Stephen Machin and Sandra McNally of the London School of Economics report their recent findings on the effectiveness of the literacy hour, from a study based on the original National Literacy Project.

There are significant basic skills problems amongst the UK adult population: according to the 1999 Moser report, one in five adults are functionally illiterate. Raising teaching standards in schools to enhance literacy skills is important to ensure this is not true of future generations. The government has tried to improve literacy levels with the introduction of the National Literacy Strategy (NLS) in September 1998.

This education policy offers a highly-structured framework for teaching, setting out termly teaching objectives for pupils aged five to 11 and providing a practical structure of time and class management for a daily literacy hour. How do we know if this works? Does it help particular groups of students? Is it a cost-effective policy?

To answer these questions we investigated the literacy hour in the National Literacy Project (1996-98), the forerunner of the NLS. With data from the National Literacy Project (NLP) we examined whether the literacy hour works by comparing attainment of key stage 2 pupils in schools where the literacy hour was introduced to pupils in similar schools where it was not.

The NLP introduced the literacy hour to a subset of schools within a number of local education authorities (LEAs) in September 1996. Due to its partial coverage, some children were exposed to up to two years of the literacy hour, while others received zero exposure. Around 400 schools were in the NLP; most of which were poorly performing schools, many in inner cities.

For our research, the NLP offers a 'treatment group' of affected pupils (i.e. those in schools where the NLP was introduced) and a 'control group' of unaffected pupils (we mostly model this as schools in geographically adjacent LEAs, but we use statistical methods to look at schools with similar characteristics to the treatment schools). We compared the attainment of pupils in these schools before and after the introduction of the NLP policy, or the 'literacy hour'. The analysis is based upon national data-sets on key stage 2 pupil attainment and on a range of school-level data sources (such as school census data provided by the LEA and school information system). It is essential to use the NLP rather than the NLS to evaluate the literacy hour because a control group of schools only exists for the former.

The data shows a larger increase in attainment in reading and key stage 2 English for pupils in NLP schools compared with pupils not exposed to the literacy hour. For example, the NLP increased the number of children obtaining Level 4 or above in key stage 2 English by between 1.8 and 3.2 percentage points over the study period. Since we observe these same students at age 16, we can consider whether the effects of the NLP persist over time - in other words, did students who attended NLP schools do any better in their GCSE exams than students in the control group of schools? We find modest but positive effects of the NLP that persist to age 16, as GCSE English performance is higher for the NLP children. This helps ascertain that the literacy hour has positive effects on literacy into secondary school and the end of compulsory education.

The question could be asked: does the literacy hour affect some pupils more than others? In particular, does it improve literacy skills for pupils at the lower end of the literacy skill distribution? We considered one aspect of this, by looking at gender differences in our data. The reason for doing so is the inferior performance of boys in English (as demonstrated both at the end of key stage 2 and in GCSE results), a gender gap in education that has been present for many years. Our analysis of the NLP shows that boys in particular benefit from the literacy hour compared to girls. Hence the NLP policy has had an impact on reducing the gender gap in literacy skills.

On whether the NLP policy was cost effective, we estimate one type of benefit arising from the literacy hour - the labour market returns associated with better reading skills. To consider this, we estimate the relationship between age 11 reading scores and wages at age 30 in the British Cohort Study. However calculated, the cumulative wage benefits over a person's working life are estimated to be considerably higher than the per-pupil cost of the policy (about £25 per annum). Given the high ratio of benefits to costs, we can say that the policy has significantly raised pupil literacy skills, and done so in an extremely cost-effective manner.

These findings are of considerable significance when placed into the wider education debate about what works best in schools for improving pupil performance. The evidence suggests that public policy aimed at changing the content and organisation of teaching can significantly raise pupil attainment and can do so cost effectively.

Subscribe to Literacy Today