Child poverty and education outcomes by ethnicity

• Children in Bangladeshi and Pakistani households were the most likely to live in low income and material deprivation out of all ethnic groups, while children in Indian households were the least likely.

• Children in Asian households were 2.5 times as likely, compared with the national average, to be in persistent low income during the period from 2013 to 2017.

• Pupils eligible for free school meals (FSM) made less progress between 11 and 16 years old than those not eligible for FSM, with national average Progress 8 scores of -0.53 and 0.06 respectively.

• Educational outcomes for Bangladeshi and Pakistani children did not follow this trend; Bangladeshi and Pakistani children who were eligible for FSM had higher Progress 8 scores than the national average.

• London had the highest Progress 8 scores for Asian, White and Mixed pupils. White pupils in the North East had the lowest Progress 8 score of all pupils (-0.28), and Chinese pupils in the East Midlands had the highest (1.22).

Child poverty in the UK is a growing issue and affects more than 4 million children. Growing up in poverty can have negative consequences for children's well-being and future life prospects, such as employment and earning opportunities (HM Government, 2014).

Young adults who suffer financial hardship as children have significantly greater than average chances (PDF, 113.95KB) of earning lower wages, being unemployed, spending time in prison (men) or becoming a lone parent (women).

There is a clear pathway from childhood poverty to reduced employment opportunities, with earnings estimated to be reduced by between 15% and 28%, and the probability of being in employment at age 34 years reduced by between 4% and 7%.

To understand the scale of child poverty, we can analyse the income that a family has available to spend or save (that is their disposable income). One commonly used measure of child poverty is children living in households in relative low income (also referred to as relative poverty). Relative low income helps us to understand the variances between low- and middle-income households at a specific point in time.

The relative low income measure provides a snapshot of the number of children living in poverty at a point in time, but does not tell us how long someone experiences poverty. To understand how long children experience poverty, persistent low income measures can be used to estimate the proportion of children living in low-income households in at least three of the last four years.

Income alone may not always reflect the extent to which a family can afford items and activities considered typical in society at a given point in time. The combination of a low-income threshold and an assessment of whether households are able to access essential goods and services (material deprivation), allows for a more holistic measure of children living in poverty.

Children living in poverty are more likely to have lower levels of educational outcomes (PDF, 930.61KB). The relationship between deprivation and education is crucial for understanding the significant impact deprivation has on later outcomes in adulthood. We measure this relationship by looking at the educational outcomes of children who were eligible for free school meals (FSM). Eligibility for FSM is related to receipt of income support benefits, such as Universal Credit. FSM-eligible children are more likely to be in low-income families than children who are not eligible.

The relationship between poverty and education outcomes is complex, and affected by multiple factors, including geography. For this reason, we will explore educational outcomes in smaller geographical areas in the context of the proportion of children living in low income locally. For example, regions in the North of England tend to have higher poverty rates (before accounting for housing costs) than regions in the South of England.

The aim of this article is to explore how both child poverty and educational outcomes vary for different ethnic groups, and to look at whether there is a relationship between these variables that is consistent across ethnic groups. In doing so, we will look at how children from different ethnic groups experience:

• poverty at a fixed point in time

• poverty that persists over time

• the inability to afford necessities

We will then assess the relative progress pupils from different ethnic groups make in education when compared with their peers who start at a similar level of educational attainment. We will use pupil-based and area-based proxies for child poverty to explore the relationship between child poverty and educational outcomes by ethnicity.

There are a number of ways of measuring poverty, and no single definition is universally accepted. This section focuses on three measures of child poverty:

• relative poverty (living in a low-income household)

• persistent poverty (living in a household in persistent low income)

• low income and material deprivation

Children living in low-income households

A household is defined as being "low income" if it has an equivalised income below 60% of the UK's median household income, before housing costs. Household income includes income from earnings, benefits, pensions and investments (and other sources), and is adjusted to take account of the household size and composition. In the financial year ending (FYE) 2018, the UK's median equivalised household income was £26,000 for a couple with no children, and £37,000 for a couple with two children under 14 years of age (before housing costs).

Children in Pakistani and Bangladeshi households were the most likely to live in low-income households. They were 2.8 and 2.4 times as likely, respectively, to live in low-income households, compared with children living in White British households, during the three-year average ending in FYE 2018.

Figure 1 shows that 47% of children living in Pakistani households, and 41% of children living in Bangladeshi households were living in low income. This was 30 and 24 percentage points higher, respectively, than children living in White British households and 27 and 21 percentage points higher than the national average.

In 2018, the Pakistani and Bangladeshi population had a higher rate of unemployment (8%) and the highest rate of economic inactivity (38%). This may explain their higher likelihood of living in low income.

In contrast, Indian and White British children were the least likely to live in low-income households; 17% of children living in Indian and White British households were living in low-income families, three percentage points lower than the national average.

The Other (36%), Mixed (33%) and Black (30%) high-level ethnic groups all have a higher percentage of children living in low-income households than the national average. This may be, in part, because these ethnic groups have unemployment rates higher than the national average (4%), at 8%, 7% and 9% respectively. In addition, the Black and Mixed ethnic groups were the most likely to have gross household income (the income that a household has available for spending after taxes and benefits are taken into account) of less than £400 per week.

Figure 2 shows how child poverty has changed between the three-year averages for FYE 2012, 2015 and 2018 for some specific ethnic groups. During this period, the national average increased slightly from 18% to 20%, and children in White British households were the least likely to live in low-income households.

During the same period, the largest decrease in the percentage of children living in low-income households was among children living in Indian households - a decrease of six percentage points, from 23% to 17%. This may be because of the decrease of Indian households having a weekly income of less than £400, from 33% to 25% for the same time period.

The largest increase was among children living in households from the Other White ethnic group, an increase of seven percentage points, from 16% to 23%.

Children living in Pakistani and Bangladeshi households were the most likely to live in low-income households through the period FYE 2012 to FYE 2018.

Figure 2: Child poverty has fallen most for children from Indian households in recent years

Percentage of children in households in low income, by ethnicity, UK, three-year average, FYE 2012 to FYE 2018

Data download

Children living in households in persistent low income

Children live in persistent low income if they live in a household that has been recorded as being in low income (an income below 60% of the median income, before housing costs are taken into account) in at least three out of the last four consecutive years.

The proportion of children living in persistent low income is dependent on the proportion of children in relative poverty in the current year, and the three previous years. Therefore, a reduction in the proportion of children in relative poverty in one year will not immediately translate to a fall in the proportion in persistent poverty. However, if the fall in relative child poverty is sustained, then this should gradually result in a fall in persistent child poverty.

Figure 3 indicates that 11% of children were living in households in persistent low income during the period between 2013 and 2017. Children are more likely to be in persistent low-income households if they are in a workless family (that is, no one in the family was working). During the same period, 37% of children in workless families were living in households in persistent low income.

Children in Asian households were 2.5 times as likely, compared with the national average, to be in persistent low income during this period. The high percentage of Asian children living in households in persistent low income may be driven by the Pakistani and Bangladeshi ethnic groups, who are the most likely to be living in low-income households.

The percentage of children in Black households living in persistent low income was six percentage points higher than the percentage of children in White households living in persistent low income.

Low income and material deprivation

Material deprivation is an additional way of measuring living standards, and refers to the self-reported inability of individuals or households to afford 21 particular goods and activities (PDF, 1.76MB) that are typical in society at a given point in time, irrespective of whether they would choose to have these items, even if they could afford them. This includes items such as warm coats for children and keeping accommodation warm in winter.

A child is considered to be in low income and material deprivation if they live in a household that has a final score of 25 or more out of 100 in the material deprivation questions, and equivalised household income below 70% of contemporary median income, before housing costs. In FYE 2018, 12% of children fell below the threshold of low income and material deprivation in the UK.

During this period, children in Bangladeshi households were the most likely out of all ethnic groups to live in low income and material deprivation, at 29%. This was almost three times as high as White households (10%), and was followed by 24% of children living in Pakistani households and 22% of children living in Black households.

Conversely, children in Indian households were the least likely to be in material deprivation and low income (5%) out of all ethnic groups.

To measure education, we use the "Progress 8" measure collected by the Department for Education (DfE). Progress 8 measures how much progress students make between 11 and 16 years, compared with other students with similar starting points. The starting points are calculated using assessments of Maths and English at the end of primary school (Key Stage 2 results), when children are usually 11 years old. Each pupil's progress is then worked out by comparing their "Attainment 8" score with the national average score for pupils who started at a similar level to them. Attainment 8 is a measure of a pupil's performance at the end of Key Stage 4 across eight core subjects, including Maths and English.

If a student's Progress 8 score is equal to the national average, their progress is in line with that of other students who started at a similar level. A score above the national average means the student has made more progress than other students who started at a similar level to them. A score below the national average means they have made less progress than other students who started at a similar level to them.

Group averages of Progress 8 scores can be calculated, for example, within schools or ethnic groups, in order to see how a group has progressed as a whole. In this article, we present averages within ethnic groups.

The average Progress 8 score for all ethnic groups during the academic year 2018 to 2019 was negative 0.03. Information on how this is calculated is available in section 7: About the data.

Figure 5 provides an overview of the average Progress 8 scores for each ethnic group. Chinese pupils were the highest performers, achieving an average Progress 8 score of 0.86.

The second highest Progress 8 score was achieved by Indian pupils (0.71) - the ethnic group least likely to experience both low income alone, and low income and material deprivation combined. Conversely, Bangladeshi and Pakistani pupils had higher Progress 8 scores than the national average, despite tending to have higher prevalence of poverty than other ethnic groups.

Traveller of Irish Heritage and Gypsy/Roma pupils made the least progress between 11 and 16 years, achieving scores of negative 1.05 and negative 0.81 respectively. Some caution should be taken when interpreting these percentages as they are based on a small number of pupils. For example, there were 141 pupils from Traveller of Irish Heritage backgrounds in the academic year 2018 to 2019.

Evidence has suggested that poverty and social deprivation in children is linked to educational underachievement. Furthermore, the relationship is somewhat circular, with educational attainment the most influential factor (PDF, 5.97MB), surpassing child poverty, for poverty in future life stages. Given that deprivation varies for different ethnic groups, it is important to consider how educational outcomes may vary as a consequence.

To analyse the relationship between poverty and education, we use two measures of poverty that are used widely in educational analysis: free school meal (FSM) eligibility - a pupil-based measure, and the Income Deprivation Affecting Children Index (IDACI) - an area-based measure. The previously used measures of child poverty cannot be explored in the context of education but FSM and the IDACI provide reasonable proxies with which to compare average Progress 8 scores.

The percentages of each ethnic group that were eligible for FSM in academic year 2018 to 2019 were broadly similar to the proportions experiencing low income, persistent low income, and low income and material deprivation.

In the academic year 2018 to 2019, 14% of all pupils were eligible for FSM. The Chinese and Indian ethnic groups had the lowest percentages of students who were eligible for FSM, at 7%.

The highest percentages of FSM eligibility were seen in White minority groups - 56% of Traveller of Irish Heritage pupils, and 39% of Gypsy/Roma pupils were eligible for FSM. 26% of Bangladeshi and 20% of Pakistani pupils were eligible for FSM, respectively 12 and six percentage points higher than the national average.

Education outcomes by ethnicity and eligibility for free school meals

This section focuses on FSM eligibility as an indicator of poverty. FSM eligibility demonstrates child poverty at the pupil level, and has the advantages of being easily collected and familiar to schools, parents and local authorities. FSM eligibility is a good indicator of child poverty, as it captures the majority of children living in poverty, however there is a small but significant group of children who live in poverty who are not captured by FSM eligibility (PDF, 831.01KB).

Pupils eligible for FSM made less progress between 11 and 16 years than those not eligible, with average Progress 8 scores of negative 0.53 and 0.06 respectively. This was also the case within every ethnic group.

Chinese pupils achieved the highest Progress 8 scores out of all ethnic groups. When looking at the progress made by pupils eligible for FSM only, this was still the case. Among FSM-eligible pupils, those from the Chinese ethnic group made the most progress out of all ethnic groups, with an average score of 0.66. The Chinese group also had the second smallest gap between the Progress 8 scores of pupils eligible for FSM and those not eligible (0.22), although this difference is based on a very small number of pupils; 123 Chinese pupils were eligible for FSM.

When taking into account gender too, as collected and defined in the school census by the Department for Education (DfE), girls from the Chinese ethnic group who were not eligible for FSM had the highest Progress 8 score (1.01) out of all combinations of ethnic groups, FSM-eligibility status, and gender. Figure 7 shows how Progress 8 scores vary according to these three variables.

The previous exploration of child poverty indicated that Bangladeshi and Pakistani children were the most likely to live in poverty out of all ethnic groups. Bangladeshi and Pakistani pupils who were eligible for FSM progressed higher than the average, with Progress 8 scores of 0.30 and 0.03 respectively. The gap between average Progress 8 scores for FSM-eligible pupils and those not eligible was 0.24 for Bangladeshi pupils and 0.27 for Pakistani pupils, narrower than for most ethnic groups.

By contrast, the lowest progress was seen among White pupils. Among those eligible for FSM, the least progress was made by Traveller of Irish Heritage pupils, with an average score of negative 1.16. This pattern was seen in the previous analysis of ethnic groups, before accounting for FSM eligibility. When taking into account gender too, boys from the Traveller of Irish Heritage ethnic group who were eligible for FSM had the lowest Progress 8 score (negative 1.51) out of all combinations of ethnic groups, FSM-eligibility status, and gender. In addition, among pupils eligible for FSM, the biggest gap between boys and girls was seen in the Traveller of Irish Heritage pupils, where girls achieved an average Progress 8 score 0.80 points higher than boys.

The White Irish ethnic group had the biggest gap between the average Progress 8 scores of FSM-eligible pupils (negative 0.51) and those not eligible (0.23). This may be driven by gender; among pupils eligible for FSM, the second biggest gap between boys and girls was in the White Irish group, where girls scored an average of 0.77 points more than boys.

White British children, who were less likely to live in poverty than children from other ethnic groups (as indicated in the previous measures of poverty), progressed less than average if they were FSM eligible (negative 0.78). In addition, White British pupils had the second largest gap in average Progress 8 between FSM-eligible pupils and those not eligible, at 0.73 points.

Gender plays an important role in the progress that pupils make between 11 and 16 years. As can be seen in Figure 6, in most ethnic groups, FSM-eligible girls made more progress than boys who were not eligible, with the exception of the White Irish, White British and some Mixed ethnic groups. Among pupils eligible for free school meals, girls made more progress (negative 0.28) than boys (negative 0.77), and this was also the case within every ethnic group.

The previous analysis showed that one-third of children in Black households lived in low income, and 22% of children in Black households lived in low income and material deprivation. Among Black children who were eligible for FSM, the average Progress 8 score was negative 0.08, however when looking at specific Black ethnic groups, there is some variation. Black African pupils achieved Progress 8 scores higher than average (0.17), while the progress made by Black Caribbean and Black other students was lower than average (negative 0.54 and negative 0.23 respectively).

Figure 6: Bangladeshi and Pakistani pupils who were eligible for FSM progressed higher than the national average of all pupils

Average Progress 8 score by ethnicity, gender and eligibility for free school meals, England, academic year 2018 to 2019

Data download

Education outcomes by geography and ethnicity

It is important to consider the geographical impact on child poverty and educational outcomes, as local resources, prosperity and subsequent geographical sorting effects mean that both poverty and education can vary at low levels of geography. By looking only at national estimates of child poverty and education, the data smooth over outlying performers.

The lowest average Progress 8 score for Asian pupils (0.24) was in Yorkshire and The Humber. For the Black (0.08) and Mixed (negative 0.14) pupils, it was in the West Midlands. The lowest average Progress 8 score of all was seen in White pupils in the North East (negative 0.28). Within every ethnic group, the region with the lowest average Progress 8 score was in the North or Midlands.

When looking at the highest average Progress 8 scores, London stands out as the region with the highest average Progress 8 scores for Asian (0.62), White (0.07) and Mixed (0.1) pupils. However, the highest average Progress 8 score for Black pupils was in the North East, and Chinese pupils in the East Midlands had the highest average Progress 8 score of all ethnic groups and regions (1.22).

Nationally, during the three-year average in the financial year ending (FYE) 2018, 31% of children in England were estimated to live in relative low income (60% below the median after housing costs) but the estimate varies by region. London had the highest percentage of children living in low income after housing costs (37%), followed by the North East (35%) and the West Midlands (34%).

The lowest percentages of children living in low income were seen in the South East and South West (25%). London housing market aside, the data suggest a North-South variation in children experiencing low income that is only somewhat suggested in regional average Progress 8 scores.

The remainder of this section focuses on the average local authority scores from the Income Deprivation Affected Children Index (IDACI, 2019) as an indicator of poverty. The IDACI is a supplementary metric to the Indices of Multiple Deprivation, which estimates the proportion of children aged 0 to 15 years living in income-deprived families (as defined by receipt of various benefits or tax credits with an income below 60% of the national median before housing costs).

IDACI scores and ranks are calculated at Lower layer Super Output Area (LSOA) level, but in this article, we are using the local authority summary statistics in order to match average Progress 8 data published by local authority.

The strength of the relationship between academic progress and income deprivation (as measured by the IDACI) varies for the different ethnic groups, but is negative in direction for some. There is a moderate negative correlation (negative 0.55) between White average Progress 8 scores and the IDACI local authority average scores. This means that as a local authority's IDACI score increases, the average Progress 8 score of White pupils decreases.

Similar but weaker correlations are seen for Mixed and Asian pupils (negative 0.33 and negative 0.19 respectively). There is no correlation between Black or Chinese average Progress 8 scores and average local authority IDACI scores (0.09 and negative 0.02 respectively). Average Progress 8 scores remain consistent regardless of average local deprivation.

Many local authorities with the lowest average Progress 8 scores for different ethnic groups also had high average local authority IDACI scores, estimating that higher proportions of children live in income-deprived families.

Knowsley ranked third out of all upper tier local authorities by average IDACI score (30.3% of children estimated to live in income-deprived families) and had average Progress 8 scores of negative 0.7, negative 0.82 and negative 0.9 for Mixed, White and Black pupils respectively. This made it the local authority with the lowest average Progress 8 scores for these ethnic groups.

Similarly, Blackpool ranked second by average IDACI score (30.7% of children estimated to live in income-deprived families) and had average Progress 8 scores of negative 0.65 and negative 0.69 for White and Mixed pupils respectively.

Middlesbrough, and Manchester also have higher proportions of children estimated to live in income-deprived families, and similarly appear in the bottom 10 average Progress 8 scores for an ethnic group.

Figure 7: Many local authorities with high proportions of children living in income-deprived families also had lower average Progress 8 scores

Average Progress 8 scores, by ethnicity, England, academic year 2018 to 2019, and average Income Deprivation Affecting

Data download

Looking at estimates of child poverty and educational progress by ethnicity, as well as specific proxies for child poverty within education, there is no clear, consistent relationship between child poverty and progress from Key Stage 2 to Key Stage 4. Where there does appear to be a link, it cannot be ruled out that it may be related to variables not considered in this work, and the educational resilience of different ethnic groups when living in poverty.

Children from Pakistani and Bangladeshi households consistently appear as being more likely to live in poverty compared with other ethnic groups. They are most likely to live in low-income households, with nearly half of children in Pakistani households considered to be living in low income. When factoring in material deprivation, they continue to be the most likely out of all ethnic groups to live in poverty. Despite this, Pakistani and Bangladeshi children who are eligible for free school meals (FSM) - the proxy for poverty in educational analysis - achieve Progress 8 scores higher than average, indicating that they make more progress than pupils from other ethnic groups who started at a similar level to them.

The likelihood of experiencing poverty for children in Black households is also notable; 30% are considered to live in low-income households, and 22% live in low income and material deprivation. There is a mixed picture for educational progress for specific Black ethnic groups; Black African and Black Other pupils make more progress than peers starting at a similar level, but Black Caribbean pupils make less progress than their peers.

Children living in Indian households are some of the least likely to be living in child poverty and the likelihood has decreased over time. Indian pupils have the second largest average Progress 8 score after Chinese pupils.

Across the board, pupils who are eligible for FSM have lower average Progress 8 scores than pupils that do not, but there is variation seen between the ethnic groups. White Irish and White British pupils have the largest gaps between average Progress 8 scores for FSM-eligible students and those not eligible, while Chinese, Black African, Bangladeshi and Pakistani students have the smallest gaps.

Looking at regional estimates of children living in low income and material deprivation, there appears to be variances between the north and south of Britain in the likelihood of poverty, but this pattern is not so clearly reflected in regional average Progress 8 scores. In the North East, for example, 17% of children are estimated to live in low income and material deprivation, and White pupils achieved an average Progress 8 score of negative 0.27, the lowest score out of all combinations of ethnic groups and regions. However, the North East is also the region in which Chinese, Black and Mixed pupils had their highest average Progress 8 scores.

There is a moderate negative relationship between local authority average IDACI scores and average Progress 8 scores for White pupils, but this relationship is either weak or null for other ethnic groups.

Many local authorities with the lowest average Progress 8 scores for different ethnic groups also had high average local authority IDACI scores, most often for the Mixed and White ethnic groups.

A child is an individual under 16 years, or someone who is 16 to 19 years, and all of the following:

• not married, in a civil partnership or living with a partner
• living with their parents or a responsible adult
• in full-time education or unpaid government training

Child poverty

In the analysis of child poverty outcomes, the ethnic background reported is that of the "household reference person", who is usually the person with the highest income. This means a child living in a Black household is not necessarily from the Black ethnic group. The definition of Household reference person may differ between data sources. More details can be found in the ONS glossary.

Details about the equivalisation of households can be found in the Households Below Average Income (HBAI) Quality and Methodology Information Report (PDF, 1.76MB).

Children in low income

This uses the Households Below Average Income: 1994/95 to 2017/18, published by the Department for Work and Pensions (DWP). These statistics come from the Family Resources Survey (FRS), which is a representative survey of around 19,000 private households in the UK.

The statistics for children in low-income households are presented as three-year averages (financial years ending (FYE) 2016 to 2018).

The ethnic groups presented are broken down into the Office for National Statistics (ONS) 2011 5 broad groups, and into the following detailed groups:

• Asian: Indian, Pakistani, Bangladeshi, Chinese, Asian Other
• Black
• Mixed
• White: White British, White Other
• Other

Details about statistical uncertainty over time can be found in the Households Below Average Income (HBAI) Quality and Methodology Information Report (PDF, 1.76MB).

Children in persistent low income

This uses the Income Dynamics, 2010 to 2017 Experimental Statistics published by the DWP. These statistics come from the Understanding Society (USoc) survey, Waves 2 to 8, 2010 to 2017. In 2016 to 2017 (Wave 8), the sample for USoc was of over 35,000 individuals in the UK.

The statistics for children in persistent low-income households are presented for a four-year consecutive period, 2013 to 2017.

The ethnic groups presented are the Asian, Black, Mixed and White. Statistics for the Other ethnic group are not presented because of small sample size.

Low income and material deprivation

This uses the Family Resources Survey (FRS), financial years ending (FYE) 2016 to 2018.

The statistics for children in low income and material deprivation are presented as three-year averages (FYE 2016 to 2018).

The ethnic groups presented are broken down into the ONS 2011 5 broad groups and into the following detailed groups:

• Asian: Indian, Pakistani, Bangladeshi, Chinese, Asian Other
• Black
• Mixed
• White
• Other

A suite of questions designed to capture the material deprivation experienced by families with children has been included in the Family Resources Survey (FRS) since FYE 2005. Respondents are asked whether they have 21 goods and services, including child, adult and household items.

Together, these questions form the best discriminator between those families that are deprived and those that are not. If they do not have a good or service, they are asked whether this is because they do not want them or because they cannot afford them. More details can be found in the Households Below Average Income (HBAI) Quality and Methodology Information Report (PDF, 1.76MB).

For each question, a score of 1 indicates where an item is lacking because it cannot be afforded. If the family has the item, the item is not needed or wanted, or the question does not apply then a score of 0 is given. This score is multiplied by the relevant prevalence weight. The scores on each item are summed and then divided by the total maximum score; this results in a continuous distribution of scores ranging from 0 to 1. The scores are multiplied by 100 to make them easier to interpret. The final scores, therefore, range from 0 to 100, with any families lacking all items which other families had access to scoring 100.

Income Deprivation Affecting Children Index (IDACI)

This uses data on receipt of various benefits and tax credits, as at August 2015, as an indicator of income deprivation.

Further detail on the exact indicators used is available in the Indices of Deprivation technical report.

The IDACI measures the proportion of children within a geography that are considered to be in low-income families. A score of 0.30 equates to 30% of children.

The IDACI is calculated at Lower layer Super Output Area (LSOA) level, from which local authority summary statistics are calculated. This research uses the average score local authority summary. Further detail on the higher area summaries measures is available in the Indices of Deprivation technical report.

Education

Progress 8 statistics come from the Department for Education (DfE) Key Stage 4 performance 2019 (revised) publication.

The ethnic groups presented in the Progress 8 statistics are broken down into six broad groups and the following detailed groups:

• Asian: Indian, Pakistani, Bangladeshi, Other Asian
• Black: Black African, Black Caribbean, Other Black
• Mixed: Mixed White/Black Caribbean, Mixed White/Black African, Mixed White/Asian, Other Mixed
• White: White British, White Irish, Irish Traveller, Gypsy/Roma, Other White
• Chinese
• Other

Some of the statistics quoted here are based on very small numbers of pupils and can change a lot from year to year. For example, only 97 pupils in the Chinese ethnic group were eligible for free school meals (FSM) in the academic year 2017 to 2018, meaning that the average score for this group can be heavily influenced by one pupil either performing exceptionally, or getting a very low score.

Progress 8 measures how much progress students make between 11 and 16 years. Each pupil's progress is worked out by comparing their Attainment 8 score with the national average score for pupils who started at a similar level to them. The higher a pupil's Progress 8 score, the more progress they have made in comparison with pupils who started at a similar level. The starting level is calculated using assessments from the end of primary school, when children are usually 11 years old.

The average Progress 8 score for all ethnic groups during the academic year 2018 to 2019 was negative 0.03. This average is not an exact "0" because:

• the national average for protected characteristics (such as ethnicity) refers to "All state-funded schools", which includes special schools, while the national average for schools is taken for "All state-funded mainstream schools", which excludes special schools

• it includes adjustments of extremely negative scores

The school census collected by DfE includes a gender variable, recording a pupil's gender as either male or female. The definition of this variable within the school census does not necessarily match the definition used for collection of a gender variable elsewhere. More information on this variable and others is available in the 2018 to 2019 school census guide.

Pupils are included in the figures for free school meals (FSM) if their families claimed eligibility for FSM at the time of the annual spring school census. This definition includes all pupils who were FSM eligible, not only those who actually received free school meals.

A child may be able to get FSM if their family receives any of the following:

• Income Support
• income-based Jobseeker's Allowance
• income-related Employment and Support Allowance
• support under Part VI of the Immigration and Asylum Act 1999
• the guaranteed element of Pension Credit
• Child Tax Credit (provided the recipient is not also entitled to Working Tax Credit and has an annual gross income of no more than £16,190)
• Working Tax Credit run-on - paid for four weeks after the recipient stops qualifying for Working Tax Credit
• Universal Credit

Children who get paid these benefits directly, instead of through a parent or guardian, can also get FSM.

A child may also get FSM if their family receives any of these benefits and the child is both:

• younger than the compulsory age for starting school
• in full-time education

Vasileios Antonopoulos, Nadyne Dunkley, Arpa Radia, Lualhati Santiago, Rebecca Williams from Race Disparity Unit.