Earlier this week's Reform published the third blog in an experimental research series on value for money in schools. That blog presented the first set of results relating to the teaching workforce, this fourth and final blog will outline the remaining findings relating to school support staff, spending on energy and pupil outcomes. To recap, the first blog in the series gave an explanation of the context and rationale for the research and the second provided a description of the clustering method used to create groups of similar schools.
To compare the decisions different school types take in their spending and staff deployment, and how their pupils perform, we grouped schools according to the characteristics of their pupils (e.g. the percentage of students eligible for free school meals in the previous six years, pupil’s prior attainment levels, etc.). However, as explained in the previous blogs, the school clusters are not as homogenous as hoped, and thus do not allow for true 'like-for-like’ comparisons. The results are therefore tentative, but still provide what we believe is a valuable perspective, and we hope others interested in value for money in schools can build on it in future research. The estimations of how far a school type is from similar schools in terms of economy, efficiency or effectiveness, are based on weighted averages across clusters, as explained in this week’s previous blog.
Spending and deployment of support staff
The method with which local authorities and academies fill in their Consistent Financial Reporting accounts does not allow for any distinction between spending on different types of support staff. Spending differences reflected in Figure 1 therefore cover a wide range of roles, from teaching assistants to librarians. However, the second and third efficiency indicators (see Figure 2) do distinguish between teaching assistants and other support staff, as workforce data is more detailed.
Primary voluntary controlled and academy converter schools are furthest below their cluster averages when it comes to spending on support staff. Interestingly, voluntary controlled schools still have the lowest pupil-teaching assistant ratios (see Figure 2), which may mean that they pay their teaching assistants less, or potentially this difference might be explained by the fact that they have more students per 'other' support staff members (i.e. higher ratios). On average academy converters have pupil-teaching assistant ratios that are 10 per cent higher than other similar primary school, meaning that they have more pupils per teaching assistants, which is consistent with their lower average levels of expenditure on support staff.
Secondary sponsor led academies spend on average 29 per cent more on support staff than the average of other similar schools, yet their pupil-teaching assistant ratio are on average 15 per cent higher than other school types. In addition, their pupil-support staff ratios are also on average amongst the highest for secondary schools across, all school types. Sponsor led academies thus have fewer teaching assistants and support staff per pupil, but higher levels of expenditure. This might be explained by the fact that many of these academies are in the capital and salaries receive a London weighting.
The use of TAs more than trebled between 2000 and 2015, seemingly driven by intentions to reduce teacher workload and facilitate greater inclusion of pupils with special educational needs (SEN). However, hiring teaching assistants is considered to be a costly and relatively ineffective measure. Studies suggest that there is a negative to no impact on student performance overall, as well as on the performance of the specific groups that teaching assistants are meant to support. It is how teaching assistants are deployed that really matters not their sheer number. A lower ratio between pupils and teaching assistants might therefore suggest that schools are not allocating their resources in the most effective way.
Energy
Energy costs are often the second highest form of expenditure in schools after staff. It is important to note that in general secondary schools will have higher energy costs than primary schools. This is true in the sample of primary and secondary schools used in Reform's analysis. As the Carbon Trust points out, the difference is mostly explained by "secondary schools’ longer hours and larger number of students, as well as more widespread use of electrical equipment in ICT, science, sports and crafts lessons."
As explained in the first blog, this economy indicator is meant as a proxy for how well schools negotiate energy contracts. Reform recognises that this is an imperfect measure as there are many factors that affect the energy efficiency of a building, which cannot be been taken into account due to lack of available data.
The most notable differences in average energy expenditure per pupil are amongst secondary schools. Secondary converter academies spend on average eight per cent less on energy per pupil than other similar schools, interestingly sponsor led academies spend on average 22.5 per cent more than other similar schools.
According to the National Association of School Business Management, "of the £millions schools spend each year on the procurement of electricity and gas…an average of 5% cost reduction can be achieved by better procurement." Additional savings could be made through building renovations and reducing consumption.
Attainment
Having an effective school system – one that manages to raise attainment, particularly for pupils from disadvantaged backgrounds – is one of the most powerful tools to deliver social and economic prosperity. Since the early years of academisation – introduced in 2002 – the school landscape across the country has changed drastically. Currently, 22 per cent of all primary and secondary schools are academies. Whilst academisation has been extensive amongst secondary schools, with 61 per cent of establishments having obtained academy status, it has been less wide-spread amongst primary schools, with only 15 per cent of primaries as academies. Raising attainment in the education system is the fundamental goal of academies, particularly the sponsor led ones. However, evidence on the effectiveness of academisation is mixed.
As explained in the first blog of this series, the average point score is a simple measure of raw attainment, however, the clustering procedure applied to the sample of primary and secondary schools should have enabled similar schools to be compared to each other (e.g. secondary schools within clusters should have similar levels of prior attainment). Importantly, these comparisons between school types do not attempt to draw causal links – that is, the school's performance is not necessarily due to the school type. An Education Committee enquiry has underlined that "current evidence does not allow conclusions to be drawn on whether academies in themselves are a positive force for change."
The five different primary school types have relatively equal levels of performance when compared to similar schools, with academy converters averaging scores about one per cent higher than other similar schools and voluntary controlled schools about 0.5 per cent lower. The picture for secondary schools is slightly different, with only three school types – sponsor led academies, foundation and voluntary aided schools – exhibiting similar levels of performance compared to the average of schools within their cluster. Nevertheless, as we have previously mentioned these results are tentative and thus no robust conclusions can be drawn.
Further research
This experimental blog series was an attempt to apply an innovative technique to cluster similar schools and understand which school types provide better value for money. Due to uncertainties in the robustness of the school groupings, the results presented in this and the previous blog are tentative.
As explained in the first blog, the financial constraints that schools face render it essential to find new ways of increasing efficiency. However, this must not come at the cost of reduced levels of attainment (i.e. effectiveness). Building a school system that manages to raise attainment, particularly for deprived pupils, is one of the most powerful tools to enable social mobility.
Schools appear to have relatively similar approaches to their expenditure on, and deployment of, teachers as shown in this week's earlier blog. However, they have some notable differences in their expenditure on teacher training and development. Most primary and secondary school types do not have drastically different levels of expenditure on their support staff except for sponsor led academies (see Figure 1). Interestingly, the way in which they deploy their teaching assistants and support staff seems to be quite different as shown by Figure 2. On average per pupil expenditure on energy is similar when school types are compared to schools within their cluster, except for sponsor led academies who spend much more than the average of similar schools (see Figure 3). Finally, the different primary school types tend to have relatively consistent average levels of attainment when compared to similar schools. The picture for secondary schools is more variable, as shown in Figure 4
Reform strongly encourages further research to better understand the drivers of value for money within schools, and we hope this experimental work provides a useful platform to build on. We would be delighted to discuss the work further.
