House of Commons - Underachievement in Education by White Working Class Children

Underachievement in Education by White Working Class Children - Education Committee Contents

2 The extent of white working class underachievement in education

Do "white working class" children underachieve in education?

22. The two main sources of data for our inquiry are:

· national data on the performance of children known to be eligible for free school meals, taken from the gov.uk website, which provides annual information on the proportions of pupils in the early years, key stage 2 and key stage 4 reaching the relevant benchmark; and

· sample-based survey data from the Longitudinal Study of Young People in England (LSYPE), which includes a measure of socio-economic status constructed from information on parental occupations, educational qualifications, home ownership, neighbourhood deprivation and FSM entitlement. The LSYPE is managed by the Department for Education, and is based on annual interviews with a nationally representative sample of the population who were aged between 13 and 14 in 2004, with an initial cohort size of 15,700.[37]

FSM data provides information on how poorer white children fare in comparison to less-poor white children, and in comparison to poorer children of other ethnicities. LSYPE data provides a view of socio-economic status (SES) as a continuous measure and shows how the educational performance of children from different ethnic groups is affected by their SES across the spectrum.

FREE SCHOOL MEALS DATA

White British ethnicity in context

23. The proportion of children eligible for free school meals varies by ethnicity. For instance, in 2012/13 around 12.5% of white British children at the end of key stage 4 were eligible for free school meals, compared to 38.5% of Bangladeshi children and 9.7% of Indian children.Table 2: Proportion of pupils at the end of key stage 4 who are eligible for free school meals, by ethnicity (England, state-funded schools (including Academies and CTCs), 2012/13, revised data)

Number of pupils Number known to be eligible for FSM Proportion eligible for FSM

White British 438,469 54,900 12.5%

Irish 1,899 288 15.2%

Traveller of Irish heritage 137 85 62.0%

Gypsy/Roma 820 392 47.8%

Any other white background 19,265 2,761 14.3%

Mixed heritage[38] 21,611 4,560 21.1%

Indian 13,543 1,308 9.7%

Pakistani 17,778 4,976 28.0%

Bangladeshi 7,676 2,959 38.5%

Chinese 2,257 168 7.4%

Any other Asian background 7,789 1,212 15.6%

Black Caribbean 8,158 2,059 25.2%

Black African 16,201 5,439 33.6%

Any other black background 3,083 924 30.0%

Any other ethnic group 10,327 3,185 30.8%

All pupils (including those for whom ethnicity could not be obtained, refused or could not be determined) 571,334 85,182 14.9%

Source: Department for Education, GCSE and equivalent attainment by pupil characteristics: National and local authority tables, SFR 5/2014, Table 2a, 14 February 2014. Note that the numbers in 'All pupils' row will be larger than the sum of the rows above it.

Although a smaller proportion of white children are eligible for free school meals than some other ethnicities, white British children still constitute the majority (64%) of the FSM group—some 55,000 children per year.

Early years

24. Table 3 shows that the attainment "gap" between FSM and non-FSM children exists pre-school, and is already larger for white British children by the age of 5 than for other ethnicities (24 percentage points). White British is the lowest performing group at this age (other than smaller white subgroups), although their performance is not significantly different from that of Pakistani FSM children.Table 3: Proportion of pupils at the early years foundation stage achieving at least the expected standard in all 17 Early Learning Goals, by major ethnic group and free school meal eligibility (England, all types of schools or early education providers that deliver the EYFSP to children in receipt of a government funded place, 2013, final data)

% Pupils known to be eligible for FSM who achieve the benchmark % All other pupils (those not eligible for FSM and for whom eligibility could not be determined) who achieve the benchmark Gap (percentage points)

White British 32% 56% 24

Irish 36% 59% 23

Traveller of Irish heritage 13% 31% 18

Gypsy/Roma 11% 18% 7

Any other white background 31% 40% 9

Mixed heritage 38% 55% 17

Indian 37% 53% 16

Pakistani 30% 38% 8

Bangladeshi 37% 42% 5

Chinese 33% 47% 14

Any other Asian background 34% 46% 12

Black Caribbean 39% 50% 11

Black African 40% 51% 11

Any other black background 41% 49% 8

Any other ethnic group (including not obtained) 37% 45% 8

Source: Department for Education, EYFSP attainment by pupil characteristics: 2013, SFR47/2013, National and local authority tables, Table 2a, 21 November 2013

Key stage 2

25. A similar pattern is seen at key stage 2. The FSM gap is larger for white British children than other major groups—only the smaller white subgroups and "any other" groupings have a larger FSM gap or a lower FSM performance.Table 4: Proportion of pupils in key stage 2 achieving level 4 or above in reading, writing and mathematics, by ethnicity and free school meal eligibility (England, state-funded schools (including academies and CTCs), 2013, revised data)

% Pupils known to be eligible for FSM who achieve the benchmark % All other pupils (those not eligible for FSM and for whom eligibility could not be determined) who achieve the benchmark Gap (percentage points)[39]

White British 74% 89% 15

Irish 60% 86% 26

Traveller of Irish heritage 33% 38% 5

Gypsy/Roma 18% 28% 10

Any other white background 57% 70% 13

Mixed heritage 80% 90% 10

Indian 82% 90% 8

Pakistani 78% 83% 5

Bangladeshi 82% 86% 4

Chinese 87% 85% -2

Any other Asian background 66% 80% 14

Black Caribbean 77% 86% 9

Black African 80% 88% 8

Any other black background 63% 74% 11

Any other ethnic group (including not obtained) 65% 73% 8

Source: Department for Education, National curriculum assessments at key stage 2: 2012 to 2013, SFR 51/2013, National tables, Table 9a, 12 December 2013

Key stage 4

26. Table 5 shows that by GCSE the gap between the performance of FSM and non-FSM white British children is considerably wider, and the difference between white British FSM children and poorer children of other ethnicities is starker (other than Traveller and Gypsy/Roma children).Table 5: Proportion of pupils at the end of key stage 4 achieving five or more GCSEs at grades A*-C including English and mathematics, by ethnicity and free school meal eligibility (England, state-funded schools (including academies and CTCs), 2012/13, revised data)

% Pupils known to be eligible for FSM who achieve the benchmark % All other pupils (those not eligible for FSM and for whom eligibility could not be determined) who achieve the benchmark Gap (percentage points)

White British 32.3% 64.5% 32.2

Irish 38.5% 74.2% 35.7

Traveller of Irish heritage 12.9% 25.0% 12.1

Gypsy/Roma 9.2% 18.0% 8.8

Any other white background 43.8% 57.3% 13.5

Mixed heritage 43.9% 67.5% 23.6

Indian 61.5% 77.2% 15.7

Pakistani 46.8% 58.8% 12.0

Bangladeshi 59.2% 67.0% 7.8

Chinese 76.8% 78.2% 1.4

Any other Asian background 52.4% 66.4% 14.0

Black Caribbean 42.2% 57.0% 14.8

Black African 51.4% 66.2% 14.8

Any other black background 43.1% 59.6% 16.5

Any other ethnic group (including not obtained) 51.5% 62.7% 11.2

Source: Department for Education, GCSE and equivalent attainment by pupil characteristics: National and local authority tables, SFR 5/2014, Table 2a, 14 February 2014

Trends over time

27. As Figure 1 and Figure 2 show, the performance of white British children eligible for free school meals has improved significantly in the last seven years, but the "FSM gap" for white children has barely changed. While the proportion of white British FSM children achieving the key stage 4 benchmark has almost doubled over the last seven years, it is still the case that around twice the proportion of non-FSM white British children succeed by this measure.

28. White British FSM children have consistently been the lowest performing group during 2006/07-2012/13, with a FSM/non-FSM performance gap that is larger than others.Figure 1: Trends in the proportion of FSM-eligible children achieving the key stage 4 benchmark, selected ethnicities, 2006/07-2012/13

Source: 2006/07-2009/10: Department for Education, GCSE and equivalent attainment by pupil characteristics in England: 2010 to 2011: National and local authority tables, SFR 3/2012, Table 2a, 9 February 2011

Source: 2009/10-2012/13: Department for Education, GCSE and equivalent attainment by pupil characteristics: National and local authority tables, SFR 5/2014, Table 2a, 14 February 2014

Figures for 2006/07-2011/12 are based on final data, figures for 2012/13 are based on revised data.Figure 2: Trends in the gap (percentage points) between the proportion of FSM and non-FSM children achieving the key stage 4 benchmark, selected ethnicities, 2006/07-2012/13

Source: 2009/10-2012/13: Department for Education, GCSE and equivalent attainment by pupil characteristics: National and local authority tables, SFR 5/2014, Table 2a, 14 February 2014

Figures for 2006/07-2011/12 are based on final data, figures for 2012/13 are based on revised data.

29. The data shows that the performance of Bangladeshi children eligible for free school meals has improved by 22.8 percentage points between 2006/07 and 2012/13, compared to only 14.9 percentage points for white British FSM children. Similarly, the FSM performance gap for Indian children has closed by 6.8 percentage points over this period, whereas for white British children it has hardly altered. Thus, while the performance of poorer children is improving for all ethnic groups, for some ethnic minorities within those groups it is improving faster than for white British pupils.[40]

30. Overall, the evidence from analysing free school meals (FSM) data is that:

· white British children eligible for FSM are consistently the lowest performing ethnic group of children from low income households, at all ages (other than small subgroups of white children);

· the attainment "gap" between those children eligible for free school meals and the remainder is wider for white British and Irish children than for other ethnic groups; and

· this gap widens as children get older.

THE LONGITUDINAL STUDY OF YOUNG PEOPLE IN ENGLAND (LSYPE)

31. Professor Strand's evidence to our inquiry drew on LSYPE data to demonstrate that a broader measure of socio-economic status (SES) presented similar conclusions to the FSM data, albeit with almost no distinction between white British children from low SES backgrounds and low-SES black Caribbean children. Figure 3 below shows that the steepness of the "SES gradient"—the extent to which SES has an impact on attainment—is greater for white British children than for other groups, and is similar for boys and girls. This reinforces the message from the "FSM gap" for white British children referred to above.Figure 3: Normalised mean GCSE points score by ethnicity, gender and socio-economic status (LSYPE dataset)

Source: Professor Steve Strand (WWC 4) Figure 2, p 2. Notes: (1). The outcome (total points score) is a measure of achievement based on all examinations completed by the young person at age 16, and is expressed on a scale where 0 is the mean (average) score for all Young People at age 16 and two-thirds of young people score between -1 and 1. (2). The SES measure also has a mean (average) of zero and the effects for low SES are estimated at -1SD and of high SES at +1SD. See Strand, S., "Ethnicity, gender, social class and achievement gaps at age 16: intersectionality and 'getting it' for the white working class", Research Papers in Education, Vol 29 Issue 2, 2014 for full details.

THE GENERAL LINK BETWEEN ECONOMIC DEPRIVATION AND EDUCATIONAL ACHIEVEMENT

32. Loic Menzies (Director, LKMco) argued that the link between economic deprivation and educational achievement applied at all levels of poverty, not just between the two groups that FSM data identifies: "[...] we have got a continuous spectrum. If you do these things by IDACI, then you see a continuous line, so I am not sure it is actually a very good idea to divide it and chop it at a particular point".[41] The Income Deprivation Affecting Children Index (IDACI) provides a more continuous measure of deprivation. The graph below plots IDACI scores for children (grouped in deciles) against their GCSE attainment measured in terms of their mean 'Best 8' points scores.[42]Figure 4: The relationship between GCSE performance (mean best 8 points) and deprivation (IDACI decile) for various ethnicities

Source: National Pupil Database 2013

33. Figure 4 confirms that the link between wealth and educational achievement exists at all levels of income—not just for the most economically deprived. As with the LSYPE data, it also shows that the "deprivation gradient"—the steepness of the line in the graph—is greater for white British students than for others; this supports what FSM data says about the effects of income appearing to be greater for this group than for other ethnicities.

34. Measures of economic deprivation and socio-economic status both suggest that white "working class" children are underachieving, and that the performance of some other ethnic groups is improving faster. But they also show that similar problems persist in a number of other minority groups.

35. Some other ethnic groups appear to be more resilient than white British children to the effects of poverty, deprivation and low-socio-economic status on educational achievement. Further work is needed to understand why this is the case. The Government should commission a project to assess why some ethnic groups are improving faster than white British children, and what can be learned from steps taken specifically to improve the achievement of ethnic minorities. This research should include, but not be limited to, the effects of historic funding and strategies, parental expectations, community resilience and access to good schools.

GENDER

36. Sir Michael Wilshaw's Unseen children speech noted that the problem of white FSM children underachieving in education was not limited to boys:

Free school meals data supports this view. Although white FSM-eligible boys are the lowest performing group overall in terms of the proportion achieving the key stage 4 benchmark, white FSM girls are the lowest-achieving group of girls. Moreover, Table 6 shows that the FSM gap for white children is slightly bigger for girls than it is for boys. Dr John Jerrim (Lecturer in Economics and Social Statistics, Institute of Education) told us that:

Professor Gillborn went further: "It would be very dangerous to slip into a situation where we are only looking at one gender and one ethnicity".[45]Table 6: Proportion of pupils at the end of key stage 4 achieving five or more GCSEs at grades A*-C including English and mathematics, by ethnicity, gender and free school meal eligibility (England, state-funded schools (including Academies and CTCs), 2012/13, revised data)

% Pupils known to be eligible for FSM who achieve the benchmark % All other pupils (those not eligible for FSM and for whom eligibility could not be determined) who achieve the benchmark Gap (percentage points)

White boys 28.3% 59.1% 30.8

Mixed race boys 39.5% 62.7% 23.2

Asian boys 48.6% 62.4% 13.8

Black boys 43.1% 57.2% 14.1

Chinese boys 74.1% 74.2% 0.1

White girls 37.1% 69.5% 32.4

Mixed race girls 48.2% 72.3% 24.1

Asian girls 57.2% 72.8% 15.6

Black girls 53.3% 67.7% 14.4

Chinese girls 79.5% 82.4% 2.9

Source: Department for Education, GCSE and equivalent attainment by pupil characteristics: National and local authority tables, SFR 5/2014, Table 2a, 14 February 2014

37. The problem of white "working class" underachievement is not specific to boys; attention to both sexes is needed.

DATA QUALITY AND AVAILABILITY

38. Statistical First Releases from the Department for Education readily allow for the analysis of FSM data by ethnicity in terms of achievement in early years, key stage 2 and key stage 4. Unfortunately figures for white FSM children for other relevant measures, such as absences and exclusions, and even key stage 5 results, are not routinely published. We have obtained some additional figures through requests to the Department for Education, but it is clear that analysis of combinations of ethnicity and FSM eligibility are not consistently available online.

39. Some witnesses were keen for better information to be collected to support analysis by social class, beyond FSM eligibility.[46] Others were more wary of the practicality and reliability of collecting information on parental occupations or other class indicators. Dr Demie cautioned that:

Dr Jerrim argued that it should be possible to join up educational performance data with information held by other government departments:

We asked the Minister for Schools (David Laws MP) about the sharing of data between Departments—he told us that some sharing can be done on an ad hoc basis at the moment, but that to do it effectively legislation would be required. He added that it would be "very sensible" for a future Government to look at this issue.[50] There are obvious issues here relating to data privacy.

40. Data relating to combinations of ethnicity and free school meals status is not always readily available in Government statistical releases. The Government should ensure that data relating to white FSM children is included in its statistical reports.

41. The Government should consider how data from a range of Departments can be combined in future to develop a more rounded indicator of a child's socio-economic status than FSM eligibility alone can provide for the purposes of targeting intervention.

42. We also heard that there could be problems with transmission of existing information between institutions. The Association of Colleges told us that "Colleges do not routinely receive data from local authorities on school pupils who were in receipt of free school meals".[51] Matthew Coffey (Director of Learning and Skills, Ofsted) told us that he had written to the Minister, Matthew Hancock, about this issue, and Sir Michael Wilshaw commented that it should be schools be expected to deliver this information as there was currently a reliance on goodwill.[52] In response, the Minister noted that Colleges do hold deprivation-related data through their distribution of the bursary, but that further action could be taken to strengthen the transfer of data between schools and colleges.[53]

43. The Government should act to ensure that FSM data (and any future revised indicator) is made available to post-16 institutions to allow effective monitoring of the progress of this group of young people.

REGIONAL VARIATION

44. The Department for Education's written evidence revealed a significant variation in the performance of white FSM pupils by local authority. Extreme examples included Peterborough, where the proportion of white FSM pupils reaching the key stage 4 benchmark was less than 13% in 2012, and Lambeth, where the equivalent figure was almost 50%.[54] Other notable geographical variations included:

· white FSM children perform unusually well in London, both in affluent areas such as Kensington & Chelsea and Westminster, and in poorer areas such as Lambeth, Hackney and Wandsworth. These areas also have the smallest gaps between white FSM pupils and other FSM pupils, and between white FSM and all other children;

· white FSM children perform poorly in a range of areas, including in cities (Nottingham), coastal areas (Isle of Wight, Southend-on-Sea) and rural areas (Herefordshire);

· there are a small number of areas where white FSM pupils outperform other FSM pupils at KS4, including Sefton, Gateshead and Wakefield, but in the overwhelming majority of cases the reverse is true—most noticeably in North Lincolnshire.[55]

45. Figure 5 shows how the proportion of FSM children achieving five good GCSEs (including English and mathematics) varies by ethnicity at a regional level. White FSM children are the lowest performing group in all regions other than the South West, where they perform slightly better than Black FSM pupils (although the Black FSM population is very small at 152 pupils at the end of key stage 4 in 2012/13).Figure 5: Regional variation in the proportion of FSM children achieving the key stage 4 benchmark, by ethnicity (2012/13, revised data, England, not including pupils recently arrived from overseas)

Source: Department for Education (WWC 42). Data relating to Chinese FSM students has been suppressed in some regions due to small populations.

Will school improvement alone close the gap?

46. Professor Strand told us that:

This view is supported by analysis in the IPPR report A Long Division, which noted that "Even if every school in the country was outstanding there would still be a substantial difference in performance between rich and poor children".[57] Ofsted data confirms that the FSM 'gap' exists in outstanding schools as well as inadequate schools.Figure 6: Percentage of pupils eligible for free school meals attaining five GCSEs at grades A* to C including English and mathematics, by school overall effectiveness judgementSource: Ofsted, Unseen Children, Figure 19 (based on open secondary schools with a published Section 5 inspection report at 31 December 2012)

47. Figure 6 shows that there is a significant difference between the performance of inadequate and outstanding schools for FSM children. Twice the proportion of poor children attending an outstanding school will leave with five good GCSEs when compared with the lowest rated schools, whereas the proportion of non-FSM children achieving this benchmark in outstanding schools is only 1.5 times greater than in those rated as inadequate. This reinforces the message from our 2012 report on great teachers that "raising the quality of teaching yet higher will have profound consequences for pupils' attainment and progress, and subsequently for their adult lives and the contributions they make to society".[58] A good school and good teaching can have a significantly positive effect on the educational attainment of FSM children, which underlines the central importance of raising school and leadership quality alongside closing the attainment gap.

37 "Welcome to interactive LSYPE", Department for Education Back

38 Includes white and black Caribbean, white and black African, white and Asian, and Any other mixed background. Back

39 The table suggests that Chinese FSM students outperform their non-FSM counterparts, but it should be noted that only 144 Chinese pupils were eligible for free school meals that year. Back

40 Strand, S., De Coulon, A., Meschi, E., Vorhaus, J., Ivins, C., Small, L., Sood, A., Gervais, M.C. & Rehman, H., Drivers and challenges in raising the achievement of pupils from Bangladeshi, Somali and Turkish backgrounds (2010) Research Report DCSF-RR226. London: Department for Children, Schools and Families Back

41 Q83 Back

42 The "Best 8" point score is based on listing each pupils' qualifications in descending order of point score, and summing these points for the top eight GCSEs or equivalents. Back

43 Ofsted, Unseen children - HMCI speech (June 2013), p 4 Back

44 Q35 Back

45 Q36 Back

46 Q13 Back

47 Q13 [Dr Demie] Back

48 Q14 [Dr Demie] Back