 |  |
| www.dg.dial.pipex.com | 843 readers since 19 Aug 2007 |
| |
| www.dg.dial.pipex.com | 843 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 14
by BP Davies, Lecturer in Social Administration, London School of Economics and Political Science A: Primary school education and the student yield of secondary schools 1. It is in some ways useful to compare the education system with an industry which has a variety of 'inputs', a number of stages, and a wide range of 'outputs'. Ideally, questions of resource allocation should be settled by calculating what combination of 'inputs' would most effectively produce the required 'output' of people of various attributes, given the technical coefficients of the production function, the relative prices of inputs and limitations or their supply. This analogy is of limited use in practice, however. There is no easy way of deciding the educational outputs required, since not only is the relative importance of different uses of outputs a controversial matter but also technical aids to judgement about the need for outputs for these uses are as yet primitive, and yield results which are unreliable. Too little is known about how inputs could best be combined to produce a given output, unless one defines output in a way that is no more than a description of the inputs. Many of the most important inputs and standards are difficult to measure. Others could be measured quite easily, but central and local authorities have not thought it worthwhile to do so. Thus variation in the rate of turnover of teachers or the number of secondary schools without science labs or library rooms are not recorded although the data exists for both and merely requires analysis and although the importance of both was emphasised in the Newsom Report. 2. There are very few indices of the final and intermediate output of the local authority education system. One can make rough estimates of the final output of school leavers at various ages. This is not an adequate measure of output since it takes no account of the quality of the school leavers. One cannot classify them by the number of O Level, A Level, ONC, HNC, CISE and other passes in various subjects groups obtained for instance. (2) It is unfortunate that we lack indices of educational success for all types of pupils, since American studies (and British experience) have shown that systems that are relatively successful with the bright children are not necessarily those which are successful with others. (3) Since practically the only output indices suitable for analysis measure the intermediate output of the system entering higher education, this note concentrates on them. Table 1 Coefficients of variation of indices of provision of primary and secondary education (1961-62) Table 2 Correlation of indices of standards of provision of primary and secondary school education and output indices, 1961-1962 3. At first sight, it might seem that variations in standards - i.e., variations in the quantities and quality of inputs - of primary school systems might have considerable influence on the intermediate output of secondary schools going on to higher education, and the final output of pupils leaving during or after a sixth form course. One reason for this is that standards of provision of primary education are more unequally distributed than most aspects of standards of provision of secondary schools. This is shown in Table 1, which uses coefficients of variation. (4) None of the coefficients of variation for secondary schools exceeded the coefficients for the equivalent indices for primary schools and for most indices the difference between coefficient values was substantial. However the proportion of pupils aged 13 in non-selective and selective schools was more unequally distributed than other important indices of secondary school provision. 4. The correlation coefficients measuring the degree of association between certain indices of primary school provision and indices of output shown in Table 2 might also imply that variations in important standards of provision of primary schools might cause difference in output. (5) Thus expenditure on teachers' salaries per pupil was clearly correlated with the proportion of the cohort staying on until their 17th year, the proportion receiving awards for universities and, to a lesser extent, the proportion receiving lesser-value awards (to other branches of higher education) and the proportion entering training colleges, although the coefficients are not high. Similarly total expenditure per pupil was correlated with the proportion receiving awards for universities, and the proportion staying on. But the degree of correlation between the indices of provision of primary education and the most important output indices was generally smaller than that between the indices of primary education and the equivalent indices of secondary education. Thus expenditure per pupil on teachers' salaries in the two types of school were correlated +0.37, expenditure per pupil on debt charges (reflecting the proportion of school places new) were correlated +0.22 and total expenditure per pupil were correlated +0.36. The proportion of classes oversize in the two types of school were correlated +0.18. The equivalent indices of provision of secondary education were in general more highly correlated with the output indices than were the indices of primary education, as is shown in the bottom part of Table 2. Thus it is likely that the direct effect of the variation in standards of provision of primary education on these outputs indices is small. Multiple regression analyses of the output indices on standards and socio-economic characteristics of areas support this conclusion. (6) 5. These results imply that it would be possible to have a slightly different pattern of variation in standards of provision between local education authority areas without affecting the pattern of variation of output of students from each authority. The pattern of variation in the yield of students, people likely to join what some have called the 'clerisy' [academics], is unlikely to be affected by the small change in the pattern of variations in standards which could be achieved without a very considerable increase in the amount of reallocable resources flowing into primary school education. Research of a different kind would be needed to test this hypothesis. These results tell one nothing about the effects of a redistribution of resources within an authority between schools attended by children who are most likely to go on to higher education and other schools. They tell one little about the effects of increasing or diminishing the total amount of resources devoted to primary school education. I argue elsewhere that a small redistribution of resources in favour of authorities with bad social conditions would leave the pattern of student yield unaffected, and that such a redistribution would be more compatible with the aim to maximise the contribution of education to economic growth than is usually thought. (7)
B: Variation in standards of provision and social conditions 6. The aim of this section is to describe the variation in standards of provision of primary school education and some other services enjoyed by children of primary school age, showing how each important aspect is correlated with social conditions which make it 'desirable' that services should be provided at a higher standard if one were to assume that the most important criterion for assessing the distribution of resources was the extent to which it accorded effective equality of educational opportunity. It is necessary to make some such assumption because the aims of some of the more important branches of the education service are many, some may conflict with one another, and the socio-economic attributes of local education authorities differ in relevance according to the aims considered most important. 7. The pattern of variation in standards provides some evidence about what groups of children receive resources, how area of residence influences children's opportunities, and how standards of provision in areas where a high proportion of children live in social conditions which reduce their educational opportunity compare with standards in other areas. Both the degree to which variations in standards accompany - are correlated with - variations in social conditions, and the relative inequality of standards are important. These two aspects must be considered together, since the assessments of a case in which there is a high degree of correlation between a standards index and social conditions can greatly depend on the degree of inequality of the standards index, and similarly a certain degree of inequality can have very different implications depending on the degree of correlation. Table 3 Intercorrelation of indices of social conditions, 1965 (i) Intercorrelation of social conditions 8. Table 3 shows the degree of intercorrelation of indices of social conditions. The indices include measures of three of the five factors distinguished by HR Simpson in Table A of Appendix II of The Home and The School as having a major influence on the educational achievement of children of primary school age. In addition to indices of social class, bad housing, and family size, there are indices of population growth, of the proportion of the population of primary school age, and of the proportion of the population born in newer Commonwealth countries. (8) All of these measure factors which can create additional needs for educational resources or higher standards of provision. (The tables for education welfare services include other indices of social conditions.) 9. Table 3 shows that, as Moser found, (9) social class indices were correlated with other socio-economic indices. Low social class, overcrowding, the absence of housing amenities, low value dwellings, and a high proportion of the population of infant and primary school age, and to a lesser extent large families and declining populations are all correlated with each other and thus interact to create an environment that obstructs educational progress. The correlation coefficients seem high enough for it to be feasible to compile an index of bad social conditions which would provide a reasonably good measure of the degree to which LEAs suffer from social conditions detrimental to the educational attainment of their pupils. (10) (ii) Primary school provision and social conditions 10. Table 4 shows the correlations between indices of standards of provision of primary schools and those of social conditions. The degree of correlation of most standards indices with the social conditions indices was low. Standards were, on the whole, neither strongly positive nor negatively correlated with social conditions detrimental to the educational development of primary school pupils. Moreover, most important standards indices were less variable than were social conditions indices. There is therefore little indication that the central government and local authorities have created a system which ensures positive discrimination in favour of (or against) the educationally difficult areas. 11. Expenditure per pupil on debt charges, reflecting the proportion of school places built since 1945, was more unequally distributed than total expenditure or expenditure on teachers' salaries per pupil and debt charges tended to be lowest in low social class areas. This is largely a reflection of the concentration of the building effort on areas with rapidly expanding populations. Authorities with a high proportion of their labour force in semi-skilled and unskilled jobs tended to have a higher proportion of their quota unfilled than other authorities, but the degree of variation in the proportion of the quota unfilled was very small. Table 4 Correlation of indices of primary school provision with social conditions 12. Teachers' salaries and total cost per pupil in 1950-1 were as weakly correlated with social conditions as they were more than a decade later. This is not surprising, since patterns of provision of local authority services show great continuity. This was as true of education as in general it was of local health, welfare and children services. (11) Table 5 shows the same pattern in primary education. One could predict over a third of the variation in teachers' salaries per pupil in 1961-2 from the pattern in 1950-1, and over four fifths of it from the pattern in 1958-9. Similarly, one could predict more than a quarter of the variation in total costs per pupil from the variation in 1950-1, and three quarters of it from the variation in 1958-9. This continuity of pattern was even more striking in the case of the 53 boroughs which Professor Moser classified as 'industrial'. (12) Among these the variation of teachers' salaries in 1950-1 accounted for one half of the variation in 1961-2, and the variation of total cost per pupil in 1950-1 accounted for 30 per cent of the variation in 1961-2. Thus patterns of expenditure on the principal items in the budget tend to persist over a long period. Table 5 Continuity in standards 13. A second feature of Table 5 is the decline in the relative inequality of the indices over time. The relative inequality of standards was slightly greater in the 83 county boroughs than in the industrial county boroughs alone, and the decline in inequality seemed to be greater. There was a similar decline in inequality of total costs per pupil up to 1958-9, but after that the tendency for inequality to decline among 83 boroughs seems to have been arrested, though the decline continued among industrial boroughs. (iii) Education welfare services and social conditions (a) School meals and milk 14. Most of the education welfare services analysed here have in common that they were provided mainly in order to assist poorer children to make more effective use of the education system. Therefore standards of provision should generally be highest where social conditions are worst. The services considered fell into five groups - school milk and meals, special education, the school medical services, nursery schools and recreation facilities. The correlation between standards of provision of school meals and milk services with social conditions indices are shown in Table 6. Variations in expenditure on meals and milk were dominated by variations in the former, the proportion receiving meals being negatively correlated (-0.26) with the proportion receiving milk. The proportion receiving dinners was highest in the high social class areas where social conditions were good. It was also high where a high proportion of the working population were women. (13) The proportion receiving milk was correlated with bad social conditions. The contrast in patterns may well have been due to the fact that a charge is imposed for meals. Thus the proportion receiving dinners tended to be low where the proportion of children in great need was high. Therefore it is likely that schools meals are not being consumed by many of those who are in most need of them. Survey research is needed to investigate whether this is so. Table 6 Correlations of standards of provision of school meals and milk services with social conditions, 1961-1962 (b) School health services 15. Table 7 shows that standards of provision of dental services were considerably more unequal than standards of provision of medical services, and that the number of dental officers was negatively correlated with low social class. Expenditure, which measures local authorities' financial effort to provide these services, was correlated with low social class, heavy industry, bad housing, and population density, but the coefficients were low. In general, the standards of provision of school medical services were uncorrelated with indices of social conditions, as were the indices for primary schools. (c) Nursery schools, special education, aid to pupils in secondary schools, and recreation facilities 16. Many authorities did not provide nursery schools. It can be seen from Table 8 that the more extensive the provision of schools, the higher the pupil/teacher ratio. Thus the authorities providing nursery education on the largest scale may be using their staff most efficiently. Provision was uncorrelated with bad social conditions. (14) 17. Expenditure on special education was high in industrial towns and in bigger authorities. Expenditure on aid to secondary school pupils was uncorrelated with social conditions. Although expenditure on recreation facilities was high in densely populated areas, it was uncorrelated with other social conditions indices. (d) Inequality of standards of provision of education welfare services 18. Indices of standards of provision of these education welfare services were much more unequally distributed than the most important indices of standards of provision of primary and secondary education. This can be attributed partly to the fact that the authorities and those using the services both have greater scope for choice in the provision and use of welfare services than in the provision and use of education itself. (iv) Related services, education welfare services and social conditions 19. As Table 9 shows, provision of day nurseries was uncorrelated (-0.11) with the provision of nursery school places, but was markedly correlated with the ratio of women to men in employment, population size and the proportion of live births illegitimate. There were lower positive correlations with population density and the proportion of the population born in the newer Commonwealth countries. Expenditure on child welfare services by health departments correlated with standards of provision of most education welfare services, particularly expenditure on school medical services. Standards tended to be high in areas where populations had been falling, where a high proportion of adult males were in manual jobs, and where housing conditions were poor, but the correlations coefficients were low. 20. Expenditure on health visiting - less unequally distributed than the other related services by health departments - tended to be positively correlated with expenditure on medical services. Standards were correlated with the proportion of women aged 35-9 widowed, overcrowded housing, falling populations and low social class. The number of child care officers was correlated with the proportion of women working and with need correlates such as the proportion of live births illegitimate and the proportion of the population aged between 5 and 11 and the proportion of the population born in the newer Commonwealth countries. The number was, however, negatively correlated with low social class. The proportion of officers trained was also correlated with the illegitimate birth rate and the proportion of the population of school age. The proportion of child care officer establishment filled was less correlated with social conditions. Thus standards of provision of these related health and children's services were not more strongly correlated with bad social conditions than were education welfare services, and standards of provision of the two sets of services were in general uncorrelated. The distribution of educational resources is not highly correlated with social conditions which generate needs for them, and the same is true for other social services for children. Table 7 Correlation of standards of provision of school health services with social conditions indices, 1961-1962 Table 8 Correlations of standards of provision of nursery school education, special education, aid to secondary school pupils and recreation facilities with social conditions indices, 1961-1962 Table 9 Correlations of standards of provision of related social services with education welfare services and social conditions, 1961-1962
Education correlation matrices: list of variables Nursery 1. Number or pupils on nursery school registers per thousand population, 1962. Primary 5. Pupils per full-time teacher in primary school, 1962.
Secondary 20. Cost of secondary school teachers' wages and salaries per pupil, 1961-2.
Student yield 37. Full value local education authority awards tenable 1961-2 at non-university institutions per thousand population, average of three age groups (17, 18, 19),1961.
Other 40. Net expenditure on special schools per thousand population, 1961-2.
Occupation and socio-economic class 60. Males in administrative, managerial and professional occupations as a proportion of economically active males, 1961.
Housing 69. Proportion of private households rented from a local authority or New Town Corporation, 1961.
Demographic 83. Per cent change in population, 1951-61.
Miscellaneous 97. Dental attendant, etc. in full-time equivalents (1960) per ten thousand pupils, 1961.
Footnotes (1) Limitations of space prevent a full statement of the results of the analysis of the 100 or so standards indices, but a fuller analysis is to be published elsewhere. See BLEDDYN DAVIES Social Needs and Resources in Local Services, London, 1967; and a forthcoming Occasional Paper in Social administration. (2) The regional figures for numbers of O and A levels gained are based on a sample which is too small to give accurate results for individual authorities. They could perhaps be used to measure the output of individual authorities if results for several years were added together. (3) See SM GOODMAN The Assessment of School Quality, New York, 1959. (4) The Coefficient of Variation is used to measure the degree of inequality of indices. The coefficient measures the relative dispersion of the scatter of readings around the mean of the readings. It is independent of the units in which the index is stated, but it can be misleading if the distribution of readings is very skewed or if the relative values of index readings in relation to zero is arbitrary. (5) The correlation coefficient is used to describe the degree of association between variables. It measures the degree to which readings which are greater or less than the mean of one variable tend to be associated with readings which are greater or less than the mean of the other variables. If high readings in one variable tend to be accompanied by high readings in the other, the sign of the coefficient is positive, while if high readings in one tend to be accompanied by low readings in the other, the sign is negative. The highest value a coefficient can take is +1.00 or -1.00, and the lowest reading 0.00. The coefficient understates the degree of correlation if the association is non-linear. (6) These analyses will be described in the forthcoming Occasional Paper, which will also deal with the complicated and fascinating pattern of variations in standards of provision of secondary and other forms of education. (7) The Social Needs and Resources in Local Services, Chapter 12. (8) JWB DOUGLAS The Home and the School, London, 1964, Appendix II (9) CA MOSER and W SCOTT British Towns, London. 1961. (10) Such an index will be presented and the extent to which it accounts for variation will be analysed in the forthcoming Occasional Paper. Op. cit. (11) See BLEDDYN DAVIES Social Needs and Resources in Local Services, op. cit. (12) CA MOSER and W SCOTT British Towns, op. cit. (13) The correlation between low social class and women's working is very low, so that it would not be a statistical impossibility for the proportion receiving dinners to be positively correlated with both to a substantial degree. (14) An analysis of variance of the towns which provided a service also showed an absence of correlation with indices of social conditions. 
|
