RESPONSE TO DAVID JESSON'S REPORT ON SECONDARY EDUCATION
IN KENT AND MEDWAY
Fred Naylor, Educational Consultant, News International 10 July 2002
The intentions of Professor Jesson, the author, and Dr Stephen Ladyman MP, who commissioned the report, are to be applauded. Educational arguments alone - relating to better performances - should determine the issue of selective v. non-selective secondary education. The ideology and politics that shroud the matter should be dispelled. This is an excellent starting point.
The Jesson Report's recommendations, described by Dr Ladyman as unequivocal, should therefore be carefully examined on the basis of its author's educational arguments alone.
Prof. Jesson's Major findings
1. "Grammar schools in Kent and Medway do less well and are performing at lower levels than other grammar schools" (emphasis added).
Prof. Jesson adduces evidence based on one year's results to show that Kent and Medway have more grammar schools at the lower end of the grammar school distribution range, and fewer at the upper end of the range - compared with the rest of the country. There is, however, no evidence to suggest that all Kent and Medway grammar schools do less well than other grammar schools (as may be implied in his description of this Major finding.)
This "Major finding", however, has no bearing whatsoever on the comparison between selective and comprehensive schools which Prof. Jesson set out to make. (emphasis added).
2 "Secondary modern schools show similar, if less pronounced characteristics".
Prof. Jesson defines a secondary modern school in a very idiosyncratic way: "we use the term to cover 'all mainstream, maintained non-selective schools in LEAs deemed wholly selective'."
Whilst all schools which fit the above description are certainly secondary modern schools, all secondary modern schools (i.e. schools whose pupils have not demonstrated the high level of academic ability required for a grammar school) do not fit this description. For example, the Wiltshire LEA is predominantly comprehensive, but contains areas, such as Salisbury, which retains its grammar and secondary modern schools. OFSTED figures reveal that there are 22 LEAs with secondary modern schools, whereas Prof. Jesson states that 11 of them are fully selective. He has, however, overestimated the number of fully selective authorities in the country. He seems to be unaware that Kent itself is not a fully selective authority, and that on his definition none of its schools can be said to be secondary modern. In addition to Kent several others in his list are far from fully selective. The Wirral, for example, is predominantly comprehensive. This mistake results in the professor having to include Wirral comprehensives in his category of secondary moderns. Clearly, any attempt to place the Kent and Medway secondary modern schools in the national distribution of such schools, on the basis of GCSE performances, is going to be vitiated by the presence of grammar-type pupils in the comprehensive schools, which have slipped in unawares.
This is a problem Professor Jesson has had for some time in handling the performances of secondary modern schools. In his article, The high price communities pay for their grammars, PARLIAMENTARY BRIEF, AUGUST 2001, he claimed that there were 15 fully selective LEAs. In less than a year he has shrunk the total of fully selective LEAs by 35%, and he has still got it far too high.
The Professor's failure to identify the number of fully selective authorities, and hence by his own definition the number of secondary modern schools, renders his investigation of the performances of secondary modern schools in these two LEAs, and in the rest of the country, completely untrustworthy.
3 "Secondary schools as a whole show a substantial degree of polarisation
in outcomes which is quite different from most other communities' experiences
of their schools; this may, at least in part, be responsible for the lower
than expected levels of performance."
The professor is right in saying that Kent and Medway's secondary schools show a substantial degree of polarisation in outcomes which is different from most other communities' experience of their schools. He went on to say (just following his Table 1). "It could be argued that this is an inevitable result of the selection process - but the point being made here is that it is a (sic) not a characteristic of most other education systems. (Other calculations show Kent & Medway more polarised than all other LEAs in the South East)." His first words reveal that he understands the nature of the selective system. Not only 'could it be argued' that the outcomes of schools in a selective system show more polarisation than the outcomes of schools in a comprehensive system; it is inevitable that they will do so. And because of this, they will differ from most other communities, because the majority of LEAs have fully comprehensive systems.
Let us consider the matter in more detail. The very nature of a multi-line system (e.g. grammar schools alongside secondary modern schools, which cater for those who do not meet the entry requirement for grammars) ensures that all the grammar schools achieve better overall academic results than all the secondary moderns. The English figures for 1999 showed that 96.4 % of all 15 year-old pupils in grammar schools achieved 5+ GCSE A* - C grades. The corresponding figure for secondary moderns was 32.8%. But this polarisation does not occur at the expense of the lower achieving schools. Polarisation at the expense of lower achieving schools could only occur if all the schools were capable of achieving the same results (i.e. had pupils with the same potential).
It is a feature of all multi-line systems, which run overlap courses,
that top pupils in the lower line schools achieve better results than
the least able pupils in the upper-line schools. This comes about because
the tests which are used to determine entry into the upper-line schools
are never capable of predicting achievement some years later. (It would
be most unfortunate if they were.) To judge the relative merits of selective
and non-selective systems it is necessary to determine which system
is the more efficient - in terms of the percentage of pupils (not
schools) who attain a given criterion.
For example, the percentage of 15 year-old pupils who achieved 5+ GCSE
grades A*- B in 2000 (see Note 1) were:
Comprehensive 22.7%
Grammar 82.6%
Secondary Modern 10.6%
Total (England) 27.15%
The comprehensive schools accounted for 90.2% of the total entry to secondary schools in 1995. There was less polarisation between their results, but these came well below the national figures. The greater efficiency of the selective system resulted, in part, from the fact that the top 11% of the secondary modern pupils outperformed the bottom 17% of the grammar school pupils.
The second half of this 3rd Main finding; "this may, at least in part, be responsible for the lower than expected levels of performance" showed that the Professor had no evidence to support his statement, and that in talking about "lower than expected levels of performance" he was unclear about the comparison he was trying to make.
If Professor Jesson is objecting to polarisation in a situation which guarantees the pupils a higher chance of success, compared with the comprehensive alternative, he is going back on his promise to eschew ideological considerations.
4. "The relative performance of pupils in Kent and elsewhere has
not been made available"
The last Major finding is true - in respect of the availability of KS2 performances. It does not, however, allow any conclusions to be drawn about performances. The professor's attempts to produce a number of tables, from which no conclusions could be drawn, was more of a PR exercise, containing elements of wishful thinking, than a serious attempt to present evidence.
There are many other criticisms that could be made of Professor Jesson's report. These are, however, of minor importance because they do not have a bearing on his Major findings. THESE MAJOR FINDINGS, as explained in detail above, contributed nothing to the debate on Kent and Medway secondary schools.
Other methodological weaknesses
1. Value-added statistics
The Professor claims to have carried out value-added exercises for KS2 - GCSE in respect of pupils who received KS2 grading in 1996. However, Schagen and Schagen - in their NFER study, The impact of Selection on Pupil Performance, presented on 19 October 2001, stated that "it is not yet possible to carry out a single value-added analysis of individual pupils progress from key stage 2 to GCSE." This would have to await the 2002 GCSE results.
2. Prof. Jesson cited an unnamed, non-LEA comprehensive school, which took pupils from Kent and had exceptional GCSE results. The pupils were said to have taken the same selection tests as the Kent pupils. In making comparisons between systems - comprehensives (taking 90.2% of the national age group) and selective schools (taking 10%) - it is quite unscientific to focus attention on a single comprehensive (especially when it is unnamed).
3. Specialist schools
The professor referred to his research work on City Technology Colleges, commissioned by the CTC Trust. He compared these with the national average. This work has been publicly criticized for failing to make allowances for the extra money the schools received, the extra length of the school day and the other background factors which can affect school performance. He certainly made no attempt to compare specialist schools with the selective system.
CONCLUSION
The hopes aroused by Professor Jesson's determination to conduct his investigation into the relative merits of the selective and comprehensive systems, on the basis of strictly educational arguments, were quickly dashed by a close examination of his evidence and the inferences he attempted to draw from it. His "Major findings", as explained in detail above, were full of errors, untrustworthy and provided no support for the idea that selective education in Kent should be replaced by comprehensive schools. The national evidence on the inferiority of the comprehensive system (as measured by GCSE performance) was ignored.
.
Note (1) The traditional criterion - 5+ GCSE grades A* - C - is
no longer valid as a measure of performance for the top 25% of the ability
range. This is because it is now achieved by nearly 50% of the age group.
The more appropriate measure of A* - B indicates the stretching
which academically able pupils achieve in grammar schools.
