Among students who continue in FTE post-16, those who were EVER6 at age 16 still make less progress and have a lower average ‘A’ level science score than their NonFSM peers.. The si
Identifying the link between disadvantage and attainment and participation in science subjects
To address this aim, this report analysed two data sets: the National Pupil Database for England (NPD) and the Avon Longitudinal Study of Parents and Children (ALSPAC).
The analysis of the NPD, presented in Chapter 2, updates the previous review as it uses the
Using the 2015 attainment scores, the study extends previous work by examining how socioeconomic status (SES) relates to pupils’ progress during compulsory education Progress is defined by whether later assessments (e.g., Key Stage 1) are in line with what would be expected from earlier attainment (e.g., end of Reception) If later attainment matches these expectations, the SES gap does not close, and SES and progress in science are not related If later attainment is below expectations, SES is related to both the initial attainment gap and to progress in science learning If later attainment is above expectations, the SES gap closes and SES is negatively related to progress in science attainment, meaning pupils from lower SES make more progress than those from higher SES.
Across post-compulsory education, this report examines how socio-economic status shapes participation after age 16 and how these participation patterns relate to A level science achievement It quantifies the extent to which A level science attainment gaps are driven by these varied patterns of post-16 participation and whether such gaps continue to develop during the 16–19 age phase The analysis highlights how participation differences contribute to science attainment disparities and tracks the persistence or evolution of these gaps as learners move through late adolescence.
Analyses of the ALSPAC data, as presented in Chapter 5, extend the findings from the previous report in two key ways First, they use a different measure of socioeconomic status (SES): the highest level of education attained by the mother Unlike eligibility for free school meals, which is a binary (yes/no) indicator, maternal education provides an ordinal scale and is considered in the literature (Gottfried, Gottfried, Bauthurst, Guerin, & Parramore, 2012) to be a more robust SES index; see Appendix 1.1 for an overview of measures of SES in education Second, the analyses here calculated school SES and investigated whether, in the UK as in other contexts, school-level SES exerts similar associations with outcomes.
17 countries, school SES and individual SES make independent contributions to the prediction of science attainment.
To explore the cause of any attainment gap with the aim of helping the EEF select projects
This report updates and extends the 2008 report by approaching the investigation of causes of the SES gap in science learning (i.e achievement and participation) in three ways
The initial approach reviewed studies that include data on socio-economic status (SES) and science learning, along with explicit hypotheses about what might drive the SES–science learning relationship The central question asks how parents’ SES translates into influences on their children’s science learning It is highly likely that the SES–science achievement link is explained by another variable, an intervening variable (see Appendix 1.5 for definitions of intervening variables, mediators and moderators), which is considered the causal factor behind the association Among studies proposing explicit mediating mechanisms, two plausible hypotheses emerged—often described as the “opportunity gap” and the “interest gap.” The review of this research is presented in Chapter 3.
A second approach to explaining the attainment gap extends the earlier review by applying a new, rigorous literature synthesis that brings together studies measuring science learning independent of SES and studies examining SES effects on cognitive skills related to science learning, such as literacy Since SES is linked to literacy, which could act as an intervening variable between SES and science attainment, the review focuses on data showing whether science learning relates to literacy and whether SES gaps exist in literacy skills relevant to science Although the exact causal role of these cognitive factors remains hypothetical, converging evidence from these diverse sources makes a plausible case for a mediating effect The findings from these studies are summarized in Chapter 4.
In our third investigation into the causes of the SES–science learning gap, we analyzed the ALSPAC dataset, which includes measures of plausible mediators of SES–science attainment identified in Chapter 4, alongside measures of both SES and science achievement We tested whether SES remained a significant predictor of science attainment after accounting for these mediators The analysis extends the 2008 Royal Society report, which did not examine potential causes of the SES-related attainment gap The results of these analyses are presented in Chapter 5.
Although research often references parents, in practice the measures are typically applied to caregivers—the person the child lives with when the child does not reside with the parents.
To identify promising pedagogies, interventions and programmes, within school and/or
This review significantly extends the 2008 report by evaluating research on pedagogies, interventions and programmes that positively impact learning for disadvantaged students It examines diverse intervention types, ranging from topic-focused approaches (e.g., electricity, buoyancy) to those aimed at developing students’ scientific thinking skills, such as the 'Let’s Think' program, which builds on CASE—Cognitive Acceleration through Science Education.
"Acceleration through Science Education" was not reviewed in the previous report, which acknowledged the need to examine the extent to which specific teaching methods can improve science attainment among pupils from lower SES backgrounds This omission highlights a key research gap in evaluating instructional strategies that help close the science achievement gap for low-SES students and informs future policy and classroom practice.
Identifying the link between socio-economic disadvantage and participation and attainment in science: An analysis of the England National Pupil Database
Chapter 2 presents the latest evidence on the relationship between pupils’ socio-economic status (SES) and their participation in and achievement in science, using EVER6 as the main measure of socio-economic disadvantage, defined as whether the student has been entitled to a Free School Meal (FSM) at any time in the last six years.
The EVER6 achievement gaps in Science are large The odds of achieving conventional benchmarks of success in science are much greater for NonFSM students than for EVER6 students;
The achievement gap in science grows over time The gap seems to be largest at the end of secondary school (age 16)
The substantial EVER6 gaps in science at age 18/19 are largely driven by low achievement in science at age 16 and low levels of participation in Full-time Education (FTE) post 16
Among students who continue in FTE post-16, those who were EVER6 at age 16 still make less progress and have a lower average ‘A’ level science score than their NonFSM peers
According to EVER6 data, the achievement gaps in science are broadly similar in size to the gaps seen in other subjects, including English and mathematics, as well as the overall score This indicates that science is not a special case when it comes to achievement gaps.
Research indicates that the earliest measure of achievement in science, recorded when a child is seven, strongly predicts later science achievement Moreover, this early science performance is itself strongly predicted by the pupil’s Early Years Foundation Stage Profile (EYFSP) score at age five.
This section presents a secondary data analysis using the England National Pupil Database (NPD) to examine science participation and achievement across England It updates and extends the Royal Society’s 2008 report, SES and Science Education The study analyzes national trends in science participation and achievement from Key Stage 1 (age 7) through Key Stage 2 (age 11), GCSE examinations (age 16), and onward to A‑level and other Level 3 qualifications (age 18/19) Socio-economic status is measured by Free School Meal (FSM) entitlement recorded at any time in the past six years, known as the EVER6 indicator.
The purpose of the analyses are to:
Compare the size of the EVER6 achievement gap over the school period from
Foundation Stage to the end of Key Stage 5 using the most recent 2015 cross-sectional data;
Evaluate EVER6 gap in progress (both in science and overall achievement) within each Key Stage;
Where available chart trends in the size of the EVER6 gap over time;
3 This chapter was prepared by Steve Strand
Compare the size of the EVER6 achievement gaps in Science to the size of the gap in other subjects;
Breakdown the EVER6 gaps in Science at KS4 by other student and school characteristics, e.g by gender, ethnicity, school type and region
The measure of Socio-economic Status (SES)
Extensive and long-standing evidence shows a medium-to-strong association between socioeconomic status (SES) and educational achievement This link is documented in White's 1982 literature review and reinforced by Sirin's 2005 meta-analysis Sirin reports a moderate overall effect size (Pearson's r ≈ 0.31), noting that the magnitude varies depending on how SES and achievement are measured, as well as student characteristics such as age and ethnicity.
Various indicators of socio-economic status (SES) are used in the academic literature, capturing dimensions such as parental occupation, household income, the parents’ highest educational qualifications, and family social resources Many of these indices require detailed interviews with parents to obtain reliable measures and are not readily available within the school setting In contrast, area-based measures like England’s Income Deprivation Affecting Children Index (IDACI) estimate the share of children under 16 living in neighbourhoods where households are entitled to state benefits A key limitation of area-based SES measures is that they may not accurately reflect the specific circumstances of the pupil’s family.
In school-based research, the primary indicator used is whether a pupil is entitled to a Free School Meal (FSM) FSM entitlement serves as a proxy for low-income households, with eligibility defined for that purpose.
• Children in Income Support households
• Children in Income Based job Seekers Allowance households
• Children in Working Families Tax Credit households whose income (excluding housing benefits) is below 60% of median before housing costs
• Children in Disabled Person’s Tax Credit households whose income (excluding housing benefits) is below 60% of median before housing costs
• National Asylum Support Service (NASS) supported asylum seekers
In January 2012, the Department for Education (DFE) in England introduced the EVER6 measure, an indicator of whether a pupil has been entitled to free school meals (FSM) at any point in the last six years This six-year look-back expands the size of the identified group from around 18% to around 28% of the school-age 5–16 population.
No single measure perfectly captures a construct, and both FSM and EVER6 are binary indicators that suffer from threshold effects around the entitlement cut-off and a lack of differentiation within the larger group not entitled to FSM (Strand, 2014) However, because EVER6 is the most widely used indicator, it remains the measure most directly linked to entitlement status.
IDACI neighbourhoods are defined using Lower Layer Super Output Areas (LSOAs), the standard geographic units used in England for deprivation analysis In 2010, England had 32,482 LSOAs, each designed to be roughly equal in size and to contain about 1,500 residents on average These LSOAs provide the neighbourhood-level framework for calculating the Index of Deprivation Affecting Children (IDACI), enabling consistent, granular assessment of child deprivation across the country.
The Pupil Premium provides funding to schools, with 2015/16 rates of £935 per secondary pupil and £1,320 per primary pupil for pupils in the EVER6 group This funding is the measure of deprivation used in reporting for primary and secondary school performance tables and is the SES metric employed by the Education Endowment Foundation (EEF) in its evaluations In the current National Pupil Database (NPD) analysis, this SES measure is used as the deprivation indicator.
According to the January 2013 school census, 27.6% of pupils from Reception through to Year 11 were identified by the EVER6 measure, higher than the 18.2% eligible at any single point in time across Reception–Y11 A detailed year-by-year breakdown of the proportions identified as FSM and EVER6 for 2013 is included in Appendix 2.1.
Achievement gaps and effect size measures
Over the last six years, this chapter reports the size of the achievement gap between EVER6 and those who have never been entitled to Free School Meals (NonFSM), using the same units in which the data were collected—National Curriculum levels, test marks, GCSE grades, points scores, or the proportion achieving a given threshold (for example, the percentage achieving Level 4 or above at age 11 or the percentage passing the EBacc in science at age 16) Given the wide range of measures and the changes over time even when the same measure is used, standardised measures of effect size are useful, and we will report Cohen's d.
D for continuous measures and the Odds Ratio (OR) for binary measures These effect size measures are described in detail in Appendix 2.2
Achievement at Key Stage 1 (Age 7)