Colin Alston questions the slippery concept of national reading standards. Failing teachers. Failing schools. Failing LEAs. Failing teacher training colleges. All four have been placed in the dock by the Office for Standards in Education and the Secretary of State. Their crime? Collectively, they are blamed for the fact that many seven-year-old children are "failing to reach the expected standards" in reading. The solution? Yet more league tables.
We need to take a step back from the obvious politics of recent pronouncements to ask ourselves what we really know about the "expected standards" of seven year-olds. The Department for Education and Employment report on test results tells us that seven-year-olds "at level 2 have reached the national standards which are expected to be within the compass of a typical pupil of that age" (my italics). The DFEE also notes that the "standards . . . at level 2 should stretch a typical seven-year-old". More surprisingly, in the light of the "measure and blame" culture implicit in Ofsted's report on inner-city literacy standards, the DFEE says: "There is no underlying assumption about the precise proportion of pupils which should be at any particular level".
So, we are told that a typical seven-year-old should achieve level 2 in the key stage 1 (KS1) reading test. This is the expected standard - the national standard. If typical seven-year-olds were distributed evenly between schools and LEAs, then it might be fair to compare raw results of one school with another, or one LEA with another. But evidence suggests that a typical seven year-old is a very elusive creature.
Just how old is a typical seven-year-old? When Year 2 pupils are being SAT upon, some are not yet seven years old, while others are approaching eight. Their achievements in reading differ sharply. In Hackney, we found that 12 per cent more of the autumn-born (oldest) children reached level 2, compared with the summer-born (youngest). In maths, the difference is more startling, with a difference of 17 percentage points in the proportions reaching the national standard. On this evidence, tens of thousands of parents across the country will be worrying needlessly that their child (or their child's school) is below average. In the new lexicon, to be below the national standard is to be failing.
Nationally, only one in six primary children is eligible for free school meals, so a typical seven-year-old is not eligible. Two decades of research have demonstrated that pupils' home circumstances and socio-economic background are powerful influences on their educational attainment - not immune to school effects but highly resistant to them. In Hackney, 61 per cent of KS1 pupils are eligible for free meals, with some schools having over 80 per cent. Compared with their peers, we find that 17 per cent fewer of the "eligible" pupils achieved the national benchmark of level 2 in reading.
A typical seven-year-old is white. The 1991 Census showed that about 5. 5 per cent of the country's population is from an ethnic minority. In Hackney, over 70 per cent of seven-year-olds are from ethnic minorities Analysis of our schools' reading test results show wide variations in the attainment of the different groups, with some scoring highly and others less so.
What language does a typical seven-year-old speak? One in ten of our KS1 pupils has less than two years' experience of the English language. Unsurprisingly, relatively few of these children reach the national reading standard in a language which they are still learning to speak.
The typical seven-year-old is not six and a half years old, socio-economically disadvantaged, bilingual, black, or new to hisher school - nor a refugee, a traveller, nor in temporary accommodation. The typical seven-year-old, the typical school, the typical LEA and the typical unicorn have something in common.
On most pupil background indicators, Hackney is exceptionally disadvantaged. So are its neighbouring boroughs, which were slated in the OFSTED report on literacy. When inner-city LEAs focus on improving levels of literacy, a unitary national standard is of limited value. Teachers and schools need something more sophisticated if they are to set targets which address individual children's needs.
This does not mean setting targets which are lower than the national "expected" standard; they could be higher but more appropriate to the child. LEA-level contextual analysis provides teachers with vital information about the variations in reading skills, which are linked to pupils' age, gender, socio-economic background, language fluency, ethnicity and mobility.
To research the factors which influence pupils' reading attainment is not to seek excuses for poor outcomes or to reinforce "low expectations". It is to seek to understand the barriers to achievement in order to overcome them. Every parent and teacher knows that children's acquisition of reading skills is not a standard or a uniform process. Most inner-city authorities are engaged in complex and ongoing research into pupils' achievement. They have to do this in order to help schools raise the standards of a diverse and shifting school population. Findings are often shared with neighbouring LEAs, but there is little evidence that OFSTED wishes to look at the question of "expected standards" in more detail.
What is the reading standard that might be "expected" of a seven-year-old with one year's experience of spoken English? What is the "expected" reading level of a bilingual Kurdish refugee? Across LEAs, nationally, what is the average "achievement gap" in reading at KS1 between pupils eligible for free meals and those who are not? Do schools in prosperous areas do better with prosperous children than inner-city schools? How many bright children are reaching the national standard but not fulfilling their potential? Answers to questions such as these will help real teachers to help real children.
In the meantime, schools are judged unfairly, LEAs are league-tabled bizarrely and parents are shovelled raw results as if they have meaning. According to Chris Woodhead, poor reading standards are the result of poor teaching. Undoubtedly, they sometimes are, and must be tackled accordingly. But our research shows that the learning world of seven-year-olds is much more complex than this. Parents have a right to better information than is currently provided by raw league tables, rewritten OFSTED reports and media hype. At the very least, school performance tables could carry the type of health warning that was edited out of OFSTED's report on literacy: "Most of these characteristics - bilingualism, poverty, pupil and staff turnover - were outside the control of the schools, and where schools faced a combination of them their task was a difficult one."
Finally, it seems that the phrases "national standard" and "national average" have become synonymous. In 1995, the national average for the five-plus A-C grades GCSE indicator was 43.5 per cent. Few people realise that 70 per cent of LEAs failed to reach this standard! The national averages for KS1 results in 1995 included those independent schools which chose to submit their results. I wonder what proportion of LEAs "failed" to hit the national (notional?) average. What price a national standard? Will the typical seven- year-old please stand up?
Colin Alston is head of research, statistics and information, Hackney Council education and leisure services.