This chapter presents analysis on the performance of different systems in mathematics around the period of upper secondary education. It draws on data from the OECD’s Programme for International Student Assessment (PISA) and the OECD’s Survey of Adult Skills (PIAAC) to identify how different systems perform in supporting students’ acquisition of mathematics competence. As well as looking at the overall performance of students and adults in each country, it analyses the performance of different groups of students and adults by gender, age, educational attainment, upper secondary programme and socio-economic background.
Mathematics for Life and Work
A Comparative Perspective on Mathematics to Inform Upper Secondary Reform in England
Abstract
Performance in mathematics is comparatively lower than in reading in several English-speaking systems, but not England (United Kingdom)
Individuals in some countries, such as Australia and New Zealand, have comparatively stronger literacy scores both at 15 in PISA and as young adults in PIAAC.
In contrast, young people at 15 and between 16-24 in England have comparatively similar performance across mathematics and reading.
Mathematics is a strength of systems with strong vocational upper secondary education
A high proportion of systems with well-developed vocational education systems and high enrolment in vocational education at 15 – such as, Austria, Belgium, the Netherlands, Slovenia and Switzerland – perform better in mathematics than reading in PISA.
Numeracy skills among young adults (16-24) reflect a similar trend with some countries with strong vocational systems having comparatively stronger numeracy scores than reading scores (e.g. Austria).
Low mathematics skills among young people in England reflects low participation in education among 16-18-year-olds
Completing upper secondary education is associated with higher numeracy skills in all OECD systems; the association is particularly high in England – suggesting that upper secondary contributes significantly to young people’s numeracy skills in England.
Yet, England has comparatively low shares of young people participating in or having completed upper secondary education, which drove down numeracy skills among young people in PIAAC 2012.
English-speaking countries have some of the greatest gender gaps in mathematics
The gap in mathematics skills in favour of boys and men is consistently greater than the OECD average in many English-speaking countries, notably in Canada, England, Ireland, and Northern Ireland (UK).
Exploring mathematics performance across countries requires internationally comparative data. While there is no international survey focused on learning outcomes in mathematics in upper secondary education specifically, the PISA and PIAAC surveys provide data about mathematics for young people who are around the ages at which enrolment in upper secondary education is typical.
The OECD Programme for International Student Assessment (PISA)
PISA samples 15-year-olds which is an age group that is generally either preparing to enter upper secondary education or has just entered it (Box 3.1). Across the OECD, 15 is the most frequent age for the start of upper secondary education (see Chapter 2). Data from PISA therefore provides a snapshot of student achievement in mathematics just at the beginning of upper secondary education. Since one of the most influential factors shaping student achievement is prior learning, and this is particularly the case with mathematics where learning is cumulative, performance in mathematics as students enter this level of education provides essential information for understanding upper secondary mathematics performance. The Readers’ Guide of the OECD's Programme for International Student Assessment (PISA) 2022 Volume I provides information to help interpret of data used1.
PISA measures 15-year-olds’ ability to use their reading, mathematics and science knowledge and skills to meet real-life challenges. PISA is organised on a 3-year cycle, and each assessment has a different main domain. The most recent cycle of PISA was 2022, which the analysis in this report uses when the main domain was mathematics. The next cycle of PISA is scheduled for 2025, with science as its main domain.
Source: OECD (2023[1]), PISA 2022 Online Education Database, https://www.oecd.org/en/data/datasets/pisa-2022-database.html (accessed on 13 February 2024)
The Survey of Adult Skills (PIAAC)
PIAAC samples 16-65 year-olds (Box 3.2). This is clearly a very large cohort including adults across generations who would have experienced upper secondary education a long time ago and at various stages of reform. For these reasons, the analysis presented in this report focuses on younger age cohorts (i.e. 16-24 year‑olds) who have recently completed upper secondary education to understand how the design and delivery of this level of education in recent years is associated with numeracy skills. PIAAC participants provide information about their age and educational attainment as well as a range of other information via the survey’s background questionnaires. Data used in this report draws on the most recent cycle of PIAAC which collected data over 2012-17. New data will be published in December 2024 (after the development and publication of this report).
The PISA and PIAAC datasets sample different countries, cohorts and use different proficiency scales, meaning direct comparisons of countries’ performance across the two surveys is not possible. However, by looking across countries’ performance this report provides a perspective on how individuals’ mathematics knowledge and skills evolve from their entry into upper secondary education, after completing upper secondary education and into tertiary education and work.
PIAAC measures 16-65-year-olds’ proficiency in key information-processing skills – literacy, numeracy and problem solving in technology-rich environments – and gathers information and data on how adults use their skills at home, at work and in the wider community.
The first cycle of the Survey of Adult Skills was administered over three rounds: Round 1 in 2012, Round 2 in 2014 and Round 3 in 2017. Most (24) countries participated in Round 1 for which the data collection took place in 2011/12. A few additional countries participated in the later rounds. The data in this report is primarily from the first round in 2012.
Preparations for Cycle 2 of the assessment began in 2018. The data collection was originally planned for 2021-22, ten years after the first round of Cycle 1 but was postponed because of COVID-19. The next round of the Survey of Adult Skills was rescheduled to 2022-23 and the results will be published in December 2024.
Source: OECD (2022[2]), Survey of Adult Skills (PIAAC), https://www.oecd.org/skills/piaac/ (accessed on 20 September, 2023).
National data on mathematics performance
This report makes more limited use of national data on mathematics performance. The vast majority (34) of OECD countries have a national examination at the end of upper secondary that measures student achievement (OECD, 2023[3]). Among systems that do not have a national examination, such as Canada, individual provinces have their own approaches to certify student achievement at the end of upper secondary education, as is the case in British Columbia. National examinations are intended to be comparable, valid and reliable for a single cohort nationally, and frequently make efforts to be comparable over time. National examination data, however, is not comparable across countries.
In addition, when students were not able to sit physical examinations during COVID-19, many systems introduced estimated teacher grades and other measures to ensure that young people could continue with the next phase of their lives (OECD, 2021[4]). While these measures enabled education systems to continue to function, they were also associated with grade inflation in many systems in 2020, 2021 and into subsequent years (OECD, 2023[5]). Given the limitations in the reliability and comparability of national data on mathematics achievement in upper secondary education, this report makes limited use of these data.
Overall performance in mathematics
In 2022, 15-year-old students in Japan, Korea and Singapore performed very highly in mathematics
In 2022, 15-year-old students in England and all the focus countries and jurisdictions – Austria, British Columbia (Canada), Denmark, Ireland, New Zealand and Singapore – scored above or around the OECD average in mathematics (Figure 3.1). Fifteen-year-olds in Singapore, a non-OECD country, scored particularly highly, scoring 39 score points higher than young people in Japan, the highest performing OECD country (OECD, 2023[1]).
OECD countries and Singapore
Note: The mean score of Canada includes the students from British Columbia (Canada).
Source: OECD (2023[1]), PISA 2022 Online Education Database, https://www.oecd.org/en/data/datasets/pisa-2022-database.html (accessed on 13 February 2024).
Around one in three students across the OECD does not achieve basic mathematical competence
On average across the OECD, just over a third of 15-year-old students (31% in 2022) scored below Level 2 in mathematics (Figure 3.2). Level 2 is the level at which students begin to demonstrate the ability and initiative to use mathematics in simple real-life situations (see Box 3.3 for a description of the mathematical proficiency levels used in PISA). It is considered the minimum level of proficiency that all students should acquire by the end of secondary education; students who score below this threshold are considered to be “low-achieving students” (OECD, 2023[1]).
Among the focus systems, Singapore (8.0%) has the lowest share of 15-year-old students not acquiring basic mathematical competence. While still slightly below the OECD average (31.1%), the share of 15‑year-olds not demonstrating basic mathematical proficiency in Austria (24.9%) and New Zealand (28.8%) was higher than in England (23.3%) (OECD, 2023[1]).
Student performance in PISA is reported on a scale. To help interpret what students’ scores mean in substantive terms, the scale is divided into levels of proficiency that indicate the kinds of tasks that students within these levels are capable of completing successfully.
Note: *Percentages reflect the OECD average
Source: OECD (2023[1]), PISA 2022 Online Education Database, https://www.oecd.org/en/data/datasets/pisa-2022-database.html (accessed on 13 February 2024)
Almost one in ten students across the OECD are top performers in mathematics
In 2022, 8.7% of 15-year-olds across the OECD on average scored at Levels 5 or 6 in mathematics (Figure 3.2). Students who demonstrate performance at Levels 5 and 6 are considered “top performers” (Box 3.3). They can develop and work with models in complex situations, and work strategically using broad, well-developed thinking and reasoning skills (OECD, 2019[6]). In Singapore (40.5%), more than two in five students are considered top performers, and the contrast between the very small share of low performing and large share of high performing students in Singapore stands out. The share of high performing students in British Columbia (12.1%) and England (12.0%) is greater than Ireland (7.2%), Austria (10.3%) and New Zealand (10.3%).
OECD countries and Singapore
Source: OECD (2023[1]), PISA 2022 Online Education Database, https://www.oecd.org/en/data/datasets/pisa-2022-database.html (accessed on 13 February 2024).
How equitably are mathematical skills distributed across 15-year-old students?
On average, boys from higher socio-economic backgrounds perform higher in mathematics. While immigrant students tend to perform lower on average, this is not the case in all systems (Box 3.4).
Across most countries in the OECD, boys outperform girls
Across the OECD on average, 15-year-old boys outperform girls in mathematics (Figure 3.3). This is the case in all the focus countries and most OECD countries generally. Among the focus countries and jurisdictions, Austria, British Columbia (Canada), as well as England, have the largest gender gaps. Across all PISA-participating countries on average, the greatest gap between boys’ and girls’ performance is at the top of proficiency scale between high performing boys and girls (OECD, 2023[1]).
The gender gap in mathematics may be influenced by gendered attitudes towards the subject, learning and careers. Girls, especially high performing girls, tend to express greater anxiety, fear of failure and less self-efficacy towards mathematics than boys, negatively influencing their performance. Some research suggests that this anxiety is not necessarily related to the subject itself and more the assessment format. Studies from Spain and the United Kingdom has found girls might perform better in classroom tests, but not in national examinations and that there are few differences between boys’ and girls’ levels of anxiety over coursework (Rogers and Hallam, 2010[7]; Azmat, Calsamiglia and Iriberri, 2016[8]). Social stereotypes may also influence parents’ and girls’ expectations around future career paths and shape girls’ expectations for their performance. In 2012, parents in all PISA-participating countries were more likely to expect their sons to work in a field related to science, technology, engineering or mathematics (Encinas-Martín, 2023[9]; Brussino and McBrien, 2022[10]).
However, Finland, Slovenia, and Norway – where girls perform higher than boys show that the gender gap is neither inherent nor unavoidable. These countries might be effective in promoting shared expectations for all learners, reducing the cultural space and acceptance of views that girls are less inclined to be strong mathematicians.
OECD countries and Singapore
Note: Gender gap is equal to the performance of girls minus the performance of boys. The mean score of Canada includes the students from British Columbia (Canada).
Source: OECD (2023[1]), PISA 2022 Online Education Database, https://www.oecd.org/en/data/datasets/pisa-2022-database.html (accessed on 13 February 2024).
Students from higher socio-economic backgrounds perform higher in mathematics on average
Across the OECD, on average, a students’ socio-economic status accounts for about 15.5% of the variation in their mathematics performance. In many English-speaking countries, such as Australia, Canada, Ireland, the United Kingdom and the United States, the strength of relationship between performance and socio‑economic status is below the OECD average (OECD, 2023[1]).
Analysis of PISA 2009 mathematics scores found that the scores of students with Chinese heritage in Australia and New Zealand were closer to those of students in Shanghai than that of their peers in Australia and New Zealand (Feniger and Lefstein, 2014[11]). Similarly, analysis of 2012 PISA mathematics scores of children with east Asian parents who were born and raised in Australia found that, on average, they scored more than 600 points on the PISA test, putting them second to only Shanghai (China) among PISA‑participating countries (Jerrim, 2015[12]).
While these studies may appear to suggest that immigrant students have a specific "culture" supporting their success it should be remembered that this is not the case for all immigrant students. The cited research focuses mainly on Asian immigrants moving to Australia, Canada and United States all of which typically have relatively demanding immigration policies. Consequently, the immigrant students in these samples are likely from highly educated and well-qualified families. These families, who are voluntarily immigrating to a new a country, are also probably highly motivated and invested in their success. Finally, they are a small population which makes it difficult to draw robust conclusions from their performance.
OECD countries and Singapore
Source: OECD (2023[1]), PISA 2022 Online Education Database, https://www.oecd.org/en/data/datasets/pisa-2022-database.html (accessed on 13 February 2024).
In which countries do 15-year-old students perform well in mathematics specifically?
Among those countries with high performance, it is important to consider whether high performance in mathematics is driven by high performance in all subjects i.e., are these education systems “all-round high performers” or are they specifically high performers in mathematics? If the latter is true, it might suggest that there are specific approaches to how mathematics is provided and taught, as well as cultural and social perceptions, which make the teaching of mathematics in these countries a national strength.
Students in countries with strong vocational systems perform better in mathematics
Countries with strong vocational systems – Austria, Belgium, Denmark, the Netherlands, Slovenia and Switzerland – tend to perform significantly better in mathematics than in reading comparative to other systems (Table 3.4). In the Netherlands and Slovenia 15-year-olds score above the OECD average in mathematics, but below it in reading. One might speculate that cultural and social perceptions around the role and importance of mathematics, as well as the content of educational programmes themselves – for example, systems with strong vocational education tend to integrate more mathematical concepts across technical domains – influence stronger performance.
East Asian countries seem to be all-round-strong performers
The other group of countries with significantly strong performance in mathematics are the east Asian countries – Korea, Japan, and Singapore. All three systems are among top performers in both reading and mathematics. This suggests that, despite some prevailing views, these systems might not have specific cultural values focused on high achievement in mathematics, but rather that learning outcomes more broadly are highly valued and well-supported.
Students in some English-speaking countries are relatively less proficient in mathematics than reading, but not in England
Some of the English-speaking countries among the focus countries, Ireland and New Zealand, perform lower in mathematics than in reading, comparative to other countries. The difference in New Zealand is particularly marked - New Zealand is the ninth OECD country in literacy and the twenty second in mathematics. Other English-speaking countries, Australia, Northern Ireland, Scotland (United Kingdom) and the United States, also perform comparatively lower in mathematics. In these systems, requirements to study mathematics, course structure, participation, and social perceptions might also explain lower performance. Interestingly, given stated perceptions around mathematics not being viewed as important, or less important in England, the country’s 15-year-olds rank higher in mathematics than reading in 2022.
OECD countries and Singapore
Above OECD average
Not significantly different from the OECD average
Below OECD average
Note: Mathematics and Reading rankings order OECD countries and jurisdictions under analysis, as well as Singapore, according to their PISA scores. The countries in bold are the comparator countries in this study. Scores across the two domains must not be directly compared, but only in relative terms, i.e., by looking at the ranking or the score difference with the overall average for the respective domain.
Source: OECD (2023[1]), PISA 2022 Online Education Database, https://www.oecd.org/en/data/datasets/pisa-2022-database.html (accessed on 13 February 2024).
Overall performance in numeracy proficiency
Adults in Austria, Denmark and New Zealand have higher numeracy skills than the OECD average
Among OECD countries, adults (16-65-year-olds) in Finland, Japan and the Netherlands have the highest numeracy proficiency. Among the focus countries, adults in Austria, Denmark, New Zealand and Canada2 also perform above the OECD average. Several countries have distinctly different results in the performance of their adults (16-65-year-olds) compared to that of their 15-year-olds. While 15-year-olds in all the focus countries perform above the OECD average in mathematics in PISA, adults in Ireland and England perform below the OECD average in PIAAC3. Singapore, one of the highest performing countries in PISA, also performs below the OECD average in PIAAC (Figure 3.5).
The significant change in some countries’ position between PISA and PIAAC cannot not be understood without accounting for the significant political, social, and economic changes that have taken place during the lives of the generations covered by the PIAAC cohort. The oldest age group that participated in PIAAC began school in the 1950s and could have entered the labour market around the 1960s. Some of the focus countries, notably Ireland and Singapore – experienced major economic growth with their Gross Domestic Product (GDP) per capita more than doubling since 1990 (World Bank, 2023[13]). Their education systems also experienced rapid and significant expansions with younger generations attending school for longer. The lower average performance of adults in these countries reflects the lower levels of schooling and related numeracy skills among older generations. Other factors that are not observable from the PIAAC data, such as cognitive decline with age and types of employment that require more or less intense use of numeracy skills, may also contribute to some of these patterns (OECD, 2019[14]). Countries where participation in the education system was already comparatively high and have had more modest increases in participation since the post-war period, such as Austria and Denmark, demonstrate comparative stability across PISA and PIAAC results.
OECD countries and Singapore
Note: Chile, Greece, Israel, Lithuania, New Zealand, Singapore, Slovenia, and Türkiye have 2015 data (Round 2). Hungary, Mexico, and the United States have 2017 data (Round 3). Data is only reported for Canada as whole rather than British Columbia but data for specific provinces is not available in the PIAAC data.
Source: OECD (2012, 2015, 2018[15]), Survey of Adult Skills (PIAAC) (2012, 2015, 2018), https://www.oecd.org/skills/piaac/data/ (accessed on 14 November 2023).
Younger adults tend to demonstrate higher numeracy proficiency
Across the OECD on average, adults aged 25-35 demonstrate the highest numeracy skills (Figure 3.6) Within this age group, individuals have had the time to complete upper secondary education and generally post-secondary education, including tertiary education, training and post-secondary non-tertiary education and have started to consolidate their knowledge and skills at work. The youngest adults in the PIAAC sample – 16-24year-olds – score only slightly lower than 25-35year-olds. The slightly lower score of 16‑24year-olds likely reflects that many in this age group are still completing upper secondary and post‑secondary education. Only in England is the difference between the youngest (16-24) and oldest (55-65) age groups – 0.6 score points – insignificant. This is likely driven by the very low scores of 16-19year-olds in England who, with a score of 249.7, have one of the lowest scores for this age group across the PIAAC sample, more than ten score points below the OECD average of 261 (OECD, 2019[14]).
OECD countries and Singapore
Note: Chile, Greece, Israel, Lithuania, New Zealand, Singapore, Slovenia, and Türkiye have 2015 data (Round 2). Hungary, Mexico, and the United States have 2017 data (Round 3). Data is only reported for Canada as whole rather than British Columbia but data for specific provinces is not available in the PIAAC data.
Source: OECD (2012, 2015, 2018[15]), Survey of Adult Skills (PIAAC) (2012, 2015, 2018), https://www.oecd.org/skills/piaac/data/ (accessed on 14 November 2023).
How does completing upper secondary education influence numeracy?
The association between completing upper secondary education and numeracy proficiency differs across countries
In all countries, attaining a higher level of education is positively associated with skills (OECD, 2019[14]). Attaining upper secondary education specifically is positively associated with numeracy proficiency in all PIAAC-participating countries (Figure 3.7). Despite mathematics being optional for 16-18-year-olds in England, and just 14.9% of 16-18-year-olds studying mathematics A level in 2021/22 (see Chapter 4), completing upper secondary education in England has one of the strongest associations with numeracy proficiency for young adults. Young adults (16-24) in England who have completed this level of education having numeracy scores that are 48 points higher on average than their peers who have not completed upper secondary education. This might reflect the comparatively high bar set by mathematics programmes and qualifications at 16 in England (GCSEs) (see Chapter 6), as well as opportunities to develop and use numeracy skills across other upper secondary programmes outside mathematics A level.
In England, the association between completing upper secondary education and numeracy scores is markedly stronger for younger age groups than for older adults (45-65). This might reflect changes in education policy including patterns of subject choice and improvements in education quality since the 1970s and 1980s. Denmark, Ireland and New Zealand display similar trends.
Mean difference in numeracy proficiency between those who attained at most upper secondary education and those who attained below upper secondary education in OECD countries and Singapore
Note: Chile, Greece, Israel, Lithuania, New Zealand, Singapore, Slovenia, and Türkiye have 2015 data (Round 2). Hungary, Mexico, and the United States have 2017 data (Round 3); For the cohort of 16–24-year-old, the values reflect the difference between those in education or with at least upper secondary and those not in education and lower upper secondary. For the cohort of 16–24-year-old, there is no data concerning Korea and Singapore; Missing OECD average. Data is only reported for Canada as whole rather than British Columbia but data for specific provinces is not available in the PIAAC data.
Source: OECD (2012, 2015, 2018[15]), Survey of Adult Skills (PIAAC) (2012, 2015, 2018), https://www.oecd.org/skills/piaac/data/ (accessed on 14 November 2023).
Low numeracy among the youngest adults in England is likely driven by low participation in education and low overall attainment
In England, completing upper secondary education is strongly associated with higher numeracy scores – to a much greater extent than many other OECD countries. Yet the country’s young people have some of the lowest numeracy scores on average. This seemingly paradoxical situation is likely explained by the comparatively high share of young adults who have not completed upper secondary and are not currently enrolled it in4. In 2012, the share of 16-24year-olds not in education and not having achieved at least upper secondary education in England (14.7%) is higher than the OECD average (10.7%) and all the focus countries (Figure 3.8). One of the main factors influencing the comparatively low levels of numeracy among young adults in England is likely to be the lower levels of educational attainment and participation.5
OECD countries and Singapore
Note: Chile, Greece, Israel, Lithuania, New Zealand, Singapore, Slovenia, and Türkiye have 2015 data (Round 2). Hungary, Mexico, and the United States have 2017 data (Round 3). Data is only reported for Canada as whole rather than British Columbia but data for specific provinces is not available in the PIAAC data.
Source: OECD (2012, 2015, 2018[15]), Survey of Adult Skills (PIAAC) (2012, 2015, 2018), https://www.oecd.org/skills/piaac/data/ (accessed on 14 November 2023).
How equitably are numeracy skills demonstrated across populations?
In England, an individual’s socio-economic background is strongly associated with their numeracy proficiency
PIAAC uses the level of educational attainment achieved by a participant’s parents as the best available proxy for socio-economic background (OECD, 2016[16]). For 16–24-year-olds in England, having at least one parent with tertiary education is associated with scoring 69 points more in numeracy (Figure 3.9). The relationship between numeracy scores and socio-economic background has also increased over time in England: young people without tertiary educated parents today are likely to have lower numeracy skills than their peers, whereas older adults in the same position in the past were less likely to be disadvantaged by not having a tertiary educated parent. There might be specific reasons why this disadvantage is so pronounced in England and why it has become greater over time. Since mathematics is not required for the duration of upper secondary education, young people with tertiary educated parents may be more likely to continue studying mathematics for longer and may have greater support to do so.
While being from a more socio-economically advantaged background is associated with higher learning outcomes in all countries, in other countries such as Canada, Ireland and Singapore, particularly for younger age groups, the association is far weaker. Policies related to how young people engage with numeracy skills, including mathematics in upper secondary education may influence these patterns (see Chapters 4 and 5).
Mean difference in numeracy proficiency between those with at least one parents who has attained tertiary education and those whose neither parent has attained upper secondary in OECD countries and Singapore
Note: Chile, Greece, Israel, Lithuania, New Zealand, Singapore, Slovenia, and Türkiye have 2016 data (Round 2). Hungary, Mexico, and the United States have 2019 data (Round 3); Missing OECD average. Data is only reported for Canada as whole rather than British Columbia but data for specific provinces is not available in the PIAAC data.
Source: OECD (2012, 2015, 2018[15]), Survey of Adult Skills (PIAAC) (2012, 2015, 2018), https://www.oecd.org/skills/piaac/data/ (accessed on 14 November 2023).
English-speaking countries have a large gender gap in numeracy skills
As in PISA, men demonstrate higher numeracy skills than women (Figure 3.10). In all countries, the gender difference is smaller among younger age groups, suggesting that shifts in perceptions of gender in society, overall attitudes and stereotypes around mathematics have influenced numeracy proficiency between men and women. Numeracy skills among older adults might also be shaped by gendered patterns of employment, with men working more frequently in technical, financial and statistical professions where more numeracy skills are used.
English-speaking countries appear to have a large gender difference. The gender difference in England and Northern Ireland (United Kingdom), Ireland, Canada, New Zealand and the United States, is greater than it is across the average of the OECD countries participating in PIAAC. Some of these countries are those where the gender gap in favour of boys in also greater than the OECD average gender gap for PISA (Figure 3.3). The gender gap might be related to cultural attitudes whereby mathematics is not necessarily perceived as a subject in which everyone needs to achieve highly in order to do well in work and life. Instead, it might be viewed as a niche subject associated with more male-dominated professions.
OECD countries and Singapore
Note: Chile, Greece, Israel, Lithuania, New Zealand, Singapore, Slovenia, and Türkiye have 2016 data (Round 2). Hungary, Mexico, and the United States have 2019 data (Round 3); Gender gap is equal to the performance of girls minus the performance of boys. Data is only reported for Canada as whole rather than British Columbia but data for specific provinces is not available in the PIAAC data.
Source: OECD (2012, 2015, 2018[15]), Survey of Adult Skills (PIAAC) (2012, 2015, 2018), https://www.oecd.org/skills/piaac/data/ (accessed on 14 November 2023).
In which countries do adults have particularly strong numeracy skills?
The only countries where adults have stronger numeracy than literacy skills are those with strong vocational systems
A number of the countries in PIAAC where adults have notably stronger numeracy than literacy skills, comparative to other countries, also have well-developed vocational systems, with higher-than-average enrolment in upper secondary vocational education. Examples include Austria, the Slovak Republic and Hungary. An exception is Denmark where only 19% of 15–19-year-old upper secondary students are enrolled in vocational programmes (OECD average 37%), although mathematics is compulsory for all vocational students (OECD, 2022[17]) (see Chapter 4). The technical content of vocational programmes tends to mean that numeracy skills are integrated across course content so students in these programmes are likely to have multiple opportunities to acquire numeracy skills. In these systems, the education system is often designed to reflect the demands of the labour force where workers have more opportunities to continue building their mathematical competence throughout their career.
OECD countries and Singapore
Above OECD average
Not significantly different from the OECD average
Below OECD average
Note: Numeracy and Literacy rankings order OECD countries and jurisdictions under analysis, as well as Singapore, according to their PIAAC scores. The countries in bold are the comparator countries in this study. Scores across the two domains must not be directly compared, but only in relative terms, i.e., by looking at the ranking or the score difference with the overall average for the respective domain.
Source: OECD (2012, 2015, 2018[15]), Survey of Adult Skills (PIAAC) (2012, 2015, 2018), https://www.oecd.org/skills/piaac/data/ (accessed on 14 November 2023).
Adults in some English-speaking countries have significantly stronger literacy skills compared to numeracy skills, but not in England
Adults in some English-speaking countries – Australia, Canada, New Zealand, and the United States – all have numeracy skills that are relatively weaker than their literacy skills comparative to other countries. This might reflect cultural attitudes towards mathematics and workforce patterns where mathematics is not perceived to be essential to achieve good outcomes in life and work.
One further reason for the relatively stronger performance in literacy skills among adults might reflect the fact that almost all adults have multiple opportunities to continue building and deepening their literacy skills daily, for example by reading a range of factual and fictional texts and writing emails and other digital forms of communication in personal and professional contexts. However, as discussed in Chapter 2, while numeracy skills are essential to everyday life, individuals with low numeracy skills might avoid these activities and miss out on opportunities to deepen their numeracy skills (Skagerlund et al., 2018[18]). As in PISA for 15-year-olds, 16-24-year-olds in England do not perform weaker in numeracy skills, compared to literacy skills, comparative to other systems (Table 3.5).
How do individuals’ mathematics and numeracy proficiency develop over time?
While PISA and PIAAC cannot be directly compared, the OECD has undertaken analysis by transforming the survey scores so that they are comparable (Borgonovi et al., 2017[19]). As individuals age and complete subsequent levels of education, one would expect their skills to rise. The growth in numeracy skills as individuals age differs significantly across countries (Figure 3.11). Among the focus countries, the increase in numeracy skills between 15-year-olds and 24-year-olds (the period during which individuals typically complete upper secondary education and may also be enrolled in or have completed tertiary education) in Austria and Denmark is greater than the average.
Reasons why the growth in numeracy skills in these countries may be higher include greater exposure to mathematics and other science, technology, engineering, and mathematics (STEM) subjects, reflecting subject requirements and course design in upper secondary education (see Chapter 4). For example, in Austria, all upper secondary students enrolled in general programmes must take mathematics for four years of upper secondary education and for at least two years in Denmark. In these two countries, STEM subjects are the main area of study for over two in five vocational students (see Chapter 4). Finally, the mathematics programmes themselves, particularly in the case of Denmark, set a high standard and integrate a large degree of higher order concepts upon which mathematical skills are built – such as mathematical reasoning and problem solving (see Chapter 6).
The overall growth in numeracy skills between PISA and PIAAC is shaped by the initial starting point. In 2003, New Zealand (525) had one of the highest mathematics scores of all countries participating in PISA. In contrast, some of the countries with the greatest increase between PISA and PIAAC, such as Norway (483) and Sweden (498), had comparatively low scores in 2003. In contrast, the PISA scores for Austria (515) and Denmark (512) at 15 were already among the highest of the participating countries. The high starting point of numeracy proficiency at 15 suggests that the growth in numeracy scores in Austria and Denmark relates to a deepening of individuals’ already strong mathematics skill as they age and complete subsequent levels of education (OECD, 2004[20]).
Comparisons in mean numeracy achievement between the PISA 2003 cohort and PIAAC participants at age 24 in the OECD countries
Note: Chile, Greece, Israel, Lithuania, New Zealand, Slovenia, and Türkiye have 2016 data (Round 2). Hungary, Mexico, and the United States have 2019 data (Round 3). Data is only reported for Canada as whole rather than British Columbia but data for specific provinces is not available in the PIAAC data. To examine literacy and numeracy achievement growth between age 15 and young adulthood, analyses are conducted on synthetic cohorts, matching data from PISA and the relevant birth cohort captured in the Survey of Adult Skills (PIAAC). PISA scores have been transformed in PIAAC scores using concordance constants estimated in Borgonovi et al., 2017.
Source: OECD (2012, 2015, 2018[15]), Survey of Adult Skills (PIAAC) (2012, 2015, 2018), https://www.oecd.org/skills/piaac/data/ (accessed on 14 November 2023); OECD (2019[21]), PISA 2018 Online Education Database, https://www.oecd.org/en/data/datasets/pisa-2018-database.html (accessed on 14 November 2023).
This text below highlights key insights from this chapter’s review of mathematics performance in upper secondary education. It focuses on insights for all countries and highlights policy pointers for England specifically.
1. Mathematics is not a relative weakness among young people in England, comparative to other systems
Depictions of mathematics in popular culture as well as some research suggest that it is culturally acceptable to not be good at mathematics in England, while achievement in reading and literacy is widely viewed as essential (see Chapter 7). Yet, data on individuals’ performance among 15-year-olds in PISA, and among 16-24-year-olds in PIAAC shows that individuals’ performance in England ranks similarly relatively to other OECD countries in reading and literacy and mathematics and numeracy in 2022 and 2012.
This contrasts with the situation in several English-speaking systems where 15-year-olds are comparatively stronger in reading, with mathematics being a relative weakness in Australia, Ireland, New Zealand, Scotland (UK) and the United States. This asymmetrical performance persists into young adulthood in Australia and New Zealand. This pattern lends weight to some of anecdotes and evidence that suggests that, in some English-speaking countries, mathematics achievement is viewed less importantly than literacy and reading, potentially influencing achievement. It is possible that this view – of mathematics achievement not being critical to life success – remains present in England, yet perhaps approaches to teaching and learning are able to compensate its potential influence on actual mathematics achievement.
While mathematics is not a relative weakness in England, neither is it a relative strength. In contrast, mathematics and numeracy tend to be a strength in systems with strong upper secondary vocational systems, such as Austria, the Netherlands and Switzerland. The results of some east Asian countries – notably Japan, Korea and Singapore – shows that some countries are able to achieve strong results in both reading and mathematics and are considered all-round high achieving systems.
Policy pointers for England
Communicate that students in England and consequently the education system, especially up to 16, perform well in mathematics from a comparative perspective. This might help to address societal views that mathematics skills in England are low, and this is acceptable.
In seeking to strengthen mathematics outcomes, draw on the experiences of countries like Austria, the Netherlands and Switzerland, where the emphasis on technical skills and their widespread integration across vocational education helps to promote mathematics achievement.
2. Promoting greater completion of upper secondary education is important for improving numeracy skills among young adults in England
The PIAAC data in 2012 suggests that young people in England – 16-24-year-olds – have low numeracy skills. In contrast to other systems internationally, the numeracy skills of this age group were not statistically different to those of the oldest age group – 45-65-year-olds. While these data seem to imply that England’s education system has made little progress in developing the numeracy skills of its younger generations, analysis shows that completing upper secondary education in England actually has one of the strongest associations on increasing numeracy skills. Rather, the low numeracy skills on average of 16-24-year-olds are related to low participation and completion of upper secondary education.
Policy pointers for England
Consider how to monitor and measure completion of upper secondary education to support greater participation at this level.
Since 2015, young people in England are required to remain in education or training up to 18. Monitoring completion of upper secondary education will be important to provide data on achievements of this policy. One of the challenges for monitoring completion in England is the absence of a standard measure of completion, unlike in many other systems across the OECD where completion is frequently measured via the achievement of an upper secondary certification or qualification. In England, young people are typically expected to achieve three A levels, but this is not a standard measure of completion determined by government policy and is rather an expectation shaped by tertiary entry demands. A further complication in monitoring completion in England is the transition point at 16, when students can choose to remain in school or move to a further education college or employer for an apprenticeship.
Consider using an upper secondary certification to signal and monitor completion of upper secondary.
In contrast to England (and other systems in the United Kingdom), most OECD systems have a general certification for completion of upper secondary education. For example, this is the case across nearly all of the focus systems - Austria, British Columbia, Denmark, Ireland, New Zealand. England’s recent announcement to consider the introduction of an Advanced British Standard could help to fill this gap in England and help to promote better measurement of upper secondary completion (Department for Education, 2023[22]).
3. England has one of the largest gender gaps in favour of boys and men across OECD countries
In 2022, England had the seventh large gender gap in favour of boys in PISA and, in 2012, the third largest gender gap in favour of young men among 16-24-year-olds. In general, many other English-speaking systems also have large genders gaps. In 2022, Australia, British Columbia, Ireland, New Zealand, Northern Ireland (UK), Scotland (UK) and the United States all had gender differences above the OECD average. While boys and men perform higher than girls and women across the OECD on average, there are a few countries that defy this trend – Finland, Norway and Slovenia in 2022 – suggesting that it is not unavoidable.
Policy pointers for England
Explore how cultural perceptions might influence the comparatively large gender gap in mathematics in England (and other English-speaking countries).
One explanation for the clustering of English-speaking countries with a large gender gap is that it might be due to cultural perceptions of mathematics, with mathematics being viewed as a more masculine domain and leading to more male-dominated occupations such as engineering and finance. One of the factors influencing lower performance among girls and women might also be their perceptions and self-belief around the subject, with girls reporting lower enjoyment and self-efficacy and greater anxiety than boys (see Chapter 7).
Look at practices in countries with comparatively low gender gaps - Finland, Korea, Japan, Norway – to explore how teaching and learning might be effective in promoting more equitable views and self-beliefs of mathematics.
References
[8] Azmat, G., C. Calsamiglia and N. Iriberri (2016), “Gender differences in response to big stakes”, Journal of the European Economic Association, Vol. 14/6, pp. 1372-1400, https://doi.org/10.1111/jeea.12180.
[19] Borgonovi, F. et al. (2017), “Youth in Transition: How Do Some of The Cohorts Participating in PISA Fare in PIAAC?”, OECD Education Working Papers, No. 155, OECD Publishing, Paris, https://doi.org/10.1787/51479ec2-en.
[10] Brussino, O. and J. McBrien (2022), “Gender stereotypes in education: Policies and practices to address gender stereotyping across OECD education systems”, OECD Publishing, Paris.
[22] Department for Education (2023), A World-Class Education System: The Advanced British Standard.
[9] Encinas-Martín, M. (2023), Gender, Education and Skills: The Persistence of Gender Gaps in Education and Skills, OECD Publishing, Paris.
[11] Feniger, Y. and A. Lefstein (2014), “How not to reason with PISA data: An ironic investigation”, ournal of Education Policy, pp. 845-855.
[12] Jerrim, J. (2015), “Why do East Asian children perform so well in PISA? An investigation of Western-born children of East Asian descent”, Oxford Review of Education, Vol. 41/3, pp. 310-333.
[3] OECD (2023), Education at a Glance 2023: OECD Indicators, OECD Publishing, Paris.
[5] OECD (2023), Implementation of Ireland’s Leaving Certificate 2020-2021: Lessons from the COVID-19 Pandemic, https://doi.org/10.1787/e36a10b8-en.
[1] OECD (2023), PISA 2022 Online Education Database.
[17] OECD (2022), Education at a Glance, OECD Publishing, Paris, https://doi.org/10.1787/19991487.
[2] OECD (2022), Survey of Adult Skills (PIAAC).
[4] OECD (2021), The State of Global Education: 18 Months into the Pandemic, OECD Publishing, Paris, https://www.oecd-ilibrary.org/docserver/1a23bb23-en.pdf?expires=1647878949&id=id&accname=guest&checksum=126F26A8C954A3C5CA84D67E2A6EFD40 (accessed on 21 March 2022).
[21] OECD (2019), PISA 2018 Online Education Database, https://www.oecd.org/en/data/datasets/pisa-2018-database.html.
[6] OECD (2019), PISA 2018 Results (Volume I): What Students Know and Can Do, PISA, OECD Publishing, Paris, https://doi.org/10.1787/5f07c754-en.
[14] OECD (2019), Skills Matter: Additional Results from the Survey of Adult Skills, OECD Skills Studies, OECD Publishing, Paris, https://doi.org/10.1787/1f029d8f-en.
[16] OECD (2016), Skills Matter: Further Results from the Survey of Adult Skills, OECD Skills Studies, OECD Publishing, Paris, https://doi.org/10.1787/9789264258051-en.
[20] OECD (2004), Learning for Tomorrow’s World: First Results from PISA 2003.
[15] OECD (2012, 2015, 2018), Survey of Adult Skills (PIAAC), https://www.oecd.org/skills/piaac/data/.
[7] Rogers, L. and S. Hallam (2010), “Gender differences in perceptions of studying for the GCSE”, International Journal of Inclusive Education, Vol. 14/8, pp. 795-811, https://doi.org/10.1080/13603110902721654.
[18] Skagerlund, K. et al. (2018), “Financial literacy and the role of numeracy–How individuals’ attitude and affinity with numbers influence financial literacy”, Journal of Behavioral and Experimental Economics, Vol. 74, pp. 18-25, https://doi.org/10.1016/J.SOCEC.2018.03.004.
[13] World Bank (2023), GDP per capita (constant 2015 US$).
← 1. The Readers’ Guide or PISA 2022 Volume I can be found on the following link: https://www.oecd-ilibrary.org/sites/4cd15712-en/index.html?itemId=/content/component/4cd15712-en#.
← 2. It is not possible to separate out PIAAC data for individual Canadian provinces. PIAAC data is presented for Canada overall rather than British Columbia separately.
← 3. OECD averages in PISA and PIAAC do not refer to the same group of countries.
← 4. Since upper secondary education is almost uniformly required to access post-secondary education, this means that these adults have not completed higher levels of education either.
← 5. Data for England is from the first cycle in 2012. Since then, England raised the compulsory education age to 17 years-old in 2013, and then to 18 years-old in 2015.