons for changes of school vary, as will the effect of the changes on children's educational progress. School factors, as well as those in the child's personality and home background, may play an important part. Further analysis of different factors in this group of children's development and home background will be possible at a future stage in relation to the possible influence of changing school and moving house. 6. School attendance Information was gathered from the schools about the children's attendance from the beginning of the Autumn Term 1964 to the Summer Term 1965, the latter being the time when the educational questionnaires were completed for the majority of the children. Teachers were also asked to give a rating of children's attendance on the Bristol Social Adjustment Guide but, at present, only the objective data are presented. In the present sample 70 per cent of the children were recorded as having an attendance rate of 90 per cent or higher. This is equivalent approximately to 18 days' absence, or less, in a school year. About nine per cent of children had an attendance record of below 80 per cent. Children who have been admitted to a particular school only recently and thus attended for a short period are included; the results for these children will obviously be less reliable. It was noted in checking the completed material that this particular question was not well answered due, perhaps, in some measure to the way in which it was asked. Although it was felt that this was unlikely to have affected the overall results seriously, these should be regarded as tentative in view of this proviso. Table 45 Percentage attendances Many circumstances in the children's home background may influence their records of attendance. Health is an important aspect and is, itself, associated with other factors in the home background, such as standards of parental care and parents' interest in their children's education. On the other hand, the influence of the school is also very important. It seems likely that just as poor school attendance may adversely affect children's progress in school, so poor progress and adjustment in school may in some cases lead to poor attendance. Further, there is some evidence (Douglas and Ross, 1965) that schools differ in the extent to which they succeed in helping children whose educational progress has been adversely affected by absence. 7. Parental interest There is considerable evidence from research that parents' attitudes to and interest in their children's education are closely associated with school performance (Fraser, 1959; Douglas, 1964; Wiseman, 1964). These attitudes and interest are themselves often associated with a number of other factors of general care which are related to children's development. For example, a mother realistically concerned with her child's education is also more likely to take advantage of other available services for his welfare. However, in this section the discussion is confined to the interest that parents show in their children's educational progress. Two assessments were made of parental interest. The first was a subjective assessment by the teachers who were asked to rate the mother's and father's interest with regard to their child's educational progress. This had the limitations of any subjective estimate. On the other hand, the teacher could take into account a number of factual indications of parental interest besides personal contact, such as the children's comments and parents' attendances at school meetings. The second assessment was of a more objective nature. Head teachers were asked if, since September 1964, the parents had taken the initiative to discuss the child, even briefly, with any member of the teaching staff. Despite the advantage of greater objectivity, this question also had its limitation because certain parents, however much they were interested in their child's educational progress, might have been unable to visit the school. The results for the subjective assessment of the teachers are presented in Tables 46 and 47. Table 46 Teachers' ratings of maternal interest Table 47 Teachers' ratings of paternal interest The difference between the assessments of the mothers' reported interest and the fathers' in the 'can't say' category would be expected as so many fathers cannot - even if they wish - visit their children's school during the day because of working hours. Even so, teachers felt able to comment on the interest of 63 per cent of the children's fathers. Approximately 75 per cent of the mothers and 46 per cent of the fathers were rated as showing some interest or being very interested in their child's educational progress. It is worth noting that the proportion of mothers (15.2 per cent) and of fathers (16.2 per cent) reported as showing little or no interest was almost the same. This seems to indicate that teachers were able to give an equally definite rating for each parent in this category despite fathers' greater difficulty in making daytime visits. The teachers' ratings show a significantly different tendency for parents to take a greater interest in the girls' educational progress than in the boys'. This is rather unexpected, as it contrasts with parental attitudes towards the education of older children; then the reverse is the case, with boys' scholastic progress being considered of much greater importance and greater numbers of boys staying on at school beyond the compulsory school leaving age. Although such trends as do appear tend to favour the girls, the difference in the parents' interest shows rather a complex pattern with no clear-cut results. It is also possible that this finding is due to some bias in the teachers' ratings. Since girls' school performance is better than boys', the teachers' ratings of parental interest may be influenced by a knowledge of the child's performance; i.e. if a child is doing well at school, the teachers may make an assumption of parents' interest in the child's educational progress. With regard to the second more objective criterion used to assess parental interest, it can be seen (Table 48) that during the current school year 57 per cent of parents took the initiative to discuss their children with one of the teaching staff. Table 48 Parental approach to school staff It would be expected that contact between staff and parents would be easier at the infant school stage, when it is more likely that parents will accompany their children to and from school and the general atmosphere is more informal. Even so, it seems that 43 per cent of the parents of this sample had not initiated any contact with school staff during the period in question. It can be considered that parental interest in their children's education will to some extent be a reflection of the school's own attitudes and interest in the children's families. A further question was asked to find out more about the amount of personal contact between staff and parents, initiated by the staff. Head teachers were asked if there had been any discussion with the parents about the child 'at the instigation of you or your teaching staff'. The replies to this question are shown in Table 78. Table 49 School instigation of discussion The staff had instigated discussion with 25.5 per cent of the children's parents. There is a difference of four per cent in the proportion of boys and girls discussed. This difference was highly significant statistically. It may, in part at least, be due to the fact that boys in the sample made less satisfactory progress in reading and were less well-adjusted in school (see Section V6). Hence, teachers would feel greater concern about their progress and thus be more anxious to talk to parents about it. There will be some parents who, as well as having initiated a discussion themselves, will also have discussed their child at the teacher's instigation. After allowance had been made for the overlap between these two groups, a final total was estimated of the number of children whose parents and teachers were reported as having had some discussion about them. This total amounted to 65 per cent of the children in this sample; thus, from September to near the end of the school year, about two thirds of the seven year olds in the Study had been jointly considered by their teachers and parents. Further analysis of parents' interest in their children's educational progress and its relationship to other factors will be made at a later stage. However, an analysis of reading ability and parental interest is made in Section VIC. 8. Parental aspiration As an indication of a more general attitude to education, mothers were asked if they would like their child to be able to stay on at secondary school after the minimum school leaving age. Table 50 Parents wanting children to stay on at secondary school after minimum school leaving age The statistical evidence shows that there is a difference between the answers given by parents of boys and by parents of girls. More parents of the boys than of the girls wanted them to stay on at school after the minimum school leaving age. It will be noted that of the criteria of parental interest so far examined, this is the only indication of more interest being taken in boys' education than in girls'. This question was answered in the affirmative by 81 per cent of the mothers. Of the remainder, only four per cent gave an unqualified 'No'. Of course, at this early age when the actual decision is many years ahead and less realistic, a socially more acceptable answer is likely to be given. Very probably, parental attitudes will undergo changes during the years until this question confronts them as an immediate issue. Nevertheless, the present result would seem to indicate a favourable climate of opinion towards continued secondary education beyond the statutory leaving age. 9. Summary (a) Most of the data in this section were obtained from 7,985 completed Parental Questionnaires. The statistically significant but not marked bias in this sub-sample suggests that findings relating to adverse conditions are likely to underestimate the incidence of such circumstances in the total cohort. (b) The present occupations of the children's fathers were grouped according to the Registrar General's five Social Class divisions, and the distribution was compared with those of the 'late returns' and of a national sample. The Social Class divisions formed the basis of a regrouping of the fathers' occupation into the five occupational groups which are used for the main socio-economic classifications in this Report. (c) Approximately six per cent of the children were not living with both of their natural parents. (d) Approximately nine per cent were living in households in which they were the only child, 77 per cent in households of two, three or four children, and 14 per cent in households where there were five or more children. (e) Approximately 11 per cent of the children were living in 'overcrowded' conditions. (f) About two thirds of the children had moved home once or more since they were born. Approximately 13 per cent had moved twice and a further 13 per cent three or more times. (g) Since the age of five, 15.4 per cent of the children had changed school once and 2.5 per cent two or more times. (h) Of 10,645 children for whom this information was available, 70 per cent were recorded as having an attendance rate of 90 per cent or higher. About nine per cent of the sample had a record of below 80 per cent attendance. (i) Some 57 per cent of parents had approached a member of the school staff in the current academic year in order to discuss their children. There was no significant difference between the parents of boys and girls in this respect. Teachers' ratings of parental interest on the other hand seemed to show that the girls' parents were more interested in their children's educational progress than the boys'. However, it is possible that the teachers' ratings may have been influenced to an extent by the better progress of the girls. Of the 26 per cent of children whose teachers had instigated discussion with parents, there were significantly more boys than girls. (j) About 81 per cent of the parents said they would like their children to stay on at school after the minimum school leaving age. References DOUGLAS JWB The Home and the School (McGibbon and Kee) (1964). DOUGLAS JWB and Ross JM 'The Effects of Absence on Primary School Performance'. Brit. J. Educ. Psychol. 35, pp. 28-40 (1965). FLOUD JE, HALSEY AH and MARTIN FM Social Class and Educational Opportunity (London: Heinemann) (1957). FRASER ED Home Environment and the School (University of London Press) (1959). GENERAL REGISTER OFFICE Classification of Occupations (HMSO) (1960). GENERAL REGISTER OFFICE Census 1961: Occupation Tables, England and Wales (HMSO) (1966). LEWIS H Deprived Children, the Mersham Experiment (Oxford University Press) (1954). MINISTRY OF HOUSING AND LOCAL GOVERNMENT Report of the Committee on Housing in Greater London (HMSO) (1965). PRINGLE ML KELLMER Deprivation and Education (Longmans) (1965). ROE MC Survey into Progress of Maladjusted Pupils (ILEA) (1965). TAYLOR Lord, and CHAVE S Mental Health and Environment (Longmans) (1964). WISEMAN S Education and Environment (Manchester University Press) (1964). WYNN M Fatherless Families (Michael Joseph) (1964).
F. BEHAVIOUR AND ADJUSTMENT 1. Introduction This section is concerned with aspects of the children's behaviour and adjustment in school and home as reported by their teachers and mothers. Here, as in other sections of this interim report, the total number of children included in different tables varies. The reasons for this have already been explained. The evidence (see Section IIB: 'Description of the Sample') that the children for whom Parental Questionnaires had been received in time for inclusion in the present Report were better adjusted in school than those for whom a Parental Questionnaire had not been received, is particularly relevant to any interpretation of results in this section. It is reasonable to assume that the former group would also be better adjusted at home. Thus, where results are presented here which are derived from mothers' reports of children's behaviour, the figures may underestimate to some extent the incidence in the total sample, when the behaviour in question is indicative of difficulties. However, there is no expectation that any such bias will affect comparisons between the sexes. There is one other general factor which should be stressed. Although all figures presented in this Report should be seen in relation to the circumstances in which the information was collected and the way in which questions were put, it is particularly important to have this in mind when assessing the validity of data about behaviour and adjustment. An attempt has been made to reduce reporting error or bias by gathering, wherever possible, current rather than retrospective data. Nevertheless, in many questions what is being assessed is essentially a facet of a relationship between two people (a mother and a child, a teacher and a child) rather than simply an objective fact about a child. In the case of the mothers' reports of the children's behaviour, these were given to a third person, usually a health visitor. This factor does not invalidate the results, but it does mean, particularly when statistics are seen as incidences, that the total framework must be borne in mind. Thus, for the sake of convenience and brevity, a sub-section is here entitled 'Children's Happiness at School'; in essence, however, the figures tell what the mothers reported about their children's happiness at school when posed a particular question on one occasion by a health visitor. It is possible that if the enquiry had been carried out by post, if the mothers had had time to consult their husbands, or if the question had been put by or to teachers, by psychologists or by a trained team of interviewers, then the results obtained may have been somewhat different. On the other hand, it would be foolish to disregard the results on this account unless there was evidence that a question on this topic, and posed under these circumstances, produced a very distorted picture. No method of enquiry is faultless, but each has limitations. Where there is possibility of distortion or bias, the implications are considered. Information was obtained about the children's behaviour and adjustment in order, first, to test certain hypotheses about general adjustment, as well as particular facets of it, in relation to other factors; to examine, if comparable information can be collected on these children as they mature, the ways in which patterns of behaviour shown at seven years of age change or remain constant at later ages; to provide some data on the behaviour of a representative group of seven year olds; and to compare the sexes at this age. Although the last point is here mentioned separately, it should, strictly, be included with the first since a relationship is examined between two variables, namely behaviour and the sex of the children. In the present section only the last two points are covered: descriptive data are given and the sexes are compared. Sub-section 2 deals with the children's behaviour and adjustment as assessed by their teachers; and in sub-section 3 results are presented for different aspects of behaviour as reported by mothers. 2. Behaviour and adjustment - teachers' assessments (a) Settling down period on starting school The head teachers were asked whether on starting school the children settled down within a month; within 1-3 months: or remained unsettled after three months. Since the children's present schools will not always have been their first ones, the results do not relate to the settling down period on first starting school; nevertheless, over 82 per cent of the children had been to one school only (see Section VF: 'Environmental Factors'). There will be some loss of reliability due to the fact that the data are retrospective by some two years in the majority of cases. There is also the possibility of some distortion of the results for individual children. Thus, if a child is now quite 'normal' and well-adjusted, there is perhaps less likelihood of an unsettled period two years ago being remembered. The results are presented in Table 80 and relate to children in 'ordinary' maintained and independent schools. The large number of children listed as 'Can't say' is most likely to be a reflection of the difficulty of obtaining retrospective information from schools because of staff changes. It will be seen that the figures for these children have not been percentaged, since there is no reason to believe that if the information had been available, they would not have been distributed proportionately in the other categories. Where a child had been at a school less than three months, the full range of possible answers was not open to the head teachers. These children were therefore coded as 'Inapplicable'. It will be seen that, as judged by this criterion, the great majority of the children (94 per cent) had settled down on starting at a first or a new school within three months. The difference between the sexes was highly significant statistically. The trend, furthermore, persisted through the three categories. The girls were reported to have settled down in school more quickly than the boys. Table 51 Settling down period on starting school The possibility of some distortion of the results for individual children has already been discussed. It was suggested that a child's present behaviour may influence a teacher's memory of a facet of his behaviour two years before. It may be that such retrospective distortion is partly or wholly responsible for the present findings. Thus, there is evidence that the boys in the sample were less well-adjusted in school (see sub-section 2(b) of the present section). The difference between the sexes demonstrated above may, then, reflect their present behaviour. However, the fact that the same difference between the sexes is evident when the mothers' reports of the children's settling down period on first starting school are examined (see sub-section 3(b)) makes it less likely that the present findings are spurious. Nevertheless, the possibility cannot be ruled out that the same factors were operating in the mothers' retrospective reporting. Under the circumstances, the evidence of a sex difference here must be regarded as tentative at the present stage. Further research is needed to confirm the results. (b) Bristol Social-Adjustment Guide scores The difficulties of obtaining a reliable and valid measure of adjustment are very considerable; much less progress has been made by psychologists in this field than, for example, in measuring intellectual abilities and attainments. Furthermore, in a study of this magnitude of seven year olds, one is confined to an assessment of the children's adjustment by some other person. As has been mentioned earlier, one is essentially obtaining a measure of a relationship between two persons; in this case a child and his teacher. This is not irrelevant since social adjustment cannot be viewed in isolation, but in individual cases there will be differences between a child's adjustment as seen by different teachers. Such differences, however, are less likely to be marked if, instead of obtaining global assessments, one asks specific questions about particular aspects of behaviour. The Bristol Social-Adjustment Guide (Stott, 1963) in effect does this. The Guide consists of a large number of descriptions of behaviour and the teacher is asked to underline those descriptions which best fit the child. Aspects of behaviour which appear to be deviant are specially coded and it is therefore possible, by summing these items for an individual child, to obtain a quantitative assessment of social adjustment in school; the higher the score, the more manifestations of deviant behaviour have been noted by the teacher. It is obvious that a child's behaviour can be deviant in many ways. Thus, he may be unforthcoming, aggressive or over-demanding. A total score masks all these different facets and therefore its use is only warranted as a relatively crude assessment which may be useful in certain kinds of analyses or as a first step before a more detailed and sophisticated analysis. At the present stage in this Study, there has been time only to examine total scores. For convenience, and as a first step, it has been decided to classify the children as proposed by Stott: children with a score from 0 to 9 are grouped and called 'stable'; those with a score from 10 to 19 are called 'unsettled'; and the children with a score of 20 or more are called 'maladjusted'. The groups are therefore operationally defined in these terms. Thus, no assumption is made that children with a score of 20 or more are maladjusted in any sense other than that defined above. The results are presented in Tables 52 and 53. A word of explanation is needed about these figures. In both tables, the numbers involved are several hundred less than the number of Guide scores available. This reduction is due to technical difficulties arising from the data processing; there is no reason to believe that the children excluded are different from the ones included in any specific way. In Table 52 the figures for children in all categories of school have been extracted from a more detailed table not presented in this Report, in which there was a breakdown by occupational group and Sex. Where there were no data on either the sex of the child (very few) or the occupation of the father, the children were excluded. This sub-sample is known to be biased in that they are better adjusted in school than those where the occupational group was not known (see Introduction to present section). Although this bias is not expected to affect a comparison of the sexes, the combined figures would be biased. However, the figures given in Table 53 include all the children in schools (including special schools) in the present sample for whom a score was available, and there is no evidence of bias in this group. Table 52 Bristol Social-Adjustment Guide Scores - boys and girls Table 53 Bristol Social-Adjustment Guide Scores - combined sexes The comparison between the sexes in Table 52 shows a highly significant difference; the girls were better adjusted in school. Approximately 60 per cent of the boys were 'stable' as against 74 per cent of the girls; on the other hand, nearly 16 per cent of the boys were 'maladjusted' compared with eight per cent of the girls. The results, strictly interpreted, apply only to this form of assessment. Nevertheless, in view of the very marked differences shown, the firm conclusion is warranted that at this age boys show more indications of poor adjustment in school than girls. These results confirm those found by Crawford (1966), who used the Guides in a Liverpool study on 773 boys and girls aged seven and eight years. Two points should be borne in mind. First, the results derived from total scores may mask other factors. Thus, despite an overall difference between the sexes, it may be that detailed examination of the data will show no differences between the sexes in particular facets of adjustment; or, even, that girls show more deviant behaviour in some ways. For example, boys may be more aggressive or over-demanding and girls may be more timid or withdrawn. This further detailed study of the data will be undertaken later. Secondly, since this sub-sample is known to be biased in relation to adjustment in school, the percentages shown in Table 52 should not be taken as representative figures for the sexes. In Table 53 the results are presented for a larger sample of children, since no other results are available for this test on a national sample of this size. There is no expectation of bias in these results. It was stressed earlier that the definitions of 'stable', 'unsettled' and 'maladjusted' are operational ones and refer only to scores on the Guides. The grouping used by Stott and others has been adopted as a first step and to permit comparison with other studies. To say, therefore, that the present findings indicate that over 13 per cent of seven year olds are maladjusted is to say no more than that over 13 per cent of seven year olds are likely to obtain a total score of 20 or more on the Social-Adjustment Guides. It would be taking too extreme a view to say that the term 'maladjusted' means all things to all men. Nevertheless, the difficulties of establishing some reliable criterion of maladjustment against which to validate an objective score such as is derived from the Guide are considerable. From a practical viewpoint there is an obvious need to establish the proportion of school children who would benefit from some form of psychological help or from psychiatric treatment. The need for such information was underlined by the Underwood Committee (HMSO, 1955) and the Scottish Education Department (HMSO, 1964). It is hoped at a later stage to throw some light on this matter with the data available on this cohort of children; and also to examine in some detail the factors which are associated with maladjustment. 3. Behaviour and adjustment - mothers' reports (a) Happiness at school The mothers were asked: 'Is the child happy at his/her present school?' The answers were pre-coded by the interviewers into the categories shown in Table 54. The results are for children in all types of school. Table 54 Happiness at school The difference between the sexes is highly significant. A strict interpretation of the results would be that girls at this age are reported by their mothers to be happier at school than boys. However, there is some indirect evidence from the present Study to indicate that this may be interpreted as a meaningful difference between the sexes in the extent to which they are in fact happy at school. Thus, the independent assessments of the teachers showed that girls are markedly better adjusted at school, and it is reasonable to assume that this would be reflected in their 'happiness'. Nevertheless, the subjectivity of this question imposes a need for caution, particularly in viewing the percentage figures in the three categories. One must consider the possibility that the mothers may have seen their answers as a reflection upon the school, although there is no reason to believe that any such tendency would have acted in one direction only. Further, it had to be left to the interviewer to make a reasoned judgement about which category to code in the light of the mother's answer. However, within the limitations of this type of question, the overall results are extremely encouraging from an educational standpoint and do credit to the teachers and schools concerned. (b) Settling down period on first starting school Information about the length of the settling down period on first starting school was obtained from the mothers. The same question was also put to the head teachers (see sub-section 2(a) of the present section), although their replies would relate to the children's settling down in their present school, which would not necessarily have been their first one. Some of the children attended a private or a local authority nursery class or school, and the results for these are presented in Table 55; and relate to the settling down period on commencing nursery schooling. In Table 56 the results are given for children who did not attend a nursery class or school. Although these results were obtained by sorting punched cards, it was possible to include a few (52) additional children so that the total for Tables 55 and 56 are slightly higher than for tables in the remainder of this section. The children involved were in all categories of school. Table 55 Settling down period on first starting school (nursery class/nursery school attenders) Table 56 Settling down period on first starting school (non-nursery class/nursery school attenders) It has not been possible at the present time to examine statistically any differences between the settling down period for those who attended a nursery class or school and those who did not. Inspection of the tables suggests no very marked overall differences, but a straightforward comparison of these tables is likely to be misleading; a more sophisticated analysis is needed which takes into account other factors which may be associated with nursery school attendance. In both tables, however, the results indicate that girls are considered by their mothers to settle down more quickly than boys on first starting school. Although confirming the results obtained in a comparison of the sexes on the teachers' assessments (see sub-section 2(a) of this section), the same reservations expressed there apply to the above results. The overall figures are not strictly comparable with the teachers' assessments given in Table 51, both because of the bias in the present sub-sample and also because the teachers were not necessarily assessing the children's settling down period on first starting school. Furthermore, any differences between Tables 51 and 56 may be due to the fact that mothers and teachers see the situation from different standpoints. (c) Developmental difficulties It was decided to seek information from mothers about developmental difficulties which occurred largely in the home. Three steps were taken to increase the reliability of the data. First, mothers were asked to say whether any of the difficulties had occurred over a prescribed period; namely, 'during the last three months'. Secondly, specific aspects of behaviour were selected so that it was reasonable to ask for a straightforward 'Yes' or 'No', although a 'Don't know' reply was included in the coding. Thirdly, if the behaviour had occurred only during a period of acute infection, the interviewers were asked to code the answer as 'Inapplicable'. The results are presented in Table 57 and relate to all children in the present sample for whom information was available, whether they were in school or not. Percentage figures only are given here and they refer to children whose mothers reported that the behaviour in question occurred during the three months prior to the interview. The small numbers of children where 'Don't know' or 'Inapplicable' was coded have been excluded from the percentaging. Although the results are presented in one table, the categories are, of course, not mutually exclusive and a child may appear under more than one category. Table 57 Developmental difficulties occurring over a three month period In examining the table, one must take account of two main factors. First, the bias in this sub-sample, discussed in the introduction to this section, may mean that some of the figures quoted underestimate the incidence in the total population of seven year olds in the country. Nevertheless, it would be unwise to over-stress the effect or the importance of this bias in considering the value of the results. Its effect is unlikely to be marked and it should be viewed within the framework of the limitations inevitably imposed upon data of this kind collected in a large survey. One is dependent upon the mothers' reports, which may be influenced by such factors as their own personalities, education, tolerance level and the attitude of the interviewer; further, the behaviour was reported to have occurred over a particular period of time which, if changed, may have produced somewhat different figures. Thus, the results in relation to many other kinds of data are of a crude nature. The fact that figures are given to one decimal place must not be taken as implying that exact information is being obtained. On the other hand, no clinician, no teacher, no social worker would claim that information he obtained is without its own limitations. The value of data gathered on a large group of children is that it provides workers in this field with a frame of reference, a 'backcloth' against which to view the normality or deviance of behaviour manifested by an individual child, or a group of children, of this age. It will also permit the changing pattern of behaviour to be studied as the children grow towards maturity, if - as is hoped - further information can be gathered upon them. Neither the limitations discussed, nor the bias, are likely to affect the overall pattern of the results, the relative incidence of different difficulties. Thus, it will be seen that the reported occurrence of a temper tantrum or of faddiness is not uncommon; about 30 per cent of the children were reported to have shown these aspects of behaviour. Sleepwalking, on the other hand, is reported as occurring only in a small minority of children of this age. The second main factor which must be borne in mind is that the figures apply to behaviour in the home as seen by mothers. For some aspects of behaviour, such as difficulties in sleeping, this is the only appropriate framework within which to view the data. However, other behaviour, such as food-faddiness or temper tantrums, manifests itself in other situations - in school, for example. It is highly likely that if information on these aspects of behaviour had been obtained from teachers, different results would have been obtained because children behave differently in different circumstances. None of these aspects of behaviour in isolation should be seen as an indication of poor adjustment. Nevertheless, it is likely that some are more symptomatic of emotional disturbance than others. In addition, a number of workers have shown that a simple summation of reported behavioural difficulties for individual children provides some measure of adjustment which is associated with poor educational performance and other factors. It is intended to investigate this further at a later stage and to see which difficulties and which combination of difficulties are most closely related to other factors, both in the home and in the school. With regard to a comparison between the sexes in the reported occurrence of the 'developmental difficulties' set out in Table 57, it will be seen that for only three items was there a significant difference. The 'P' values are given, and for reasons which are explained in Section IV: 'Statistical Analysis', the level of 0.01 has been selected as the one which in this Report is used as the criterion of statistical significance. There is clear evidence from the present findings that, as reported by mothers, more boys than girls show reluctance to go to school; that more boys have temper tantrums at this age; and that more girls are considered to have poor appetites. (d) Other aspects of children's behaviour In addition to the behaviour reported by mothers which has been discussed in the previous sub-section, the mothers were asked about other aspects of behaviour. Although in the Parental Questionnaire and in this Report the two groups of items are presented separately, there is no clear-cut distinction between them from a psychological viewpoint. The developmental difficulties already discussed were so described because this seemed an appropriate term. Nevertheless, some of the aspects of behaviour discussed in this section might equally well have been described as difficulties. One major reason for separating these two groups of items in the questionnaire was that for some aspects of behaviour it seemed appropriate to ask for a straightforward answer as to whether or not the behaviour had occurred during a prescribed period; for the behaviour discussed here it seemed more appropriate to enquire whether 'at the present time' it happened 'frequently', 'sometimes' or 'never'. The results are presented in Table 58 and relate to all children in the sample for whom the information was available, whether they were in school or not. The figures are given in percentage form and where an answer was coded 'Don't know' or 'Inapplicable', it has been excluded from the percentaging and also from the table. Table 58 Other aspects of children's behaviour Of course, the same factors as were discussed in the previous sub-section should be considered in taking account of the above results. A number of aspects of behaviour appear to be common at this age: disobedience and fighting with other children were reported to occur frequently or sometimes in about 60 per cent of the children. In contrast, twitches or mannerisms were reported in less than eight per cent. The reported incidence of nail-biting (26 per cent) was about half that obtained by Birch (1955) in a local study utilising teachers' reports and more objective criteria. Preferring to do things alone rather than with others stands out as an aspect of behaviour which was reported to occur frequently in a substantial minority of the children (over 22 per cent). In all but three items, the difference between the sexes was statistically significant or highly significant when tested for trend. The two most marked differences were both in aggressive behaviour; twice as many boys as girls were reported as destroying their own or others' belongings; and about 70 per cent of the boys fought with other children, whilst only 44 per cent of the girls did so. In general, it was the boys who more often showed deviant behaviour, but more of the girls were nail-biters and sucked their thumb or finger during the day. As with the behaviour difficulties described in the previous sub-section, it is intended to examine the aspects of behaviour discussed here in more detail at a later stage. 4. Summary (a) On starting at their present school, 94 per cent of the children in 'ordinary' maintained and independent schools in the sample were judged by their teachers to have settled down in school within three months. There was evidence that girls settled down more quickly than boys in infant classes, but in view of the fact that the data were retrospective and may have been influenced by present behaviour, this conclusion was regarded as tentative. (b) The evidence was very clear that girls at this age in all categories of school are markedly better adjusted than boys when a relatively crude overall assessment is made in terms of the total score of the Bristol Social-Adjustment Guide. It was noted that this overall assessment may mask other differences between boys and girls when more specific aspects of behaviour are examined. This more detailed analysis will be done at a later stage. (c) Approximately 93 per cent of a sub-sample of 7,985 children in all categories of school were reported by their mothers to be happy in school. There was also a highly significant tendency for more girls to be reported as happy at school than boys. (d) Of the children who had attended a private or local authority nursery class or nursery school, some 94 per cent were reported by their mothers to have settled down within three months of first starting nursery schooling. Of the children who had not attended a nursery class or nursery school, about 91 per cent were reported by their mothers to have settled down within three months of first starting school. No statistical analysis was made to test any difference between these two groups of children in relation to the settling down period because a straightforward comparison might have been misleading. No conclusions were therefore drawn. The evidence from a comparison of the sexes in both groups, however, suggested that girls settled down more quickly than boys. (e) Percentage figures were given for the reported incidence of nine 'developmental difficulties' for boys and for girls. In only three was there a significant difference between the sexes: there was evidence that more boys than girls are reluctant to go to school; more boys than girls have temper tantrums; and more girls are felt to have poor appetites. (f) Information was gathered from mothers about 14 other aspects of children's behaviour. In all but three items there was a significant difference between the sexes in reported incidence. In general, boys at this age are more often reported to show deviant behaviour at home. References BIRCH LB The Incidence of Nail-Biting Amongst School Children. Brit. J. Educ. Psychol., Vol. XXV (1955). CRAWFORD A Department of Psychology, Liverpool University (unpublished communication) (1966). MINISTRY OF EDUCATION Report of the Committee on Maladjusted Children (HMSO) (1955). SCOTTISH EDUCATION DEPARTMENT Ascertainment of Maladjusted Children (HMSO) (1964). STOTT DH The Social Adjustment of Children; Manual to the Bristol Social-Adjustment Guides (University of London Press Ltd.) (1963).
VI: INTERRELATIONSHIP OF THE VARIABLES A. EDUCATIONAL FACTORS 1. Reading and arithmetic in relation to occupational group (a) Introduction There is overwhelming evidence for a relationship between general socio-economic factors and general educational progress. What are therefore now needed are more investigations of particular aspects of the environment in relation to particular facets of educational progress and, indeed, to child development in general. At the present stage in this Study there has not been time to examine many aspects of the environment in relation to educational factors; and such statistical analyses as have been carried out have been relatively straightforward. More elaborate analyses will be carried out later. In this section two facets of educational performance - in reading and in arithmetic - are analysed in relation to the occupational group of the children's fathers or the male head of the household. (b) Occupational group and attainment in reading The ability to read is of central importance to children's progress in school. Once mastered, it is a skill which opens many doors. Whilst most schools now lay more stress upon activities, upon learning through experience and upon oral work in the classroom than was once the case, it remains true that a child who finds reading difficult is gravely handicapped in almost all spheres of his school work. Other investigations have shown a relationship between the socio-economic level of a child's family and his reading ability, although few studies have been made of this relationship for children in infant schools. It is not difficult to think of reasons for this strong relationship. In general, parents in higher socio-economic groups are more intelligent than those in the other socio-economic groups; thus, their children will tend to be more intelligent and, as a result, better readers. Although no general consensus of opinion exists about the precise contribution which heredity makes towards intellectual functioning, there can be no reasonable doubt that it does playa part (Burt, 1960). In addition to heredity, and interacting with it, is the influence of the environment in which a child grows. A home in which books and reading material of all kinds are an integral and valued part of daily life is one which is most likely to lay the foundation for the rapid acquisition of reading skill by a child. Reading for him is seen as something which gives his parents enjoyment and which has a purpose. He comes to school, then, wanting to read and may already have made a start. Motivation is a potent factor, but the influence of the home environment does not rest there. The written word is a symbol which enables the author to convey his thoughts, his ideas and concepts to the reader. If the reader finds difficulty in understanding these thoughts, then any ability to articulate the words will be of little value. True reading ability, then, is more than merely a skill in decoding written symbols; it is an ability to use this skill in understanding what the author is trying to convey. It follows that a home which fosters this true reading ability best is one in which the level of verbal expression is at its highest; where there is a feeling for the spoken word as a tool for conveying precise meaning; where the vocabulary is rich and varied; and where children are stimulated by questions about the world around them and by explanations appropriate to their age. The factors which have been discussed are complex and extremely difficult to quantify, particularly in large scale investigations such as the present one. Nevertheless, it is reasonable to assume that in general the socio-economic level of the home, much easier to categorise, will be related to these factors. A great deal of information is known about the occupation and education of the parents and the grandparents of the children in the present Study. It is hoped to use this information to arrive at some measure of the social-educational-economic background of the children in order to investigate at a later stage the relationships between this and other factors in the children's development. Further, it is hoped within the limitations set by the relative crudity of some of the data - virtually inescapable in any large survey - to examine the relationship between particular aspects in the environment and particular facets of children's development. This, however, needs a far more elaborate statistical analysis than there has yet been time to carry out. As a first step, the relationship is here examined between the occupational groups of the children's fathers-as an index of the socio-economic level of the family-and one aspect of children's development, namely reading ability. In sub-section (c) of this section the same analysis is made of arithmetical ability. The results are given in Table 59 for all the children in the present sample in 'ordinary' maintained and independent schools for whom all the requisite information was available. As the index of reading ability, the children's scores on the Southgate Reading Test are used because of their objectivity. Children with a score of 28 or more out of a possible 30 are grouped and operationally defined as 'good readers'; those with a score less than 28 are grouped and called 'medium and poor readers'. Percentage figures only are given in the table below. Table 59 Southgate test score and occupational group of the father The difference between the two reading groups in relation to occupational group is highly significant. There is thus clear evidence that in children of this age the lower the 'status' of the fathers' occupations, the poorer the reading attainment of the children. However, there is statistical evidence that this 'falling off' of reading attainment through the occupational groups is not uniform. An inspection of the table and Figure 8 suggests that this is because there is little difference between the reading performance of children whose fathers have non-manual occupations (occupational groups 1 and 2), but that these children are sharply differentiated from the children whose fathers are skilled, semi-skilled or unskilled workers (occupational groups 3, 4 and 5). Nevertheless, the figures indicate that reading performance continues to 'fall away' through the three manual groups. Although the overall tendency is clear and statistically highly significant, and the fact of some departure from this general tendency has also been shown, more sophisticated analysis is needed to establish the validity of the speculations advanced in the above paragraph. This analysis will be done at a later stage. It will be recalled that the Southgate Test is a measure of word recognition and not of reading comprehension. It may be that the relationship between occupational group and reading for comprehension is stronger than that between occupational group and word recognition. Later analysis of the results for the two other assessments of reading ability obtained in this Study should throw some light on this question. Figure 6 Southgate test score and occupational group of father Figure 7 Problem arithmetic test performance and occupational group of father (c) Occupational group and attainment in arithmetic Much less attention has been given by research workers to the relationship between arithmetic ability and environmental factors, compared with the number of studies of reading. This is understandable since reading plays a more important part in determining children's general progress in school than does arithmetic. Nevertheless, arithmetical ability does impinge upon many other school subjects and it is probably true to say that its importance both in school and in later life will grow along with the rapid technological advances of modern life. It was decided, therefore, to investigate the relationship between the children's arithmetical ability and environmental factors. As with reading ability, the first step was to examine arithmetical ability in relation to socio-economic status. Again, the occupational group of the father is taken as the index of socio-economic status at this stage. The Problem Arithmetic Test score is used as the measure of arithmetical ability. The results are given in percentage form in Table 60. All children in the present sample who were in 'ordinary' maintained and independent schools are included where all the requisite data were available. Children with a score on the Problem Arithmetic Test of 8 or more out of a possible 10 are grouped and operationally said to have 'good arithmetical ability'; those with a score of 7 or less have 'medium and poor arithmetical ability'. Table 60 Problem arithmetic test score and occupational group of father As the table shows, the proportion of children with 'good arithmetical ability' in the present sample fell from about 28 per cent in occupational group 1 to about 14 per cent in occupational group 5. The evidence from the statistical tests is that this tendency is highly significant and one may conclude that in children of this age performance on a test of problem arithmetic is related to the socio-economic status of the family: the lower the status of the fathers' occupational group, the lower the scores of the children. However, as with the corresponding analysis of reading ability, this overall tendency for performance to 'fall away' through the occupational groups is not uniform. An inspection of Table 60 and of Figure 7 showing the pattern of results for children of 'good arithmetical ability' in graph form, suggests that there is a differentiation between children in occupational groups 1,2 and 3 (children of non-manual workers and skilled manual workers) but little, if any, difference between the performance of children in occupational groups 3, 4 and 5 (children of manual workers) in terms of the criteria adopted. However, the comments in the preceding paragraph must remain speculative at this stage. Further statistical analysis must be carried out to clarify the position. (d) Summary The evidence from the two preceding sub-sections points clearly to a relationship between the occupational group of children's fathers and children's attainment both in reading and in arithmetic at the age of seven. In both analyses there was evidence that this general relationship is not uniform from one occupational group to another. These should be seen, however, as two separate analyses; it is not valid to make direct comparisons between the two, even at the speculative level. It will be recalled that the criterion of 'good' reading ability included about 40 per cent of the children, whereas the children with 'good arithmetical ability' were the top 19 per cent of children, as assessed by their scores on the Problem Arithmetic Test. At a later stage comparisons will be made between reading and arithmetic ability in relation to socio-economic factors, but this will necessitate different and more elaborate statistical techniques. References BURT C Interaction of Heredity and Environment in Regard to 'Measured Intelligence'. Brit. J. of Educ. Psychol., Vol. XXX, Part 3 (1960).
B. SCHOOL VARIABLES 1. Length of schooling, educational attainment and adjustment in school (a) Introduction There is evidence from other investigations, reviewed by Pidgeon (1965), that younger children in an age group are at a disadvantage educationally. Thus, summer-born children have been shown to predominate over autumn-born children in the lower streams of schools; in special schools for educationally subnormal children; and in remedial groups for backward readers. It has been suggested that one of the important factors in this situation is the length of schooling received. Within a given administrative situation most children start their schooling at specified times in the school year, usually at the beginning of a term, according to their dates of birth. In some areas children commence at the beginning of the term in which they attain the age of five or, even, at the beginning of the school year in which they become five; in others they start school at the beginning of the term following their fifth birthday. However, whatever the procedure, except for one intake of children per year, older children start school first and the younger children follow at intervals during the school year. This means that the older ones have the advantage not only of greater maturity by reason of their age but also of a longer period in school. Under most circumstances the assessment of the relationship between the length of schooling and subsequent educational performance presents some difficulties because the children who spend the most time in school are also the older ones. Due allowance must therefore be made for the age of the children and, ideally, allowance must also be made for any possible interaction between the age of the children and the length of schooling they have received. In the present Study, however, the children are all of the same age. An opportunity presented itself, therefore, for examining the educational attainment and social-adjustment of children nearing the end of their infant schooling who were of the same age but had had varying lengths of schooling. It was decided to restrict the analysis to two major groups: those who had started school when aged between 4 years 6 months and 4 years 11 months, and those who started school between the ages of 5 years and 5 years 6 months, as reported by their mothers. Since all of the children in the cohort had their fifth birthday in early March 1963, the latter group would have started school in the summer term 1963, and it is known that the majority of the former children started the term before, in January 1963. These two groups are here referred to as 'early starters' and 'late starters'. Children who had attended a nursery class or a nursery school were excluded from the groups. In addition, it was decided to take into account the occupations of the children's fathers since this may be an important variable. The occupational group of the father is an indication of the socio-economic status of the family; further, since in general parents of higher socio-economic status are also more intelligent, their children will in general be more intelligent than children of parents in lower occupational groups. Thus, it is important to ascertain not only whether children with the advantage of an early start in their schooling are doing better in school some two years later, but whether any such relationship exists independent of the socio-economic status of the children's families. It may be, for example, that the children of professional parents, because of their higher intelligence and the intellectual stimulation in the home, are able to make good the deficit of later starting when compared with children of the same socio-economic status who start school earlier. On the other hand, such children may, because of these same characteristics, be able to maintain an advantage over children of the same socio-economic status who have started later. Another reason why allowance should be made for the occupational group is that children whose fathers are in, say, occupational groups 4 and 5 are more likely to live in areas where schools are crowded and where there is a particularly acute teacher shortage. In such areas, schools may more often admit children later rather than earlier because of staffing difficulties. If this were the case then a simple comparison between 'early starters' and 'late starters' might produce a spurious result since the latter group might contain a larger proportion of children from lower socio-economic groups. Thus, any difference between the two groups might be associated with socio-economic factors rather than the length of schooling received. In this section, then, the 'early starters' and the 'late starters' are compared for reading ability as measured by the Southgate Reading Test; for arithmetical ability, as measured by the Problem Arithmetic Test; and for social-adjustment, as assessed by the total score on the Bristol Social-Adjustment Guide. Children included are those in the sample in maintained infant, junior with infants or all-age schools, or in independent schools catering wholly or mainly for children who are not handicapped. Children excluded are those who had attended nursery schools and those for whom all the appropriate information was not available when the data were processed (e.g. age of starting school, occupational group of the father). (b) Length of schooling and reading attainment (i) Analysis ignoring socio-economic factors The most straightforward way of examining the relationship between 'early' and 'late' starting in school and reading attainment in the present sample is simply to compare the reading performance of the two groups when they were tested near the end of their infant schooling. Thus, the children were divided into three categories on the basis of their scores on the Southgate Reading Test: the 'good readers' with a score of 28 to 30; the 'medium readers' with a score from 16 to 27; and the 'poor readers'. The results are given in Table 61. Table 61 Length of schooling and Southgate reading test score It will be seen that the 'early starters' were better readers than the 'late starters' when this criterion of reading performance was used. The difference between the groups in reading ability is highly significant. Thus, one may conclude that children of the same age who make an earlier start with their schooling (aged 4 years 6 months to 4 years 11 months) are better readers, as measured by a test of word recognition, near the end of their infant schooling than those who make a later start (aged 5 years to 5 years 6 months). In the present sample the difference in the length of schooling between the two groups compared was, for the majority of the children, only one term. (ii) Analysis allowing for socio-economic factors Although the overall relationship between length of schooling and reading ability at seven years of age has been demonstrated, this relationship may not be the same for children from all socio-economic backgrounds. In Table 62 the 'early starters' and 'late starters' are grouped according to the occupational group of their fathers. Further, the particular statistical analysis used necessitated dividing the children into two groups according to their reading ability. It was decided that it would be most appropriate to compare the 'good readers' with the 'medium and poor' readers. Table 62 Occupational group of the father, length of schooling and Southgate reading test score It will be seen that in each occupational group the 'early starters' had a higher proportion than the 'late starters' of 'good readers' and a smaller proportion of children of 'medium and poor' reading ability. The evidence from the two statistical tests carried out (see Section IV: 'Statistical Analysis ') is that there is a highly significant difference between the 'early starters' and the 'late starters', independent of occupational group. Thus, the conclusion is warranted that children of the same age who start school 'early' (aged 4 years to 4 years 11 months) are better readers near the end of their infant schooling - irrespective of the occupational groups of their fathers - than children who start school 'late' (aged 5 years to 5 years 5 months). (c) Length of schooling and arithmetic attainment (i) Analysis ignoring socio-economic factors Once again the most straightforward analysis is presented first. The 'early starters' and 'late starters' were divided into three categories according to their score on the Problem Arithmetic Test. Those with a score of 8 to 10 were defined as having 'good arithmetical ability'; those scoring from 3 to 7 had 'medium arithmetical ability'; and the children scoring 2 or below were defined as having 'poor arithmetical ability'. The percentaged results are given in Table 63. Table 63 Length of schooling and problem arithmetic test score The difference between the 'early starters' and 'late starters' in arithmetical ability, as measured by this test, is highly significant. Thus, if no account is taken of socio-economic factors, the conclusion is warranted that children of the same age who start school 'early' (aged 4 years 6 months to 4 years 11 months) achieve better scores on a problem arithmetic test near the end of their infant schooling than children who start school 'late' (aged 5 years to 5 years 5 months). (ii) Analysis allowing for socio-economic factors As with reading ability, a more detailed analysis was carried out which allowed for socio-economic factors in so far as these are assessed by the occupational groups of the fathers. Table 64 Occupational group of the father, length of schooling and problem arithmetic test score For this statistical analysis it was necessary to place the children into two groups on the basis of their scores on the Problem Arithmetic Test. It was decided, as with the corresponding analysis of reading ability, to place the children with 'good arithmetic ability' (with a score of 8 to 10) in one group; and those with 'medium and poor arithmetic ability' (with a score below 8) in the other. The percentage figures are given in Table 64. There was a higher proportion of 'early starters' than 'late starters' in each occupational group with 'good arithmetical ability' and a smaller proportion with 'medium and poor' arithmetical ability. The evidence from the two statistical tests carried out (see Section IV: 'Statistical Analysis ') is that there is a highly significant difference between the 'early starters' and the 'late starters', independent of occupational group. The conclusion is warranted that, in terms of performance on a test of problem arithmetic, children of the same age who start school 'early' are as a group more advanced near the end of their infant schooling than children who start school 'late', irrespective of the occupational group of their fathers. (d) Length of schooling and adjustment in school (i) Analysis ignoring socio-economic factors As in the previous two sub-sections, the simplest analysis is presented first; 'early starters' and 'late starters' are compared in terms of their adjustment in school without regard for any factors except for the length of their schooling. The two groups were divided into three categories on the basis of total scores on the Bristol Social-Adjustment Guides. Those with a score from 0 to 9 were called 'stable'; those with a score between 10 and 19 were operationally defined as 'unsettled'; and those with a score of 20 or more were called 'maladjusted' . The results in percentage form are given in Table 65. It will be noted that the total number of children involved in this analysis is some 300 less than in the corresponding analyses for reading and arithmetic. This is because fewer Bristol Social-Adjustment Guides were available for data processing when this was carried out. Table 65 Length of schooling and adjustment in school As the above table shows, there was a higher proportion of 'stable' children amongst the 'early starters' and a smaller proportion of 'unsettled' and 'maladjusted' children. The difference between the 'early starters' and the 'late starters' is statistically significant, in percentage form. There is thus evidence that in terms of the criteria adopted, children of the same age who start school 'early' are better adjusted in school near the end of their infant schooling than children who start school 'late'. (ii) Analysis allowing for socio-economic factors In Table 66 results are given, separately, for 'early starters' and 'late starters' in the five occupational groups. The difference between the adjustment in school of 'early starters' and 'late starters' was analysed independent of the occupational group of their fathers. In order to carry out this particular statistical analysis, 'unsettled' and 'maladjusted' children were grouped together to contrast with 'stable' children. Table 66 Occupation of the father, length of schooling and adjustment in school It will be seen that in each occupational group there was a higher proportion of 'stable' children amongst the 'early starters' and a smaller proportion of 'unsettled and maladjusted' children. The evidence from the two statistical tests (see Section IV: 'Statistical Analysis ') is that there is a highly significant difference between the adjustment in school of the 'early starters' when compared on this criterion with that of the 'late starters'. One may conclude that children of the same age who start school 'early' are better adjusted in school near the end of their infant schooling than those who start school 'late', irrespective of the occupational group of their fathers. (e) Summary The highly significant results obtained in the three previous sub-sections only permit one, strictly, to make inferences in terms of the assessments used. Nevertheless, the conclusion is warranted that children of the same age who commence full-time infant schooling before the age of five are, as they approach the transfer to junior schools or classes some two years later, more advanced educationally and better adjusted in school than those who commence school after the age of five, irrespective of the socio-economic status of their families. It must be borne in mind that this relationship is not necessarily a causal one; one cannot conclude without further evidence that the relatively poorer performance of later starters is the direct effect of less time spent in school. It may be that both the age of commencing school and educational attainment at seven years of age are related to another factor or factors. For example, if it were true that earlier starting in school were more common in the south of England than in the north and also that educational standards were higher in the south, this may account wholly, or partly, for the demonstrated relationship. This regional factor, and others, will be investigated at a later stage. It may be thought that even if a causal relationship could be demonstrated, the topic is of academic interest and has few practical implications since in a given area it is normal for children of the same age to start school at the same time. Further, any effect of length of schooling might be assumed to diminish and even to disappear as children grow older. Nevertheless, despite the fact that children of the same age in a particular area generally start school at the same time, some borderline has to be drawn and children whose ages differ by only a few days or weeks have different lengths of time in school. Furthermore, since there is evidence (Pidgeon, 1965) that younger children in an age group are at a disadvantage, educationally, compared with older ones, it is important to investigate the factors which contribute to this situation. If the length of schooling is one of these factors, then it may be possible to take practical steps to ameliorate the position. Finally, although it appears to be a reasonable assumption that any effects of length of schooling may diminish with time, the assumption would need to be verified. It may not be true for all children under all circumstances. For example, the importance of the children's level of intelligence may be an important factor, as may the size of organisation of the junior schools or departments which they transfer. References PIDGEON DA Date of Birth and Scholastic Performance. Educational Research, Vol. 8, No.1 (National Foundation for Educational Research) (1965).
C. ENVIRONMENTAL FACTORS 1. Parental situation (a) Introduction Research evidence supports the thesis that the general progress of children who are not living with both their own natural parents is likely to be less satisfactory than that of children who are. The greater part of research in this country is based on findings from children who have shown difficulties which have required special action. The present Study provides an opportunity to compare children living in 'atypical' family circumstances, irrespective of whether or not they are receiving any specialised help, with a large group of children of the same age and otherwise similar background. For the sake of brevity the group of children living with their own parents will be described as 'normal' and those not living with both natural parents will be described as 'atypical'. It might be expected that a comparison of the children from 'normal' family settings with those from 'atypical' ones would show certain differences in growth and development. Children not living with both natural parents are more likely to have been subject to adverse experiences which may play an important part in their emotional development or their general progress. Both the total set of circumstances that lead to children living in an 'atypical' parental situation and the children's growth and development within the 'atypical' setting are likely to be associated with socio-economic status. Comparisons were made, therefore, between children from 'normal' and 'atypical' home backgrounds for reading ability as a major indication of progress at school. The comparison between the two groups was made for boys and girls and within occupational group. The proportions of the children living in 'normal' and 'atypical' parental situations within each occupational group are given in part (b) of this section. The 'normal' group included those children reported by their mothers to be living with or cared for by both natural parents. The 'atypical' group covered children reported as having only one natural parent because of illegitimacy, desertion, divorce, separation or death; those with one step-parent; as well as children who were adopted, fostered or in care. Children whose parental situation was not known have been excluded from this second group, although it is likely that a number of them will eventually be found to belong to it. An analysis of the different types of parental situation within the 'atypical' group in relation to other factors will be possible at a future stage. Information about the children's families was gathered on the Parental Questionnaires. In the totals from the punched card analysis (see Section V: 'Environmental Factors') six per cent of the children for whom there was information were living in an 'atypical' situation. This is likely to be an underestimate of the proportion in the total cohort because mothers of children with no male head of household are more likely to be in full-time work, so that in consequence their completed Parental Questionnaires would have a greater chance of being among the 'late' returns not included in the present sample. In this section the totals were all derived from computer analysis and will vary according to the number of children within the two groups for whom there was appropriate information. All the analyses were of children both in 'ordinary' maintained and independent schools and of children in 'special' schools. The 82 children whose schools could not be classified and the very few who were resident in hospital were not included. The two groups of children were compared wherever possible for three variables: sex; occupational groups of the fathers; and Southgate Reading Test Scores. Two-way tables showing the position in the sample as a whole are given before tables showing three way breakdowns. (b) Parental situation and occupational groups About a third of the children living in an 'atypical' family situation were those without a father or a male head of household and so could not be included in the results presented in Table 67. (It is realised that this group of children in households without fathers may have some special characteristics.) Table 67 Occupational group of fathers by parental situation The proportion of children living in an 'atypical' family situation showed a highly significant and consistent increase from occupational group 1 through to occupational group 5. Of the children in occupational group 1, 2.7 per cent were not living with their own parents, compared with 6.8 per cent of children in occupational group 5. (c) Parental situation and Southgate Reading Test score The association between Reading Score and Parental Situation was tested for boys and girls separately. The children's scores on the Southgate Reading Test (for detailed description see Section III: B.3) were classified into three groups: 'Poor readers' with a score from 0 to 15; 'Medium readers' with a score from 16 to 27; and 'Good readers' with a score from 28 to 30. Table 68 Parental situation and Southgate reading test for boys There was a highly significant association between the two parental situation groups and the three reading groups. More boys in the 'normal' group had high reading scores (35.4 per cent) than those in the 'atypical' group (23 per cent) and, conversely, among the latter the proportion of 'poor readers' was higher; 29 per cent compared with 20.7 per cent. Table 69 Parental situation and Southgate reading test for girls For girls, the same overall pattern was found; there is a significant association with reading in the two groups with a higher proportion of girls who were 'good readers' (47.1 per cent) in the 'normal' group than in the' atypical' group (34.1 per cent). (d) Parental situation, occupational group and Southgate Reading Test score Three occupational groups were formed for this analysis, as at this stage in the Study the numbers of children in the 'atypical' group were too small for a five-fold classification. The two non-manual occupational groups (1 and 2) formed the first new grouping and the partly skilled and unskilled workers in occupational groups 4 and 5 formed the second new grouping. Table 70 Occupational group, parental situation and Southgate test score (boys and girls combined) For occupational groups 1 and 2 combined, as well as for occupational group 3, the difference in the children's reading scores in the two parental situations was highly significant or significant; children in the 'normal' group had consistently higher reading scores than those in the 'atypical' group. However, in occupational groups 4 and 5 this was not the case, there being no significant difference in the reading scores of the 'atypical' and 'normal' groups. Two possible interpretations of this result are, first, that an 'atypical' parental situation may be only one of a number of potentially adverse environmental factors which occur relatively more often in occupational groups 4 and 5 and so is less discernible as a factor in isolation in relation to the children's reading progress. Secondly, it is possible that the patterns of family and neighbourhood life in these two occupational groups mitigate some of the possible difficulties of children in an 'atypical' situation. (e) Summary (i) There was evidence from the present Study of an association between parental situation and socio-economic status; there being a consistent tendency for the number of 'atypical' parental situations to increase from occupational group 1 through to occupational group 5. (ii) An association was demonstrated between poor reading ability and 'atypical' parental situation which was highly significant for boys and significant for girls. (iii) There was a highly significant association between reading ability and parental situation for occupational groups 1 and 2 combined and a significant association with occupational group 3; the 'normal' group showing better reading ability. However, there was no such association within occupational groups 4 and 5 combined. 2. Parental approach (a) Introduction Parental attitudes to their children's education are related to children's progress in school (for example, Fraser, 1959; Douglas, 1964). Douglas gives evidence that this relationship is highly correlated with his index of socio-economic status; further, the relationship still holds good within the different social classes, irrespective 'of standards of living, size of family and academic record of the school'. The children in the present Study are somewhat younger than those in the majority of studies on this subject. Only one criterion of parental interest could be used at this stage of the Study. The objective fact, as reported by head teachers, of whether or not parents had initiated discussion about their children was preferred to the more subjective assessment of the parents' interest made by the teachers. Initiating discussion with their children's teachers is only one indication of parents' concern, and there will be some parents who find it more difficult to visit the school than others; for example, mothers with other children who are very young. The over-anxious or over-concerned mother who tends to make too frequent enquiries cannot be differentiated under this criterion either. Head teachers were asked: 'Since September 1964 have the parents taken the initiative to discuss the child, even briefly, with you or any member of your teaching staff?' For the sake of brevity, children whose parents had initiated discussion will be described as the 'Approached' group and those whose parents had not initiated discussion will be described as the 'Not Approached' group. The replies analysed in the tables in this section concern children both in 'ordinary' maintained and independent schools and in 'special schools'. The children whose schools could not be classified or who were resident in hospitals were not included. The association between parental approach and the children's progress has been tested in relation to one major factor in educational progress, namely reading ability. The reading ability of children in the two groups ('Approached'/'Not Approached') was considered in relation to sex and occupational group. (b) Parental approach and occupational group The results in Table 72 show that about 70 per cent of parents in occupational group 1 initiated discussion with teachers, compared with some 46 per cent in occupational group 5. In all occupational groups, except occupational group 5, a higher proportion of parents had initiated discussion than had not. Table 71 Occupational group of father and parental approach (boys and girls combined) The results of the statistical test (trend) showed that there is a highly significant tendency for the proportion of parents who have approached the school to decrease from occupational group 1 through to occupational group 5. However, the statistical test for departure from a linear relationship showed that this overall tendency is not consistent through all the paternal occupational groups and, thus, that the decrease from one occupational group to another in the proportion of parents who have approached the school is not uniform. An inspection of the table indicates that this departure from the overall tendency for 'parental approach' to fall through the occupational groups is because the proportions in the 'Approached' and 'Not Approached' groups in occupational group 3 and occupational group 4 are virtually the same. (c) Parental approach and Southgate Reading Test score The children included in the following tables were those with Parental Questionnaires returned completed in time for analysis for this Report. The association between reading score and Parental Approach was tested for the total sub-sample and then for boys and girls separately. The children's scores on the Southgate Reading Test (for detailed description see Section III B.3) were classified into three groups: 'Poor readers' - scoring 0 to 15 on this test; 'Medium readers' - scoring 16 to 27; and 'Good readers' scoring 28 to 30. Table 72 Parental approach and Southgate reading test score (boys and girls combined) There were significantly more 'good readers' and fewer 'poor readers' and 'medium readers' in the 'Approached' group than in the 'Not Approached' group for all the children in this sub-sample. Table 73 Parental approach and Southgate reading test score (boys only) Table 74 Parental approach and Southgate reading test score (girls only) The results in Tables 73 and 74 show that both for boys and for girls there is a highly significant association between reading ability and parental interest, as assessed by these criteria. In the 'Approached' group about 40 per cent of boys and 51.4 per cent of girls were 'good readers', compared with 26.6 per cent of boys and 40 per cent of girls who are 'good readers' in the 'Not Approached' group. (d) Parental approach and Southgate Reading Test score within occupational groups Table 75 Occupational groups, parental approach and Southgate reading test score (boys only) As can be seen in Table 75, for the boys in each occupational group the proportion of 'good readers' in the 'Approached' group was higher than the proportion of 'good readers' in the 'Not Approached' group. Conversely, the proportion of 'poor readers' is lower in the 'Approached' group than in the 'Not Approached' group in each occupational group. For example, in occupational group 1, about 52 per cent in the 'Approached' group were 'good readers', compared with 36 per cent in the 'Not Approached' group; and only eight per cent were 'poor readers' in the 'Approached' group, compared with 21 per cent in the 'Not Approached' group. For each occupational group a statistical test was carried out to determine whether the difference between the reading performance of the two groups was significant. The difference was highly significant for occupational groups 1, 3 and 4. Thus, in these three occupational groups, and in terms of the criteria adopted, children whose parents initiated discussion with teachers are better readers than those whose parents did not. In occupational groups 2 and 5 the results of the statistical testing were not significant, although the differences were in the same direction as in the other three occupational groups. These results may be due to the relatively smaller number of boys in occupational groups 2 and 5. Further interpretation of these results must await more detailed statistical treatment. Inspection of Table 75 also suggests a difference in 'parental approach' between the parents of 'poor readers' in the manual and non-manual occupational groups. In occupational groups 1 and 2 more (numerically) of the parents whose children were 'poor readers' were in the 'Approached' group than in the 'Not Approached' group. In occupational groups 3, 4 and 5, however, the reverse was the case, with fewer parents of 'poor readers' in the 'Approached' group than in the 'Not Approached' group. Table 76 Occupational groups, parental approach and Southgate reading test score (girls only) As will be seen in Table 76, the overall pattern of results was similar to that of the boys. In each occupational group a higher proportion of girls in the 'Approached' group were 'good readers', compared with the proportion of 'good readers' in the 'Not Approached' group. There was a lower proportion of 'poor readers' in four of the occupational groups among the children whose parents had approached the school than among the children of parents who had not initiated discussion. Statistical evidence showed the differences between the two groups, 'Approached' and 'Not Approached', to be highly significant in occupational group 3 and to be significant in occupational group 1. The difference between the two groups approached the level of statistical significance in occupational group 4. As with the boys, the differences in reading ability were not significant between the two groups of 'parental approach' in occupational groups 2 and 5. The results presented in Tables 75 and 76 above indicate that the criterion of 'parental approach' as an index of parental interest is associated with reading ability and that this association both for boys and for girls is to some extent independent of socio-economic factors. However, the association justifies a more elaborate statistical approach, which should also take account of the other data on parental interest which are available. (e) Summary (i) There is a highly significant tendency for the proportion of parents who have approached the school to decrease from occupational group 1 through to occupational group 5, although this trend is not wholly consistent. (ii) The differences between the 'Approached' group and the' Not Approached' group are highly significant in relation to reading scores for the boys and for the girls. Boys and girls in the 'Approached' group have better reading ability than in the 'Not Approached' group. (iii) The differences tended to be in a similar direction within occupational groups for boys and girls in relation to reading scores. The differences between the 'Approached' group and the 'Not Approached' group were significant in relation to reading scores for boys and girls separately in occupational groups 1 and 3 and for boys in occupational group 4. References DOUGLAS JWB The Home and the School (McGibbon and Kee) (1964). FRASER ED Home Environment and the School (University of London Press) (1959).
D. BEHAVIOUR AND ADJUSTMENT Adjustment in school in relation to occupational group The relationship between children's emotional and social adjustment and other factors is one of the most important yet complex aspects of child development. It is important because of its practical implications, not only for children's progress in school and general happiness in childhood, but also for their later integration into adult society, for their careers and their own adequacy as parents; it is complex not only because there are many factors involved which interact with each other, but also because most of these factors are difficult to 'measure' in a meaningful way. In this section the children's adjustment in school, as assessed by the total score on the Bristol Social-Adjustment Guides, is related to the occupational group of their fathers. As has been explained elsewhere in this Report, the occupational group is seen as a crude index of the socio-economic status of the family. This variable might seem to be less relevant to a consideration of children's adjustment than, say, to their reading attainment. Certainly, the possibility of a close relationship with attainment in school seems more predictable. The general level of verbal skills, the intellectual stimulation and professional and leisure interests of parents in non-manual occupations contrast sharply with the general pattern to be found in many homes of unskilled manual workers. But what of emotional and social adjustment? It seems likely, as was indicated in the opening paragraph, that any relationship between adjustment and socio-economic factors will be more complex. A number of studies have shown that patterns of child rearing differ between 'social classes'. It is unlikely, however, that any comparison between children from different socio-economic backgrounds will merely reflect this. It is known, for example, that many of the circumstances most likely to have an adverse effect upon children's adjustment occur more frequently in homes of lower socio-economic status. However, in any study of complex variables, the most straightforward relationships have first to be examined so that allowance can be made for these in subsequent analyses. In view of the marked difference between the sexes in adjustment, as assessed by the total score on the Social-Adjustment Guides (see Section VG), it was decided to study the association between adjustment and occupational group separately for boys and girls. The children were divided for the purpose of this analysis into two groups on the basis of their scores on the Guides: those with a score of 0 to 9 ('stable') and those with a score of 10 or more ('unsettled and maladjusted'). In Tables 77 and 78 the results are given in percentage form for all children in the present sample in any category of school for whom all the appropriate information was available. Table 77 Social adjustment and occupational group of the father for boys Table 78 Social adjustment and occupational group of the father for girls It will be seen in both tables and in Figure 8 that the proportion of 'stable' children shows a tendency to fall from occupational group 1 to occupational group 5. In both cases the statistical tests are highly significant, indicating that for both boys and girls of this age there is an overall tendency for adjustment in school to be poorer in children from homes of lower socio-economic status. Figure 8 Social adjustment and occupational group of the father (N = 6,467) However, there was an interesting difference between the results for the two sexes. In the case of boys of this age there is strong evidence for a decrease in 'stable' behaviour the lower the status of the paternal occupational group and no evidence that this 'decline' is not consistent through the occupational groups. For girls of this age there is the same evidence of a decrease in 'stable' behaviour through the occupational groups, but there is also evidence that the 'drop' from one occupational group to another is not uniform. It will be seen in Table 78 and Figure 8 that the proportion of 'stable' girls in occupational group 2 in this particular sub-sample was higher than that in occupational group 1. In occupational group 3 the proportion dropped considerably, but there was virtually no difference between the picture there and in occupational group 4. There was then another drop to occupational group 5. In the absence of further statistical analysis no conclusions can be drawn about the patterns of results in this sub-sample. The fact that there were more 'stable' girls in occupational group 2 than occupational group 1 is unexpected and may be due to chance fluctuations. What does emerge from the analyses of the total scores from the Bristol Social-Adjustment Guides is, first, that girls at the age of seven are markedly better adjusted in school than boys (see Section VG). Secondly, for both sexes there is a strong tendency for the least well-adjusted children to be in lower socio-economic groups. Thirdly, it appears that for boys the relationship between socio-economic factors and adjustment in school is consistent, whereas for girls the relationship is not uniform and appears to be more complex. It is highly likely that socio-economic factors, as assessed by the occupational group of the father, are related to children's adjustment in school because these factors are themselves associated with other circumstances in the social and physical environment of the family and neighbourhood; and also with the physical and mental health of the parents. A more detailed analysis of some of these circumstances in relation to children's adjustment is planned.
VII: FIRST FINDINGS: SUMMARY AND DISCUSSION A. EDUCATIONAL, BEHAVIOURAL AND SOCIAL ASPECTS 1. Introduction Undoubtedly there were many disadvantages in having to produce an early report, not least of which lay in the need to hedge conclusions around with repeated reminders of the preliminary and incomplete nature of both the sample and the analysis. However, there accrued at least one advantage. In the past, the delay between the completion of studies, especially large scale ones, and their findings being published has tended to be very long. This enables practitioners and administrators to assert that events may have in the meantime brought about such changes that the need for action had been lessened if not eliminated; or that a new investigation would have to be mounted to take account of changed conditions - thus postponing action. In the case of the National Child Development Study, the interim findings presented in this Report have become available 18 months from the beginning of the project and about 12 months after the data on the children began to be collected. Thus, the information is so recent in origin that even in a period of comparatively rapid social and educational change the findings describe the situation as it actually is at present. In recent years there has been increasing interest in the complexity of all the influences which affect children's development, and research techniques have grown more subtle, more comprehensive and more rigorous. 'The time has certainly come for some cooperative scheme of research into the innumerable questions that arise. A single investigator can do little, except make a few limited and tentative experiments and sketch what appears to be the most valid methods of inquiry. The teacher, the medical officer, the social worker, the psychologist, the statistician - all need to lend their expert knowledge'. These prophetic words are from Burt's (1937) classic survey which has never been paralleled, in this country or anywhere else. Yet 27 years later, Wiseman (1964), discussing the relationship between environment and educational progress, has to state that 'research is still largely a matter of preliminary survey and exploration, seeking new insights to form the bases of more productive hypotheses. We know very little about the mechanisms underlying the variations of a multitude of environmental factors: with no firm grasp of these, too many research workers attack a small and ill-defined sector of the field, armed with little but a hunch or a prejudice and using whatever variables may come conveniently to hand'. The National Child Development Study has taken account of both Burt's and Wiseman's strictures: conceived as an inter-disciplinary project, it is sponsored jointly by four bodies; financed by a number of government departments; it depends on the cooperation, indeed active collaboration, of all local authorities in England, Scotland and Wales, which was granted to a most generous extent; and the composition of the research team itself was also multi-disciplinary. Moreover, hypotheses were formulated before any of the material was available for analysis. The Study has also largely overcome another common difficulty, that of sampling. It is well known that there are strong regional differences because various parts of the country differ widely economically, culturally and socially (Floud, Halsey and Martin, 1957; Derrick, 1961); and one study (Ferrez, 1961) suggests that - at least in France - 'geographical factors are even more important than the social ones'. Fortunately, sufficient financial support was given to this Study to make it possible to follow up the entire cohort. It could, of course, be argued that there may still be some sampling bias since all the children were born in one week in March; they could conceivably have certain characteristics which differentiate them from children born in anyone or all of the remaining 51 weeks of 1958. Whether and to what extent this is the case will have to await the time when the formidable financial and organisational difficulties involved in mounting more than one national perinatal mortality survey in anyone year have been overcome. The fact that inter-disciplinary research is still largely in an early, almost preliminary, phase means that one of the main productive results of large-scale studies such as ours is the posing of questions to form the stimuli for further investigations. In particular, there are two kinds of study which should follow from it: first, surveys comparable in scope and methodology should be mounted every 10 years or so; this would make possible the observation on a national scale of changes in child development in relation to changes in the economic, social, medical and educational sphere. Secondly, such large-scale studies should prepare the way for smaller but much more detailed and intensive inquiries into individual differences in the development of children's personality, mental and physical growth, health and educational progress. In fact, plans for both these types of investigation have already been prepared (together with financial estimates), and it must now be hoped that the necessary funds will become available. 2. Overview This interim report has two aims: the first and main aim was to give as detailed a picture as possible of a large national sample of seven year old boys and girls. This is presented in Section IV: 'Descriptive Statistics'. Of course, data of this kind are normative and no value judgements are made as to what is desirable or undesirable, but attention is focused on what is taking place. When the material for the total sample has been analysed, the picture will then become more reliable and meaningful. The second aim was to make at least a beginning with the much more complex task of unravelling the influence of and relationship between a wide range of factors (including pre- and post-natal conditions) and children's physical and psychological development, educational progress, adjustment and behaviour at home and school. What, then, are the more salient findings thus far? (a) Settling at school It would seem that the majority of children settle down within the first month of starting school. However, a sizeable proportion (some 25 per cent) remain unsettled up to three months or longer; there is a very significant difference between the sexes, boys taking longer to settle down than girls. The results indicate that the schools of about one third of the children use some form of introductory attendance prior to the commencement of full-time schooling. Our data do not readily lend themselves to an exploration of the extent to which such schemes have any effect in shortening the subsequent settling down process; but at least theoretically one would expect this to be the case. (b) Parental interest The parents' interest in their children's education and their contact with the school was assessed in three ways: first, teachers' ratings of parental interest; these indicated that parents showing little or no interest are in a minority (some 16 per cent), there being no difference between mothers and fathers in this respect. Secondly, teachers were asked whether during the current school year parents had taken the initiative in discussing their child with a member of the staff; a rather high proportion had not done so (43 per cent); when this question was explored in relation to fathers' occupational group, it was found that the higher the socio-economic status the greater the proportion of parents who had spontaneously sought an opportunity to discuss their child with a member of staff. Thirdly, asked whether they would wish their child to remain at school beyond the minimum school leaving age, the great majority of mothers replied in the affirmative (over 80 per cent). Of course, these three criteria are not only rather broad and crude but also involve different degrees of subjectivity. But it looks as if the parents of about half the sample showed an active interest and involvement in their child's schooling, while a minority appear to be lacking such interest. A first exploration of the relationship between one of the indices of parental interest in the child's education and tested attainment in reading showed there to be a significant association: the proportion of good readers was higher among those children whose parents had themselves initiated some contact with the school, and this was true also for boys and girls separately. Then the relationship was examined between parental interest and reading attainment within each of the five occupational groups separately; the same association was found within occupational groups for boys in occupational groups 1, 3 and 4 and for girls in occupational groups 1 and 3. (c) Educational attainment Even at the early age of seven years, girls were found to be significantly better at reading than boys; this held true whether the yardstick was an objective test or the level of the reading book the child was able to manage or the teacher's judgement of the child's reading ability. When this was examined in relation to socio-economic level - assessed in terms of the occupational group of the children's fathers - there was a highly significant difference in the direction predicted on the evidence of previous studies: the lower the occupational status of the fathers, the poorer the reading attainment of the children. Though the children in this Study are all the same age, they have experienced varying length of schooling because administrative arrangements for starting school differ in different parts of the country. This provided an opportunity to compare those who had been admitted to infant schools before the age of five ('early starters') with those whose attendance commenced after their fifth birthday ('late starters'). Looking at the reading attainment of these two groups, a significant difference in favour of the 'early starters' was found, even though for the majority the difference in length of schooling had only been one term. This difference between 'early' and 'late' starters was found to be independent of occupational group. One other environmental variable was explored in relation to reading attainment; namely, whether the child lived with both his natural parents or not, the former being referred to as the 'normal' and the latter as the 'atypical' family situation. The 'atypical' situation included not only 'one-parent' families but also children who had one step-parent or who were adopted, fostered or in residential care. It was found that reading attainment was significantly lower for those whose family situation was 'atypical', and this was equally true for boys and girls. When the same question was examined in relation to socio-economic status, a rather more complex pattern emerged: for the higher occupational groups (1, 2 and 3) the result was the same, namely, reading attainment being higher for children living with both their natural parents; but in occupational groups 4 and 5, the proportion of poor readers did not differ whether the children had a 'normal' or 'atypical' family background. Now to turn to attainment in arithmetic. Here, boys were found to be superior to girls on a test of problem arithmetic. Again, there was a relationship between attainment in this subject and paternal occupational group; the lower the latter, the lower the children's score on the test. Length of schooling was also significantly associated; higher arithmetic scores were attained by 'early' than 'late' starters. When fathers' socio-economic status was taken into account, the same relationship between starting school 'early' and good arithmetic attainment was found to exist independent of occupational group. The few researches that have considered the differences between reading and arithmetic in their response to environmental effects have produced somewhat conflicting results; some showed that reading is more prone to this than is arithmetic (Thorndike, 1951; Burt, 1955; Davis and Kent, 1955; Lynn, 1958), while others do not find this to be the case (Wiseman, 1952; Kemp, 1955). At this stage there has been no analysis of possible differential effects; the present results do, however, indicate that at the age of seven both reading and arithmetic are related to environmental aspects. Further, an 'early' start in the infant school is associated with higher attainment in both subjects some two years later, irrespective of parental socio-economic status. Three other, more general abilities were explored by means of teachers' ratings. For both 'oral ability' and 'creativity' girls were rated higher by their teachers than boys, while the position was reversed regarding 'awareness of the world around'. At this stage time did not permit any more detailed examination of these abilities or their relationship to other variables. (d) Behaviour and adjustment These were assessed in two ways: by asking mothers about the behaviour and developmental difficulties of their children; and by obtaining information from teachers by means of the Bristol Social-Adjustment Guides, which were completed for each child. So far, only some preliminary analysis of this material has been possible. For most aspects of behaviour, there was evidence that a greater proportion of boys, as reported by their mothers, show difficulties at home. However, there are some exceptions; for example, more girls than boys suck their thumb or fingers during the day and bite their nails. The differences between the sexes appeared to be most marked for aggressive types of behaviour. It is worth noting that a sizeable proportion of seven year olds are reported to be faddy over food; and to throw the occasional temper tantrum (a little under 30 per cent in each case). In these two aspects of behaviour there were no significant differences between boys and girls. However, more boys are reluctant to go to school than girls. Deviations from normal behaviour which many clinicians would consider as likely indications of some degree of emotional disturbance occurred only among a small minority. Thus, between one and eight per cent of children were reported to show the following behaviour frequently: difficulty in settling to anything for more than a few moments; destroying their belongings or those of others; frequently disobeying parents; being upset by new situations; and bodily twitches or mannerisms. In this context it is interesting to recall a recent study by Glueck (1966) dealing with the 'Identification of potential delinquents at 2-3 years of age'. The three behaviour traits which, when manifested at an early age, markedly distinguished later delinquents from non-delinquents were 'extreme restlessness, destructiveness and non-submissiveness to parental authority'. Thus, these traits are considered by Glueck to have high predictive value, not only for distinguishing future delinquents but also for diagnosing maladjusted or 'malfunctioning' children. When the children's behaviour in school was assessed by means of the Bristol Social-Adjustment Guides, there was again a highly significant difference between the sexes: the proportion of boys being rated as 'maladjusted' was twice as great as that of the girls; the converse was also true, namely, significantly more girls were being rated as 'stable' than boys. At a later stage qualitative differences in the type of deviant behaviour shown by boys and girls respectively will be explored. However, using the criterion of the Guides, it looks as if about 13 per cent of seven year olds show behaviour indicative of 'maladjustment'. When adjustment was examined in relation to socio-economic status, it was found that the proportion of 'stable' children decreased the lower the occupational group of the fathers. Though this was the case for both sexes, the 'decline' from one occupational group to the next was uniform for boys but not for girls. 'Social-adjustment' was then examined in relation to length of schooling. A significant difference was found between 'early' and 'late' starters, the latter showing more indications of poor adjustment than the former. This relationship between 'early starting' and better adjustment was also found to hold independent of occupational group. (e) Need for special provision This term is used here in the widest sense to include facilities within and outside the ordinary school, as well as provision for all kinds of difficulties, be they educational, emotional or physical. First, to consider backwardness in reading. There is some evidence from our results that there has been a decrease in the proportion of poor readers during the past ten years (Morris, 1959). Nevertheless, a considerable number of children in this sample of seven year olds were described by their teachers as being either non-readers (about three per cent) or poor readers (about 24 per cent). An even larger proportion (almost half the sample) had not achieved a sufficient mastery of this subject near the end of their infant schooling to use it as an effective tool for further learning; rather, reading must continue to be specifically taught, because full mastery of the skill has not yet been attained. Thus, on the evidence of the children's present reading abilities about a quarter will need a continuation of 'infant methods' if they are to progress with this basic subject; moreover, a proportion of them will probably be unable to succeed unless given general educational help of one kind or another, not merely help with reading. To ascertain the likely size of this group, teachers were asked two questions: how many of the children were at present receiving special help because of educational or mental backwardness and how many would benefit from such help if it were available. The answer to the first question was five per cent and to the second question eight per cent; in each case the proportion of boys being significantly greater than that of the girls. Thus, in the teachers' judgement some 13 per cent of the children would derive benefit from educational help additional to what could be provided by class teachers themselves. With regard to special educational help, including special schooling, teachers were of the opinion that within the next two years this would be a likely requirement for some five per cent; this was in addition to the half per cent already attending special schools. However, for a number of reasons, this figure of five per cent is likely to be an underestimate. One other direct method was used to arrive at some estimate of the need for special provision; this was by finding out from teachers how many children had been referred to outside agencies because of behaviour difficulties or lack of progress. The figure reported, 9.5 per cent, is again likely to be an underestimate: some children will have been referred to an outside agency unbeknown to their teachers; and, perhaps much more important, a lack of diagnostic and treatment facilities has a curtailing effect on the number of referrals. Such a lack exists in many areas with a consequent shortage of places in special schools and long waiting lists for an examination in school psychological services and child guidance clinics, where these exist. Conversely, it is a well-known phenomenon that if a new special school or child guidance clinic is opened, there is a steep rise in the number of referrals. Lastly, there is one indirect way of trying to assess the need for psychological consultation and possibly treatment. The extent of the developmental and behaviour difficulties reported by the mothers and the proportion of children assessed on the Social-Adjustment Guides to be 'maladjusted' at school would indicate that a minimum of five per cent have quite serious adjustment problems, while at least a further 10 per cent show stress symptoms of various kinds. Only further analysis will provide evidence on the degree of overlap between educational and emotional problems as well as on their relation to the whole range of physical handicaps. What seems already beyond doubt is the fact that there is a need for special provision during the second year in the infant school if educational and emotional problems are to be dealt with as soon as they are recognised by teachers and parents. 3. Some pointers to policy and practice Starting school presents for most children a major step forward in independence, but also a major departure from their previous pattern of life. Might not a more widespread adoption by infant schools of the practice of 'introductory attendance' both ease and speed up the process of settling down? It is now widely recognised that parental interest in the child's education plays a vital part in satisfactory progress. Are not infant schools the most natural and logical starting point for fostering such interest on a much wider scale? This would involve giving some priority to them in terms of staff and money. Perhaps increasing responsibility for enlisting parental interest should also be undertaken by school welfare officers and health visitors? Or should teachers who have taken one of the newly established courses for joint social work/teacher training be encouraged to work in infant schools? Could voluntary workers include in their service to the community the task of freeing parents of large families to visit their children's school by looking after those who remain at home? 'Early starters' have higher attainment and better adjustment than children who start school about a term later. Perhaps this advantage is comparatively short-term; only subsequent examinations of the same children will tell. But if the advantages turn out to be long-term, should attempts be made to ensure an earlier start, especially for the culturally and socially underprivileged? Low socio-economic status is associated, even by the age of seven, with low educational attainment and high 'maladjustment'. If equal educational opportunity is to become a reality, ought not pre-school education, specially geared to the needs of culturally deprived children, to be given high priority? The well-known pattern of boys being more backward in reading and showing a higher incidence of behaviour difficulties was found to exist already by the age of seven years. Is there a continuum of vulnerability stretching right back to prenatal and perinatal days? If later work does establish such a link, would boys' greater vulnerability indicate a need for differential child-rearing and educational practices? Or is their relatively inferior performance a result of current child-rearing and educational practices? For example, could it be related to the fact that it is largely women who care for and educate boys during the early years? Would more male teachers in infant schools have a beneficial result on boys' educational progress and adjustment? About 45 per cent of children appear not to reach full mastery of reading skills by the time they are due to leave infant schools. This means that the teaching of reading needs to be continued at the junior level; also there should be some continuity in teaching methods and reading schemes. This clearly has implications for both policy and practice. Is the present age of transfer the most appropriate? Must there be a uniform transfer age? Is there effective practical recognition, both in teacher training and in appointing junior school staff, of the need for infant school reading methods beyond the age of seven? The proportion of educationally backward and emotionally maladjusted children appears to be high enough by the second year in the infant school to warrant greatly increased provision for their needs. Such early provision, including diagnosis and treatment of various kinds, would be in line with the increasing emphasis which is being placed on early detection and prevention. The focus of such early preventive work should be on socially and culturally underprivileged children, especially boys, since by the age of seven their needs are clearly the greatest, at least numerically. Hitherto, a much greater proportion of children from the higher socio-economic groups find their way into child guidance clinics, while a much higher proportion of boys from the lower socio-economic groups eventually appear in juvenile courts. Of course, poverty, broken or disrupted family life, housing difficulties, especially of large families, all contribute to this pattern of backwardness and maladjustment. At the same time, early remedial and psychological treatment has much to offer, particularly if it could be part of a more comprehensive scheme for family-centred, preventive and rehabilitative work. What kind of diagnostic and treatment centres would be most appropriate for dealing with those in greatest need? How can parental cooperation be obtained? And is it essential? What should be the relationship between preventive social, educational and psychological services? Need there be better integration of policies, services and practices? References BURT Sir C The Backward Child (University of London Press) (1937). BURT Sir C 'The evidence for the concept of intelligence'. Brit. J. Educ. Psychol., 25, pp. 158-77 (1955). DAVIS DR and KENT N 'Intellectual development in school children, with special reference to family background'. Proc. Roy. Soc. Med. 48, pp. 993-5 (1955). DOUGLAS JWB The Home and the School (MacGibbon and Kee) (1964). FERREZ J 'Regional inequalities in educational opportunity'. In HALSEY AH (Ed.) Ability and Educational Opportunity, Organisation for Economic Cooperation and Development (1961). FLOUD JE, HALSEY AH and MARTIN FM Social Class and Educational Opportunity (London: Heinemann) (1957). FRASER ED Home Environment and the School (University of London Press) (1959). GLUECK ET 'Identification of potential delinquents at 2-3 years of age'. International J. Soc. Psychiatry, 12, pp. 5-16 (1966). KEMP LCD 'Environment and other characteristics determining attainments in primary schools'. Brit. J. Educ. Psychol., 25, pp. 67-77 (1955). LYNN R 'Disparity of attainment in reading and arithmetic'. Brit. J. Educ. Psychol., 28, pp. 277-80 (1958). MORRIS JM Reading in the Primary School (National Foundation for Educational Research) (1959). THORNDIKE RL 'Community variables as predictors of intelligence and academic achievement'. J. Educ. Psychol., 42, pp. 321-38 (1951). WISEMAN S Education and Environment (Manchester University Press) (1964).
B. PHYSICAL AND MEDICAL ASPECTS 1. Scope and value of the preliminary analysis New methods of surveillance and investigation of school children's health should be constantly evolving. Easier adaptation to changing circumstances occurs where a large national cohort is used to monitor the proportion of children requiring treatment for physical ill-health. Where this reveals previously undetected or untreated cases, action is indicated - either towards prevention through health education or towards earlier diagnosis of major and minor handicaps by more screening tests and increased surveillance from infancy onwards. Follow-up of such a national group of children throughout their childhood can also throw some light upon new ways of establishing an optimum procedure for assessing health and growth. As no interim study of the children had been undertaken since birth, the seven year old follow-up included a detailed developmental and medical history as well as a full physical and medical examination. This must not, of course, be taken to imply that a complete medical history and examination is felt to be ideal, or even practicable, as a routine procedure for British school children at seven years of age. Procedures at this age might include selective screening tests; a questionnaire or personal enquiry of parents and teachers; or a medical examination of vulnerable groups. Emphasis on high-risk children is typical of a growing number of school health services. The present preliminary results also indicate a need for increased medical staff trained in child development; closer contact between parents and educational and school health services; more screening procedures; and added efficiency in early diagnostic procedures to 'identify' major and minor handicaps in pre-school and infant school children. Pre-school records of the present sample show that only in infancy was observation optimal for early diagnosis. Over 80 per cent had attended infant welfare clinics. Between one and five years considerably fewer (38 per cent) had been taken to a pre-school or toddler clinic; an unknown number would have been seen by their general practitioner. No attempt has been made on this incomplete sample to examine the nature of previously undiagnosed disabilities. At a later stage it is hoped to investigate these cases as well as the age at diagnosis of known handicaps and the educational and medical facilities provided. However, it is doubtful whether even a follow-up of more than 15,000 children will yield a sufficient number of major handicaps to evaluate the efficacy of 'treatment' for separate conditions. Interest in the handicapped children will centre mainly on their perinatal correlates. The present Report covers mainly the field of minor disabilities and deviations and the assessment and testing of special senses. In many spheres the results may be seen as an adjunct to the statistics assembled in the reports of local authority school health services and summarised in the biennial reports of the Chief Medical Officer of the Department of Education and Science. The results are more valuable where national figures are otherwise unavailable and also where gross regional or local variations in reported incidences indicate ambiguities of definition or interpretation. The present Report, thought to be regarded as provisional, does demonstrate the feasibility of gathering national data on the health of children through the school health services. By using a standard format and modern data-processing techniques, it has been possible to present interim results in a very short period. The pre-coded answers both for the history obtained from the mother and for the medical examination covered relevant aspects of every system which could readily be reproduced. The medical examinations were carried out without special apparatus or conditions by school medical officers. The information gathered was more comprehensive than would normally be considered necessary or possible at routine examinations, but regular shortened inquiries, pre-coded and rapidly data-processed, would allow decisions to be taken from a consideration of results while still current. The present Study, then, may in part be of value as a pilot for a national system of recording and retrieving data on the health of school children. 2. Descriptive statistics Information was obtained on past development and illnesses. This was retrospective as the children were not followed between birth and seven years. For this Report careful selection was made in view of its potential unreliability. Most of such data are used only in sex comparisons. However the retrospective data were gathered very much as any routine medical history. Data in Appendix 1 (not included in this version) allow the reader to calculate past incidences if he wishes. He may be surprised at the reported frequency of certain conditions of which a history is often considered abnormal. One such condition is nocturnal enuresis; about one in nine of the children were reported as being wet by night more than occasionally between five and seven years. A much smaller proportion (4.4 per cent) had daytime incontinence after three years and only 1.2 per cent were reported to have soiled by day after four years. Much is written currently on the ill-effect of periods of separation due to hospital admission and the present policy is moving towards unrestricted visiting of children in hospital. The wide importance of this subject is shown by the high proportion of children who had been admitted to hospital by seven years of age for various conditions. Any effect on social adjustment will be investigated later. Strikingly high among causes for admission were accidents and other injuries both at home and elsewhere. This underlines the need for better education in accident prevention. Sex differences in Section VE ('Medical and Developmental Sex Comparisons') confirm for many morbidity factors what is already known for perinatal, infant and childhood mortality, namely the greater vulnerability of boys. For example, boys attended more often at child guidance clinics and speech clinics and showed more facial tics, stammers and other speech defects, as well as being more accident-prone and, historically, later in developmental 'milestones', including walking, talking and bladder control. Disorders of special function such as hearing, vision and speech are among the most important educational considerations at the age of seven. Section VC, therefore, summarises the results of investigations into these functions. The extent to which defects of hearing and speech had impinged upon parental consciousness was indicated by a past history of hearing difficulty in 10 per cent and some speech abnormality in 16 per cent. The workload carried by the corresponding medical services was considerable, though it could only be estimated retrospectively. Apart from school visual or auditory tests, 12 per cent of mothers reported that the children had attended for special visual examination and eight per cent at hearing or audiology clinics. The strain placed upon the childhood dental services was underlined by the fact that three quarters of the children were reported as having already come under the school or general dental services by seven years of age. Many would feel, however, that all children should have dental assessment, and the importance of this is emphasised by the fact that one in five were found on examination to have a minimum of eight or more decayed, missing or filled teeth. Minor visual impairment was also reported to be quite frequent. Approximately one in seven of the children were assessed by the medical examiners to have a visual defect, but only 0.3 per cent were judged as handicapped for normal schooling and everyday activities; some six per cent of the sample were found on examination to have a squint (manifest or latent); over five per cent had uncorrected visual acuity of 6/12th or less in the right eye, and a similar proportion in the left eye. And what of treatment? Six per cent of children wore or had already required glasses, 25 per cent of whom had corrected vision of 6/12th or less in one or other eye. Should more children be regarded as potentially visually handicapped and put near the front of the class? About five per cent of seven year olds were judged by the medical examiners to have a minor degree of hearing impairment. Analysis of the clinical hearing test of these children and of their audiograms should throw further light upon this. Speech difficulties clearly gave concern to parents. One in six of the children were reported as having had current or past stammer or stutter (6.2 per cent), or other speech difficulty (10 per cent). Over two per cent had received speech therapy by the age of seven. A stammer or stutter was observed on examination in about one per cent. In a rating of the children's speech intelligibility by the medical examiners, about 14 per cent were judged to be not fully intelligible, but only 1.4 per cent were assessed as having a moderate or severe impairment of speech. At the present stage, time has permitted the analysis and presentation of only a part of the medical information which has been gathered. In addition, some results have been withheld until data on the 'late returns' can be processed because of particular danger of bias or misleading conclusions. In a later report it is intended not only to fill these gaps but also to examine the correlations between current medical, physical, educational, psychological and social factors and also, of course, to utilise the very comprehensive perinatal data in an investigation of the relationship between factors at or before birth and subsequent health and development. The Tables For reasons of space, most of the tables and all of the questionnaires appearing in the text of the main Report have been omitted in this version. The tables reprinted here are: A1 Southgate Reading Test scores A2 Problem Arithmetic Test scores A3 Teachers' ratings of reading ability A33 Past history of specialist clinics and services A40 Assessment of speech intelligibility 
|
