Written evidence submitted by the Education Endowment Foundation (EEF)

Headline findings:

  • There is large variation in the strategies used in schools to improve pupil attendance.
  • The EEF’s recent rapid evidence assessment found evidence of promise for strategies including parental engagement approaches and responsive interventions that target the individual causes of low attendance.
  • The overall quality of evidence on attendance is weak; there is limited high quality evidence on approaches to support attendance and the majority of studies have taken place in the USA. This is an area in which further research is required.




  1. About the EEF

The Education Endowment Foundation (EEF) is an independent charity dedicated to breaking the link between family income and educational achievement. The EEF aims to raise the attainment of children facing disadvantage by generating, synthesising and mobilising the evidence on what works to close the attainment gap.


  1. Overview

Poor school attendance is a significant problem in England. In the 2021 autumn term, the most recent period for which data is publicly available, 23.5% of all pupils were persistently absent (defined as missing 10% or more of sessions) and 1.4% were severely absent (meaning they missed at least 50% of sessions). This is of particular concern to the EEF because evidence suggests many students with poor attendance are likely to be those from disadvantaged backgrounds​. 33.6% of pupils who were eligible for free school meals were persistently absent in autumn 2021, compared to 20.0% of pupils who were not eligible. 

Research has found that poor attendance at school is linked to poor academic attainment across all stages (Balfanz & Byrnes, 2012; London et al., 2016) as well as anti-social characteristics, delinquent activity and negative behavioural outcomes (Gottfried, 2014; Baker, Sigmon, & Nugent, 2001).

In 2022, the EEF conducted a rapid evidence assessment (REA) on attendance[1]. The REA examined the existing research on interventions that aim to improve pupils’ school attendance and the characteristics of these interventions, based on a systematic search of existing literature.[2] In doing so, the review provided some insights into strategies that could improve attendance outcomes, the overall quality of the evidence base and areas where more research is required.

It is important to note the limitations of the review and the wider evidence base on attendance:


  1. Summary of key findings on attendance

3.1.   There is large variation in the strategies that have been researched with the aim of improving pupil attendance. The EEF’s REA identified 8 different categories of intervention[4] and several additional strategies that were not included. The characteristics of interventions/approaches differed since they were attempting to address different barriers to attendance. For example, non-attendance due to bullying is more likely to require tailored, one-to-one support than non-attendance due to motivation.


3.2.   The overall quality of evidence on attendance is weak, and more research is required. The overall methodological quality of many of the studies was low. Few studies were rated as low risk of bias. Risk of bias aside, there are concerns about the transferability and relevance of much of the academic research for schools in England. Almost all of the studies took place in the USA, and there is little research on commonly used attendance strategies (e.g., the use of attendance officers in English schools).


3.3.   There is some evidence of promise for parental engagement approaches and responsive interventions that target the individual causes of low attendanceThe average impact was larger for targeted parental approaches. Responsive interventions (in which multiple interventions and target approaches are employed specifically to meet the needs of individual pupils) were also found to be effective. There may be crossover with these approaches and the approaches used in English schools by attendance officers. While these results are promising, the quality of studies means that they should be treated with caution.


3.4.   Many attendance interventions do not have sufficient evidence to reach a conclusion on effectiveness. There was not enough evidence to reach a conclusion for the efficacy of mentoring or behaviour approaches. Whole class teaching of social and emotional skills did have a positive impact overall, but the quality of the evidence is poor and there was significant variation in results, with the statistical uncertainty being consistent with negative impacts or larger positive impacts.


  1. Specific approaches

The following sections explore the evidence for specific approaches to support attendance.

4.1.   Mentoring


Mentoring interventions involve pairing young people with an older peer or adult, who acts as a positive role model. In general, mentoring aims to build confidence and relationships, to develop resilience and character, or raise aspirations, rather than to develop specific academic skills or knowledge (EEF Teaching and Learning Toolkit, 2021). In all the included studies in this review, mentors needed to meet with their mentees at least once a week and sessions varied in length. The subject of mentoring conversations varied and included attitudes towards school, exploration of a pupil’s extracurricular interests, providing general emotional support or encouragement and aspirations for the future and careers.


What does the evidence say?


The picture is mixed in terms of the impact of mentoring programmes on pupil attendance. The range of effects reported for the impact of mentoring on pupil attendance was large.[5] Because of this, and because of the small number of includable studies and high risk of bias, the EEF did not undertake a meta-analysis of mentoring studies in the REA.


It should be noted a larger evidence base on mentoring interventions does exist. For example, the EEF’s Teaching and Learning Toolkit identifies 44 studies that examine the impact of mentoring on academic attainment rather than attendance. The meta-analysis has found that the impact of mentoring varies but, on average, it is likely to have a small positive impact on attainment. However, these effects tend not to be sustained once the mentoring stops.


How secure is the evidence?


Overall, the evidence base linking mentoring interventions and pupil attendance is limited in size and has serious methodological flaws. Furthermore, all studies included took place in schools in the USA, where the drivers of school absence may vary and the potential for mentoring to influence pupil attendance may be different to the UK context. Research is needed to understand the impact of mentoring programmes in England.


4.2.   Parental engagement


Parental engagement interventions are those that involve parents in supporting and encouraging their children to attend school. The studies included in the review identified two distinct types of parental engagement interventions: communication and targeted planning support with family members and/or guardians.


Parental communication interventions aim to increase awareness of the consequences of absenteeism or target commonly held parental misbeliefs undervaluing the important of regular attendance. These interventions aim to increases guardians' role in improving student attendance by building awareness or eradicating misconceptions. Communication can occur in a range of ways including email, phone-calls, text messages and post, varying in quantity and time of day. Typically, these messages state information about the importance of attendance and add in specific information about the child’s attendance history.


Targeted parental engagement interventions are responsive in nature and include approaches that involve staff having discussions with parents to gain information about the reasons for low attendance and collaboratively planning support students and their families need to overcome attendance barriers. These interventions are usually more intensive with families having access to multiple services such as counselling, mentoring, resources and family activities.


What does the evidence say?


The overall impact of parental communication studies was a very small positive impact. Six of the seven studies included in the meta-analysis were considered low risk of bias, primarily because they employ large sample sizes (typically over 5,000 student) and control for confounding factors. While the number of studies included in this analysis is small, and they were mostly conducted in the USA, we can have some confidence in this finding.


While the overarching premise of the interventions in this category was to communicate and send messages to parents, the design models varied considerably. Some of the interventions consisted of one message type that was sent to parents e.g., nudge letters focusing on the importance of students’ attendance to their learning and the number of days the school had missed the previous year. Other interventions comprised of different types of messages to improve attendance. There was not enough data to compare the effectiveness between these types of interventions.


The overall impact of targeted parental engagement studies was small and positive. The small number of studies and the varying risks of bias mean this finding should be interpreted with caution. There is also not enough research to identify which specific elements of the targeted parental engagement approaches were most impactful on average.


A key component of the interventions was building effective partnerships between schools and parents, through discussions, meetings and conferences. The purpose of these partnerships was to identify issues leading to absences and collaboratively source effective solutions to the problems. All interventions in this category were multi-component in nature, featuring a range of services available to parents and pupils. Some examples of services include counselling, parent training and sharing of resources.


How secure is the evidence?


For communication intervention studies, none were rated as high risk of bias and larger sample sizes were used in comparison to other intervention categories, indicating a subset of higher quality studies.

However, the findings overall do have limited scope for generalisability, in particular applicability to the English context since all but one of the studies took place in the USA. Further research is required to see if these interventions have a positive impact on attendance in English schools.


4.3.   Meal provision


Meal provision interventions refer to school breakfast or lunch programmes where these meals are provided to pupils at no cost. Meal provision interventions aim to improve pupils’ experience of school, and subsequently assist good attendance. The evidence on meal provision may be relevant to the committee's consideration of the impact of school breakfast clubs and free school meals on improving attendance for disadvantaged pupils.


What does the evidence say?


Despite the large number of total studies, there was high variation between the different types of analyses, which means that an overall meta-analysis could not be conducted. The meal provision studies identified through the systematic search suggest that the provision of universal breakfast programmes has had either a null or small positive effect on pupil attendance, although some studies found that impacts were not identifiable in every year of the programme’s implementation, or were not statistically significant.


Some studies examined the impact of breakfast provision within the classroom in comparison to providing breakfast in the school cafeteria. The results were either positive or null. A promising area for future research might be around how to maximise the impact of breakfast clubs. EEF is currently conducting a rapid evidence assessment specifically focussed on breakfast clubs which will be published in the Spring.


One study was conducted in England and found a non-significant positive result of the Magic Breakfast programme on school attendance. It is important to note that the evaluation of the Magic Breakfast programme found an average of 2 months’ additional progress for pupils in Key Stage 1. An impact on attainment was not seen for pupils in Key Stage 2 (Crawford et al, 2019).


How secure is the evidence?


We rated all included meal provision studies as having either some risk of bias or low risk of bias. Compared to other areas of the review, these studies tended to have large sample sizes of pupils and schools and employed more rigorous non-experimental methods.


4.4.   Extra-curricular activities


Extracurricular activities interventions are those that provide additional educational opportunities outside of the regular curriculum. These approaches aim to increase student engagement in school which may then increase overall school attendance. Interventions included a variety of athletic and non-athletic activities. Athletic activities included a range of sports, while non-athletic activities included music, drama, play activities and others. All extracurricular activities were delivered after school. The evidence on extra-curricular activities may be relevant to the committee's consideration of the role of the Holiday Activities and Food programme and other after school and holiday clubs, such as sports, in improving attendance and engagement with school.


What does the evidence say?

There is limited evidence that extracurricular activities increase pupil attendance. Of the seven studies identified, while five studies reported that extracurricular interventions had a small positive impact on pupil attendance, effect sizes varied. One study found a null impact on attendance. However, these findings need to be read with caution due to small sample sizes and selection bias.

How secure is the evidence?


Due to the limited strength of the evidence, we have not been able to extract tangible findings on the impact of extracurricular activities interventions on attendance. There were a small number of extracurricular activities studies and there were high concerns of risk of bias. Again, all identified studies were carried out in the USA.


Further to this, it is difficult to assess the impact of extracurricular activities as four of the trialled interventions included additional components such as academic support, meal provision, and parental engagement. More research is required to establish a clear impact on improvements in attendance.



  1. Other approaches

5.1.   Teaching social and emotional learning (SEL)

Teaching SEL aims to build social and emotional skills and outcomes that have been shown to be correlated with school attendance. Interventions are often delivered by school counsellors or classroom teachers and are delivered either in regular classroom time or as part of school assemblies.

The REA found a very small positive impact of whole class teaching of social and emotional skills on attendance overall but the result was not statistically significant.

None of the studies took place in English schools and only two studies were considered to have a low risk of bias. The limited quality of the studies and the questions around transferability of findings mean that the evidence summarised should be treated with caution. While the teaching of social and emotional skills may have other benefits, more research is required to understand whether, and in what contexts, these interventions have an effect on pupil school attendance.

5.2.   Responsive and targeted approaches

Many approaches to improving attendance do not have a specified “intervention” but instead aim to be responsive to the reasons for low attendance by an individual pupil. These approaches are often multi-component and may involve one to one support for the pupil that has low attendance. One of the key characteristics across all of these approaches is monitoring and identification of pupils that need attendance support and the reasons for low attendance.

The overall impact of targeted and responsive interventions was a small positive impact. The small number of studies and some concerns over risk of bias mean that this result should be treated with caution. There is not enough research to identify which specific interventions within the multi-component approach were most impactful on average.

5.3.   Incentives and disincentives

The review identified some interventions that encourage pupils to adopt certain behaviours by either offering rewards for good behaviour (incentives) or punishments for poor behaviour (disincentives). Underlying all the incentive and disincentive interventions is the idea that pupils can be prompted to sustain good behaviour through extrinsic sources of motivation.

Incentive interventions often promised certain rewards to pupils at the start of the academic term or year that would be redeemed at the end of this time window (e.g., prizes, financial assistance with future college fees, or cash), contingent upon pupils sustaining good behaviour throughout this time. The disincentives used in interventions came mostly in the form of penalties for pupils, and their parents and carers.

While most studies in this area found that the trialled interventions increased pupil attendance, effect size estimates varied in strength and statistical significance. One study (Bernard, 2014) also found that the punitive disincentive approach it tested led to a decrease in attendance, when compared to a non-punitive counterfactual intervention.

5.4.   Other approaches

Other approaches that have been used by schools to try and improve attendance include academic interventions; big structural changes to the school; and provisions of services such as mental health support. The evidence for these approaches is significantly weaker than other interventions described – this does not mean that the approaches are less effective but does indicate that they have been researched less than other interventions with regards to attendance outcomes.

  1. Current EEF activity

To further build the evidence base around attendance and exclusions, in March 2022 EEF launched a joint funding round in partnership with the Youth Endowment Fund, A Safe, Positive Place to Learn: Improving attendance and reducing exclusions’. This round was set up to find, fund and evaluate programmes and practices in England and Wales that could both keep children safe from involvement in violence and improve academic attainment by reducing absenteeism. In the coming weeks, the EEF and YEF will announce the programmes and evaluations that are going to be funded through this round.



  1. References


Baker, M. L., Sigmon, J. N., & Nugent, M. E. (2001, September). Truancy reduction: Keeping students in school. Juvenile Justice Bulletin. Retrieved from https://www.ncjrs.gov/pdffiles1/ojjdp/188947.pdf


Balfanz, R., & Byrnes, V. (2012). The importance of being in school: A report on absenteeism in the nation’s public schools. Education Digest, 78(2), 4–9.


Crawford, Edwards, Farquharson, Greaves, Trevelyan, Wallace, and Clarissa White (2019) Magic Breakfast Updated Evaluation Report, Education Endowment Foundation, London. Accessed from: https://d2tic4wvo1iusb.cloudfront.net/documents/projects/Magic_Breakfast_report.pdf?v=1630925849


Education Endowment Foundation (2018). Toolkit: Mentoring. [Online] Accessible from: https://educationendowmentfoundation.org.uk/education-evidence/teaching-learning-toolkit/mentoring


Patel, R., Kay, J., Blausten, H., Madgwick, H. and Stevenson, J. (2022). Attendance Interventions: Rapid Evidence Assessment [online]. Available: https://d2tic4wvo1iusb.cloudfront.net/documents/pages/Attendance-REA-report.pdf?v=1647348064


Gov.uk (2021/22) Pupil absence in schools in England: autumn term (online). Accessed at: https://explore-education-statistics.service.gov.uk/find-statistics/pupil-absence-in-schools-in-england-autumn-term


London, R. A., Sanchez, M., & Castrechini, S. (2016). The dynamics of chronic absence and student achievement. Education Policy Analysis Archives, 24, 112. doi:10.1080/10824669.2014.962696


Gottfried, M. A. (2014). Chronic absenteeism and its effects on students’ academic and socioemotional outcomes. Journal of Education for Students Placed at Risk, 19, 53–75.


February 2023



[1] Rapid evidence assessments summarise what is known from the research literature. Rapid evidence assessments provide a more structured and rigorous search and quality assessment of the evidence than a literature review but are not as exhaustive as a systematic review.

[2] The EEF searched for impact evaluations published since 2000 that evaluated an intervention with a primary goal of increasing school attendance and that reported on a measure of pupil attendance or absenteeism. A protocol for the rapid evidence assessment, including more detail about the methodological approach taken, was published on the EEF website and is available here (Kay et al. 2021).

[3] The quality of evidence was determined by conducting a risk of bias assessment. Each included study in the review was appraised on factors that could systematically affect the conclusions of the study, e.g. sampling approaches. An adapted version of the Cochrane risk of bias 2 tool was utilised.

[4] Mentoring; parental engagement; responsive and targeted approaches; teaching of social and emotional skills; behaviour interventions; meal provision; incentives and disincentives; and extra-curricular activities.

[5] From a large negative impact on attendance for one of the mentoring programmes evaluated in Hilbert (2020) to a large positive effect reported in McQuillian and Lyons (2016).