One of the main outcomes of the investment into a new school or a number of new buildings will be to improve the learning environment, with an ultimate goal of raising levels of attainment. The success of a new school or a new set of buildings could be measured in a number of ways. What will be discussed here is one of those ways.
Previous research conducted by NFER found that in the first new school building to open, pupil attitudes to school and the school environment improved substantially.
However, to date there has been only limited research on attendance and attainment outcomes for young people, mainly because the initiative is relatively new and it is debatable whether there has been enough time for impacts at this level to be seen. This research is an attempt to look at the most recently available attendance and attainment data.
The research is based on Partnerships for Schools’ published list of completed BSF projects as of 21 July 2010 and the latest available version of the National Pupil Database. It suggests that whilst at this stage the progress made by pupils in BSF schools is not at the same rate as similar pupils in non-BSF schools, there may be some early signs of improved attendance outcomes.
In order to investigate the impact new school buildings are having on attainment outcomes we used pupil level data from the National Pupil Database to construct multilevel models. The latest data available was for GCSE results in 2009, and so our models only considered BSF projects which had opened by April 2009 – before that year’s GSCE examinations had taken place.
We considered a number of different model specifications, including controls for pupil prior attainment and the school’s contextual value added (CVA) score for the previous year. In all cases our models showed that pupils at BSF schools make, on average, less progress than would be expected based on their in-take and past performance.
Analysis of absence data involved the year 11 and year 9 cohorts in the 2008/09 academic year. Models were again constructed including a range of pupil characteristics and school-level indicators. An important difference to the attainment model was that the analyses were unable to control for a pupil’s ‘prior absence’.
However, we have included the school-level absence rates for the previous cohort of year 9/11 pupils, meaning that our models consider changes in absence rates between the 2007-08 and 2008-09 academic years – although this change is not pupil to pupil.
The year 11 model showed that after controlling for all significant pupil characteristics, and all significant school level characteristics, there was no significant difference in the level of absence between pupils in BSF schools and pupils in non-BSF schools. This finding was the same for models using authorised and unauthorised absence as the outcome. Within the year 11 model there were approximately 10,000 pupils in 50 BSF schools.
The year 9 model showed very similar findings with no overall difference in levels of absence between BSF schools and non-BSF schools, once the models had controlled for all pupil and school level characteristics.
However, when we considered the different types of BSF project, we found that after controlling for all significant characteristics, pupils in schools with a mixture of rebuild and refurbishment had, on average, significantly less unauthorised absence, when compared to other similar pupils in non-BSF schools. Although a significant finding, this is only for six schools and approximately 1,000 pupils and so should be treated with a degree of caution.
For the 2009 GCSE results pupils at BSF schools made less progress than other pupils, even when pupil and school-level prior attainment is taken into account. No such difference was found in overall attendance rates at BSF schools. Where combined new building and refurbishment projects had
been completed, unauthorised absence rates even seemed to be slightly better – although with the available data only school-level prior performance could be controlled for.
As attainment falls it is not uncommon to see a similar fall in attendance, and so we may have expected a negative relationship in the year 11 absence model as this cohort of pupils in BSF schools made significantly less progress between Key Stage 2 and GCSE than other similar pupils.
There is clearly a need for more research into these outcomes as this could be the first sign that pupils in BSF schools are improving their attendance. Or, it could just be the continuation of an already existing trend. Our findings for year 9 in particular add some weight to the belief that although BSF schools have not had an impact on attainment, they may be starting to impact on behaviour.
This study is based on a single cohort and its findings should be considered in this context. But the available evidence, although limited, clearly suggests new schools are seeing positive changes in pupil attitudes to school and their resulting behaviour. It would be a normal expectation that these changes would eventually have an impact on raising standards and that we would see
improvements in attainment over and above what is seen in schools that have not benefited from this investment.
In combination with research that can reliably look at these attainment and absence outcomes, other areas also have to be investigated. How does a new school impact on the local community? Does it have benefits for increased access to adult education and learning? Do new buildings encourage different teaching styles? Do pupils develop different skills that are currently not assessed?
All of these are issues that could be addressed, and probably should be, if we are to understand the real impact of building new schools, on pupils, teachers and the community. Only after determining all the effects, can we then decide whether the investment has been worth it.