The Long term cost of numeracy difficulties – Every Child a Chance Trust & KPMG (2008)
This report was commissioned by the Every Child a Chance Trust and the context for the report was the Every Child Counts initiative, providing expert numeracy teaching to seven-year-old children who have failed to master basic numeracy skills. The aim was to:
- review the research of long term consequences of numeracy difficulties for individuals and society
- estimate the public costs that result
- estimate the return on investment of early intervention to address numeracy difficulties
KPMG estimates the costs to the public purse arising from failure to master basic numeracy skills are up to £2.4 billion every year.
Costs related to individuals with numeracy difficulties only (not co-occurring with literacy difficulties) are estimated at up to £763 million each year.
In KPMG’s judgement the overall return on investment for every pound spent on the Every Child Counts programme, avoiding double counting in relation to children with co-existing literacy difficulties, is likely to be between £12 and £19.
Numeracy difficulties in the adult population
Numeracy difficulties are very common in the adult population. In the government's 2003 Skills for Life survey more than 8,000 adults across England had their maths skills tested. From this data 15 million adults were estimated to have numeracy skills at or below entry level 3 - equivalent to the skills expected of an 11-year-old. Of these, 6.8 million had skills at or below entry level 2, the standard expected for a nine-year-old.
A Basic Skills Agency study in 1997 found that 74% of 37-year-olds have problems with division, 57% with subtraction, 15% could not manage their household accounts and 8% could only manage their household accounts with difficulty
An on-line survey of 2006 adults aged 18 and above, conducted in 2008 by YouGov for the Every Child a Chance Trust , found that more than one in four adults admitted they had difficulties with mental arithmetic, and over a quarter sometimes struggled to add up prices in their heads when shopping.
Nearly half (47%) wished they had learnt more maths at school and just over half (51%) of mothers said that they struggled to answer mathematical questions which their children ask them.
Women were much less confident (or perhaps more honest) than men - 34% said they had trouble working out sums in their heads, compared to 18% of men.
The survey found that difficulties with maths spread across social classes and all ages. 3% of ABC1s and 4% of C2DEs said they had to struggle with mental arithmetic in the shops most of the time. One third of the lower social groups (33%) said they felt uncomfortable in shops some of the time as compared to 25% of the top social groups. The over 55s were the most confident (77%) compared to 65% of 25-34 year-olds, who were the least confident.
Difficulties from the start
Numeracy failure starts early and becomes entrenched if not tackled. Data from longitudinal studies conducted by the Centre for Research on the Wider Benefits of Learning shows that those who are very low attainers at 7 tend to remain so at 11, more so in mathematics than in literacy.
Predictably, children from socially advantaged homes do better at 7 and 11; however, for children of parents with lower levels of education, doing well at 7, particularly in maths, is more important (i.e. more predictive of later attainment than for other groups).
The researchers conclude that ‘an emphasis on basic numeracy skills may particularly benefit children from socially disadvantaged backgrounds’.
Consequences for school children
Children leaving Key Stage 1 at seven without having mastered the most basic numeracy skills will in almost all cases be identified by their primary school as having special educational needs and be placed on the ‘School Action’ or ‘School Action Plus’ stages of the national SEN Code of Practice. By the age of 11, 34% of children with very poor numeracy skills will have Statements of special educational needs.
There is a significant link between poor numeracy and antisocial behaviour. A number of researchers have demonstrated the increased incidence of numeracy problems in children with social, emotional and behavioural difficulties.
An analysis by the Department for Children, Schools and Families of all pupils permanently excluded in Year 9 in the 2004-5 academic year data showed that pupils who entered secondary school with very low numeracy skills but good literacy skills (below National Curriculum (NC) Level 3 in Maths but not English) had an exclusion rate twice that of pupils entering Key Stage 3 at NC Level 4 or above in Maths, i.e. those with low numeracy skills but with average literacy skills were twice as likely to be excluded from school as those whose maths skills matched their literary achievements.
Truancy is also linked to poor attainment in maths. The DCSF analysis of Year 9 pupils showed that those entering secondary schools with poor maths skills are over twice as likely to truant as are those who enter with age-appropriate skills.
Recent surveys show widespread concern amongst employers about their employees’ basic skills. The CBI’s 2008 audit surveyed 735 firms employing 1.7 million people between them. Over half said they were concerned that they will not be able to find enough skilled people with the right qualifications in future. Around a quarter were investing in remedial literacy and numeracy training.
Although there is a clear correlation between literacy and employment, for numeracy and employment the correlation is even stronger, with nearly four out of ten economically inactive adults having very poor numeracy skills (adult national qualification Entry Level 2 or below).
Adults with adult qualification Level 1 numeracy or above (the equivalent of National Curriculum Level 3+ in school-age children) earn on average 26% more than adults with skills below this level. When controlling for education level, social class, parental interest in the child’s education and type of school attended, there is still a 10% earnings premium for numeracy.
Researchers at the London School of Economics have noted that the raw wage premium from having adequate numeracy skills (Level 1 or above in the adult skills framework) is actually greater now than it was in the early nineties. They conclude ‘the increase in the supply of literacy and numeracy skills since the early 1990s has been at least matched by the increase in demand for these skills, causing the return to these skills to remain stable’.
In their 1997 study of the impact of poor basic skills on 37-year-olds in the NCDS cohort, Bynner and Parsons, looking only at those who left school at 16 so as to control for the influence of extended education and resulting qualifications, found that by the age of 37 women with very low numeracy had on average 8.4 years in full-time employment compared to 10.6 years for those with average numeracy skills and 12.8 years for those with good skills. Men with very low numeracy had 15.5 years of employment compared to 18.9 years for those with average numeracy skills and 19 years for those with good skills.
In the same study, 19% of men with very low numeracy skills were unemployed or sick, compared to 3% of those with good numeracy skills. Those with poor numeracy skills at 37 had entered jobs earlier, at a time when those with better skills were still in education. By the age of 23, however, those with poor numeracy began to part company with the more skilled groups, and were more likely to be unemployed. The gap had widened with age, and at 37 those with very low numeracy skills were much less likely to be employed than those with good or average skills.
People with very low numeracy skills, compared to those with good skills, were two to three times more likely to be living in a household where both partners were out of paid employment.
Those with poor numeracy skills were much less likely to have ever received work-related training. 61% of women with very low numeracy had never been promoted at work compared to 34% of those with good numeracy. For men, the figures were 50% and 29%.
Looking internationally, a Canadian study found that numeracy is ‘generally a statistically significant determinant of labour market status, whilst literacy is most often not statistically significant.’
Grinyer found that poor numeracy, unlike poor literacy, is not associated with poor health. Other analyses, however, have found that having adult qualification Level 1 numeracy skills or above reduces the probability of having long term health problems by between six and nine percentage points, even allowing for educational levels and family background.
Adult Level 1 numeracy skills or above are for men associated with a 6% to 10% lower probability of being depressed. For women there is a 2% to 5% lower probability.
Bynner and Parsons asked 37 year olds in the NCDS study to report on their physical and psychological health over the last year. 11% of men with very low numeracy skills were classified as depressed, compared to 7% of those with average numeracy skills. Amongst women, 18% of those with very poor numeracy were depressed compared to 8% of those with average skills.
Adults with numeracy difficulties are much more likely to live in disadvantaged housing conditions than those with Level 1 or above skills. 10% of women with numeracy difficulties have experienced a period of homelessness, compared to 5% of all women. Women with very poor numeracy are more than twice as likely as women with Level 1 or higher skills to have been a teenage mother and three times more likely to have 4+ children at age 34.
The initial Basic Skills Agency assessment offered to all prisoners at the start of custodial sentences indicates that over two-thirds (65%) of prisoners have difficulties with maths (at or below Level 1 in the adult national qualifications framework). The figure for literacy difficulties is lower, at 48%. 25% of juveniles in custody have a numeracy age below that of the average seven-year-old.
Parsons, using data from the British Cohort Study (BCS 70) and National Child Development Study found that for women, poor numeracy skills were significantly correlated with criminality even after controlling for social disadvantage, poverty, disruptive family environment, poor education experiences and early signs of emotional and behaviour problems.
For men the picture was less extreme although the link between poor numeracy and the number of times they had been arrested, however, was still statistically significant when family background and childhood poverty measures were held constant.
For women, poor numeracy was an independent predictor of:
- poor physical health;
- a belief that they lacked control over their lives;
- the probability of being out of the labour market (regardless of how many children they had)
- or, if in work, of being in an unskilled or semi-skilled job;
- the probability of living in a household where no-one works.
For men, poor numeracy, even when their literacy was good, indicated:
- less likely to be in a company pension scheme
- increased risk of depression;
- increased probability of having been suspended from school, or arrested and cautioned by the police
- The Employment Equation: Why our young people need more maths for today’s jobs (2013)
- The Mathematics Skills of Schoolchildren: How does England compare to the high performing East Asian jurisdictions? (2013)
- A world-class mathematics education for all our young people (2011)
- 2011 Trends in International Maths and Science Study (TIMSS)
- A synthesis of reports related to mathematics education in England
- Society Counts: Quantitative Skills in the Social Sciences and Humanities – British Academy 2012
- Report on research into maths and science teaching in the Shanghai region (2013)
- Improving student achievement in mathematics (2000)
- Parents' Beliefs about Children's Maths Development and Children's Participation in Maths Activities (2012)
- Number and Calculation. Getting the best results