Impact Journal Logo

What we learned from evaluating the National Tutoring Programme (NTP) 2020/22: How schools can maximise the benefits of tutoring

Written by: Roland Marden and Pippa Lord
8 min read

Set up in 2020, in response to the disruption experienced during the COVID-19 pandemic, the National Tutoring Programme (NTP) aimed to help disadvantaged children in England catch up on learning affected by partial school closures. Since 2020, the NTP has reached a large number of pupils, including many disadvantaged pupils. Midway through its third year in January 2023, the programme had funded a total of 3.3 million tutoring courses in primary and secondary schools across England (DfE, 2023), with 87 per cent of schools participating in 2021/22 (NAO, 2023). At the time, although the programme did not achieve anticipated targets for reaching disadvantaged pupils, these pupils were more likely to receive tutoring than other pupils, with around half of pupils receiving tutoring in 2021/22 classified as disadvantaged and 25 per cent of all disadvantaged pupils receiving tutoring compared to 14 per cent of all pupils generally (NAO, 2023). 

Post-COVID-19, many schools have engaged with tutoring as a way of improving pupils’ attainment. Among schools participating in 2021/22, 81 per cent reported that tutoring had improved pupils’ attainment (Lynch et al., 2022), and recent research with a nationally representative sample of schools in England found that most senior leaders using the NTP (2022/23) believe that it has allowed their school to provide additional support to disadvantaged pupils and is improving their attainment (Moore and Lord, 2023). Reports from school leaders show that tuition is among their priorities for pupil premium spending: 22 per cent of school leaders reported that providing more one-to-one or small group teaching was their main priority (Sutton Trust, 2023). That said, not all schools are engaged in the NTP and tutoring is not yet embedded within schools as a method of supporting disadvantaged pupils (Moore and Lord, 2023). Moreover, recent research has found that many schools leaders are not convinced about the cost-effectiveness of tutoring or whether it is a long-term solution to closing the attainment gap (Moore and Lord, 2023). 

Additionally, the funding subsidy provided by the NTP reduced from 75 per cent to 60 per cent in the 2022/23 academic year and will be reduced to 50 per cent in 2023/24. Going forward, schools will want to use money spent on tutoring as wisely as possible.

The current design of the programme also gives more freedom to schools to decide how best to provide tutoring for their pupils. Funding is given directly to schools and then they choose whether to recruit tutors via the NTP supplier routes – tuition partner organisations and the academic mentoring programme – or do it themselves (known as ‘school-led tutoring’). The experience of the 2021/22 academic year, when 81 per cent of NTP tutoring courses were delivered via school-led tutoring, suggests that schools favour the option of setting up and managing tutoring themselves (DfE, 2022). With schools increasingly taking on this role, it is more important than ever that decisions are informed by evidence.

Working in partnership with Kantar Public and University of Westminster, NFER has been leading evaluations of the NPD since the programme’s inception in 2020 – first for the Education Endowment Foundation (EEF) with our partners, and subsequently for the Department for Education (DfE). Researching the programme for three years has put us in a privileged position of being able to investigate the impact of tutoring and observe the approaches that schools have used to implement it. Our evaluations have not been without their challenges, and there remains the opportunity to further fill the gaps in the evidence base around tutoring. 

One notable challenge was in assessing the impact of the programme on attainment for pupil premium (PP) pupils. These pupils are a key target for the programme and central to the goal of narrowing the attainment gap between disadvantaged pupils and their peers. While schools were encouraged to select PP pupils for NTP tutoring, in practice fewer than half (46 per cent) of the pupils who took part were PP in 2020/21 (Lord et al., 2022). Of all disadvantaged pupils, only a minority received tutoring. This meant that when the attainment of PP pupils in NTP schools was compared with PP pupils in non-NTP schools, many of these pupils in the ‘intervention’ group had not actually received tutoring. This ‘dilution’ issue made it challenging to detect any impact that may have been present. Not surprisingly, our analysis found that PP pupils in NTP primary and secondary schools made similar progress in English and maths compared with PP pupils in non-NTP schools. Where we were able to analyse outcomes in schools with higher proportions of PP pupils receiving tutoring (70 per cent or more), the results indicated that tutoring had a positive impact on Year 11 GCSEs (teacher-assessed grades) in English and maths (Oppedisano et al., 2022). 

Key findings and recommendations

Despite the challenges outlined above, findings from the 2020/21 impact analysis provided an emerging outline of factors associated with effective tutoring (the 2021/22 impact analysis had not been completed at the time of writing and is scheduled for publication in autumn 2023). These findings were complemented by the qualitative research conducted in the first two years of the programme detailing the experience of schools participating in the programme. Overall, our evaluation research has given us some valuable insights that can help school leaders and teachers to maximise the benefits of tutoring (see NFER Classroom, 2023).

Strategic aim and pupil selection

A key consideration when setting up tutoring is establishing its strategic aim and considering its role in supporting disadvantaged pupils. The wider evidence suggests that tutoring may be particularly beneficial for disadvantaged pupils (EEF, 2018). Schools should identify the type of pupils that should receive tutoring in the planning stage. This step helps to establish support for these pupils as the strategic purpose of the tutoring programme.

Group size 

Deciding on the group size and the mix of pupils in tutoring sessions is also important. Tutoring groups usually range from one to one to small groups of up to six pupils. The most common group size delivered by tuition partners in 2020/21 was small groups of three, and this represented over 70 per cent of tutoring sessions (Coulter et al., 2022). Our analysis did not find any straightforward relationship between group size and outcomes but, interestingly, in primary schools, small-group tuition in English was associated with higher attainment than one-to-one tuition in English (Lord et al., 2022). The wider evidence suggests that either one to one or small groups of up to three pupils is most effective, depending on the circumstances of the pupils. When group size increases above six, there is a noticeable reduction in effectiveness (EEF, 2022). Our research found that small groups can be particularly effective for pupils of similar ability. Whereas one-to-one tuition is suitable when a pupil has unique learning needs and will benefit from tailored teaching. Schools reported that one-to-one tuition is particularly beneficial for pupils with SEND (special educational needs and disabilities) and with English as an additional language (EAL) (Lord et al., 2022).

Session scheduling

How tutoring sessions are scheduled is also important. Tutoring should aim to be additional to classroom teaching. Unless there are special circumstances, it is best to avoid taking a pupil out of their subject class for tutoring in that subject. One obvious way of making tutoring ‘additional’ would be to schedule sessions outside of school hours. However, our findings were mixed on how the scheduling of tutoring was related to outcomes. One possible explanation might be related to the challenge of getting pupils to attend outside of the school day, leading to lower attendance. 

Some schools may be able to make after-hours tutoring work. But there are also approaches to scheduling tutoring during the school day that help to retain this valued ‘additionality’. One approach worth considering is to substitute tutoring sessions for non-core subjects, e.g. PE, art or music. This can work but care needs to be taken to avoid pupils regularly missing the same subject or their favourite subject. Some schools report that a rotating timetable works well to achieve this.

Duration and frequency of sessions

Receiving more hours of tuition was related to better assessment scores in English in primary schools, and better GCSEs (teacher-assessed grades) at year 11 in English and maths (Lord et al., 2022). For maths in primary schools, we found that delivered more frequently and over a short timespan were associated with higher maths scores (Lord et al., 2022). A key takeaway is that short and
frequent sessions are more likely to maximise pupils’ attention and maintain learning momentum (EEF, 2021a, 2021b).

Selecting tutors

Another important consideration is selecting suitable tutors. The wider evidence on tutoring suggests that tutor subject knowledge and teaching expertise are beneficial for learning outcomes (Lepper and Woolverton, 2002). Our analysis suggested that teaching expertise was particularly important for primary pupils, where having a tutor with QTS (qualified teacher status) or a PGCE (postgraduate certificate in education) seemed to be beneficial for English tuition (compared with other postgraduate qualifications). On the other hand, for secondary Year 11 pupils, tutors with other postgraduate qualifications were associated with tutees achieving higher GCSEs than tutors with other qualifications (i.e. undergraduate qualifications only or PGCE/QTS). Although these findings are not conclusive, it seems that teaching expertise trumps subject knowledge at primary level, whereas higher subject knowledge trumps teaching expertise at the more advanced secondary level.

Alignment to class teaching

One of the key messages from our research is that tutoring should be properly integrated into the teaching offer at a school and not treated as an add-on. The content delivered in sessions should be aligned to the subject curriculum being taught in the classroom. This is particularly important for maths tutoring, where introducing new methods can be confusing for pupils; the maths methods taught by tutors should be the same as those used in the pupil’s maths class. For English, the situation is slightly different, because there is a shared foundation of skills that is often applicable, e.g. reading, vocabulary, spelling and grammar.

Schools have found information sharing and communication with tutors to be a key part of ensuring that content is aligned with the curriculum. Subject teachers should ensure that information about the pupil’s subject knowledge is shared with the tutor. The tutor can then use this to tailor their sessions to the needs of the pupils.

Monitoring quality

As is the case for teaching generally, it is beneficial for schools to be proactive about monitoring tutoring. This will likely include monitoring pupil attendance of sessions, tutoring quality and the impact of tutoring on pupils’ progress. This may require dedicated staff trained in supervising tutors and skilled in developing effective tutoring practice (Cullinane and Montecute, 2023). This could also be an area in which subject teachers could be involved, monitoring attendance and progress of pupils from their classes.


With the NTP subsidy reduced and schools expected to fund and organise tutoring themselves, we are at a crossroads for the future of tutoring in the English education sector. In the effort to get learning back on track following the impact of the pandemic, many schools have taken advantage of the NTP to embrace tutoring as a valued teaching tool. Schools may continue to look to include tutoring as part of their high-quality teaching offer. We believe that teachers need more nuanced evidence about the cost-effectiveness of different models of tutoring to ensure optimal decision-making. At NFER, we look forward to helping to inform school choices in this area through our ongoing research and evaluation.

    0 0 votes
    Please Rate this content
    Notify of
    Inline Feedbacks
    View all comments

    From this issue

    Impact Articles on the same themes

    Author(s): Bill Lucas