​Pack them in?

When we talk about class size it becomes a really contentious issue. Ministers are criticised for an agenda of raising them; Education Endowment Fund (EEF) curate research evidence that suggests that reducing class size is costly and has little impact on outcomes for learners (1); and teachers will say that in certain contexts ‘big’ classes are unmanageable and detrimental to learning. All I have just stated is based on no actual quantity and derives from an ill-defined concept and speaks to an emotionally charged dialogue. Raising or reducing class size from what to what? Is there a baseline from which this comparative conversation has meaning? The press and profession will use these emotive terms and then exemplify extremes, but what matters? What numbers are we actually talking about so what numbers make a difference?

In this SMARTcurriculum blog I am not seeking to raise the educational, pedagogical arguments for the impact of class size on outcomes, simply the efficient use of the funding provided and the assumptions of trying to achieve one goal within a definable constraint. If there were to be a change in policy and the related funding appropriation then the metric could change, but for now the principle stands the current funding assumes a capacity of delivery.

Most Integrated Curriculum Finance Planning (ICFP) systems talk of a benchmark class average of 27 or 28. For ease I am going to use 27 as a start point as it allows a degree of generosity and flexibility, others use 28 which, as you will see, has implications on the structures we design. Why 27 as an average class size across a year group? Class units of 30 are normally how a school is described and constructed in the UK. We use it to inform the ‘Forms of Entry’ for a state school based on units of 30 children, however, this is not a practical number to use across a whole curriculum. Let us illustrate with an example of a Year 7 programme with a 50-period cycle.

46 lessons will be used for the core subjects; English, Maths, Science, Humanities, Languages, Arts, Drama, Music and will normally exist in units of 30, if the school is full. For a subject like Design & Technology based in specialist workshops, food preparation rooms, and textiles rooms will typically be organised into groups of 20 learners. We accept some debate on this as, increasingly, more groups are increased to 23-25, we will illustrate base on 20. The allocation of lessons would be 4 periods in this 50 period cycle.

To calculate the average class size for one class within this year-group. 

((46 x 30) + (4 x 20))/50 = 29.2. (Clearly there is some space to reduce other elements too).

Assuming this year-group is full and the curriculum delivery structure is appropriately planned, then the average class size will be above the average 27 threshold and for this school to be at the level it has the capacity to drop to 28 in the core subject groups. To use the base of average 27 across a whole year group is therefore realistic and respects those schools that manage higher levels of mid-term admissions and mobile populations or have spaces on their roll.

Now, if that is achieved in a year group the number of classes that are planned with an average of 27 or higher will equate to an efficient curriculum. Many example we see here score a negative efficiency score. So we have set out the reason for 27 as the base level for class size calculation, and why some will raise it to 28.

Lets look at the funding principles that now follow this structure;
  • In KS1 and KS2, 27 pupils bring in approximately £81,000 (27 x £3,000) in per-pupil-funding (this will vary across the country unless common funding arrives). If we assume lower than normal Special Educational Needs (SEN) and Pupil and Sports Premium Grant (PPG,SPG), then this can act as a raw figure. To teach the one class we need at least 1.1 teachers (allowing for 10% PPA allowance.)
    • If the average salary for a member of staff is £50,000including employment on-costs, the actual teacher salary would be £55,000. The teacher salary will be out to 68% of the grant income (we are not using on costs in this illustration and we have zeroed the ‘other income’).
    • However, if we have employed a non-teaching headteacher and give other senior staff leadership time the average teacher need to teach the one group will increase, it is more likely that we will employ 1.4 staff for this class and now the salary is £70,000. We have now moved from 68% of the income spent on teachers to 86% spent on teachers.
  • In KS3 and KS4, 27 pupils bring in roughly £100,000 (27 x £3700) in per-pupil-funding. If we assume lower than normal SEN and PPG, then this can act as a raw figure. Again, to teach the one class we need at least 1.1 teachers, allowing for 10% PPA allowance. If the average salary for a member of staff is £50,000 (with employment on costs) the teacher costs would be £55,000. The teacher cost is now 55% of the grant income.
    • Again, if we have employed a non-teaching headteacher and give other senior staff leadership time the average teacher need, to teach the one group, is 1.4 staff members are needed and now the cost rises to £70,000. We have now moved from 55% of the grant income spent on teachers to 70% spent on teachers.
  • In KS5, 17 pupils bring in roughly £85,000 in per pupil funding. If we assume lower than normal SEN and all learners are considered as full time learners with appropriate learning aims, then this can act as a raw figure. To teach the one class we need at least 0.825 teachers, allowing for 10% PPA allowance and the fact that these students are not in class for 100% of the weeks learning time. Again, if the average salary for a member of staff is £50,000 (with employment on costs) the teacher costs would be £41,250.
    • Again, in employing a non-teaching headteacher and other senior staff leadership time the average teacher need, to teach the one group, is 1.05 staff and now the cost is £43,312.
    • You can see here why schools believe that having a sixth from enhances the school budget- and it will, if the class average is 17. The truth is that many schools the average is well below 10. This then becomes a different story; the Sixth form costs the school and becomes a burden.

Clearly in Key Stage 1 there is a maximum of 30 (3) in a class before you need to put a second teacher in the room, that is a legal precedent that is set because the pupil teacher ratio is set at a maximum so that there are not massive class sizes.  Obviously in nursery provision there is a ratio that is set for safeguarding and good teaching of what is a very important stage of education. As an aside, a lot would say now that if we get that bit right and if we could resource that better a lot of the issues that happen later in education would not happen if that was done more effectively.

So, across the education system the 27 optimum average class size and structural 30 principle is what most schools should be built around. We also know that there are different issues in the Post 16 education where it would be very unusual to have classes of 30 or 27. Based on the document from the Department for Education (DfE) on class size projections (2), we generally work on 17 being the optimum for a sixth form group. It can vary between 15 and 17, but 17 is generally the number that most systems look to as the average class size across a sixth form.

Rather than just saying that 27 in a class is what’s normally achieved across a school, there are some variations in how that works in different contexts. Obviously, if a school has a sixth form and there are some smaller classes, the average class size for a school will be smaller. If it is an 11-16 school, it is more possible to get classes with 27 as an average if the school roll is full and it is more possible in middle schools and junior and infant schools because they tend to be built around that class size.

One of the biggest challenges to that process though is local authorities and schools negotiating Pupil Admission Numbers (PANs) that are not established around the units of 30. We have seen recently are Primary schools where their PAN is set at 72, which is not divisible by 27 or 30, and we will discuss this further in a later blog. But simply to ask is that 2 classes of 36 or 3 classes of 24?

The funding comes with an expectation of school class size- stepping away from this basic unit is significantly contributing to the funding issues we are facing in 2019.


Research information to peruse.

1. Reducing Class Size, August 2018, Education Endowment Foundation ‘Teacher toolkit’

2. Class Size and Education in England: evidence report, December 2011, Department for Education DFE RR169.

3. Spring School Census 2019, Infant Class Size Validation, version 1.0, January 2019, Department for Education.

4. Class size does make a difference – but 30 is not a magic number, January 2016, Peter Blatchard, IOE London Blog.

5. ‘What Doesn’t work in Education. The politics of Distraction.’ 2015, John Hattie Melbourne Australia. Pages 10-11

6. Small class sizes for improving student achievement in primary and secondary schools: a systematic review, 2018, Campbell Corporation.


Please contact us for more information of SMARTcurriculum and the analysis it can provide info@cj-learning.com

Published by Chris Jones

CJ Learning Ltd is a collective of education and leadership specialists who bring a level of excellence to their work to make a significant difference in the lives of young people and communities through curriculum development and implementation strategies. CJ Learning Technologies is an Education Technology company providing the SMARTcurriculum Method in the form of an online Application to provide metrics and development strategies for curriculum and staffing provision.