This chapter explores students’ self-beliefs, such as mathematics self-efficacy and mathematics anxiety, and how these relate to students’ attitudes and dispositions to learning. In addition, it examines the interaction between liking mathematics and feeling confident in it, and the role of teachers’ support. It also analyses the interplay between students’ uncertainty about the future and anxiety towards learning, and the impact of teachers’ attitudes towards students on students’ anxiety.
PISA 2022 Results (Volume V)
Learning Strategies and Attitudes for Life
PISA
Abstract
For Australia*, Canada*, Denmark*, Hong Kong (China)*, Ireland*, Jamaica*, Latvia*, the Netherlands*, New Zealand*, Panama*, the United Kingdom* and the United States*, caution is advised when interpreting estimates because one or more PISA sampling standards were not met (see Reader’s Guide, Annexes A2 and A4).
Students’ self-beliefs about how well they can perform on specific tasks or subjects are formed when they are young and follow them over the course of their lives (Bandura, 1997[1]). While these beliefs are partly formed based on students’ past performance, once they are formed, it is difficult to modify them. These self-beliefs will affect students’ attitudes towards learning, the development of their skills and their decisions about educational pathways and careers. Students who feel confident and do not associate anxiety with certain subjects are more likely to continue to study them and pursue a career related to them (Bong and Skaalvik, 2003[2]; Wang, Eccles and Kenny, 2013[3]).
Under the umbrella concept of self-belief is self-efficacy: this is the extent to which students believe in their ability to engage in certain activities and perform specific tasks even when facing difficulties (Bandura, 1977[4]). Students who believe they can carry out certain tasks feel less anxious and stressed about them and vice versa. Self-efficacy and anxiety are often, even if not always, concepts that represent two sides of the same coin, with one negatively affecting the other. Since mathematics often provokes students’ stress and lack of confidence, this chapter focuses on mathematics self-efficacy; that is, confidence in solving mathematical tasks or its opposite, mathematics anxiety, as measured by PISA 2022.
This chapter examines students’ mathematics self-beliefs, how these concepts vary across and within different countries and economies, and how they are associated with different strategies for sustained lifelong learning and students’ motivations.
While 15-year-olds who participated in PISA 2022 reported feeling particularly confident solving mathematical problems in the classroom, they feel less confident about those related to real-life events. This is the case especially in Viet Nam and Thailand. PISA 2022 also finds that 15-year-olds in most countries and economies (except for Korea, Singapore and Thailand) show higher levels of mathematics anxiety than they did in 2012 and that they feel anxious not only about their grades and failing in mathematics but dealing with mathematical tasks in general. Most education systems that show the lowest levels of self-efficacy, such as Argentina, Brazil, Brunei Darussalam, Cambodia, Chile, Costa Rica, the Dominican Republic, Guatemala, Malaysia, Mexico and the Philippines, also show the highest levels of mathematics anxiety.
Students who are more confident and students who are less anxious ask more questions when they do not understand something being taught. They are more proactive in their learning than students who are less confident and those who are more anxious, respectively. They are also more motivated and they especially love learning new things.
All learning strategies for sustained lifelong learning and motivations are related to students feeling more confident solving mathematical tasks and less anxious about mathematics, regardless of their performance. This is particularly true for strategies related to proactive behaviour and motivations to learn. Albania and Uzbekistan show the largest associations between self-efficacy or anxiety and most learning strategies or motivations.
Teachers can alleviate students’ anxiety about mathematics by building positive relationships with their students. This is especially true in El Salvador, Georgia, Kazakhstan and the Ukrainian regions (18 of 27), where the association between student-teacher relationships and students’ anxiety levels is particularly strong. Teachers who are extra supportive can also help students like mathematics more. While in top-performing countries, large shares of students do not necessarily consider mathematics as one of their favourite subjects, overall, there is a positive relationship between performance and having mathematics as one’s favourite subject.
Self-efficacy plays an important role in motivating students and helping them succeed. When students do not believe in their abilities to do specific tasks, they will likely not put the effort into completing them successfully. A lack of confidence in doing mathematics may impede students’ lifelong learning as they will face mathematics-related problems throughout their lives. Such a lack of confidence may also influence the choices they make as they will be less likely to choose studies or careers that require mathematical skills or are related to mathematics in any way.
Students who are more confident are more likely to set challenging goals for themselves, make more of an effort and persevere longer. On the contrary, students who experience low levels of self-efficacy need to have much higher levels of self-control to succeed as they are less likely to be motivated to learn (Klassen and Usher, 2010[5]; Schunk and Pajares, 2009[6]). Mathematics self-efficacy is positively associated with all socio-emotional skills measured in PISA 2022, even when accounting for students’ and schools’ socio-economic profile and mathematics performance (see Table V.B1.4.11). In line with the literature, a large association was especially found with persistence1, meaning that persistence could also potentially help students’ confidence in their abilities.
A lack of confidence in applying mathematical skills to real-life contexts might hinder students’ opportunities for lifelong learning
Across OECD countries, students reported feeling particularly confident solving mathematical tasks that are usually seen in the classroom.2 For example, 70% or more agreed or strongly agreed that they feel confident solving different types of equations. On the contrary, students reported feeling less confident solving mathematical problems related to real-life events; for example, less than 50% agreed or strongly agreed that they can confidently calculate the power consumption of an electronic appliance per week or find the actual distance between two places on a map (see Table V.B1.4.1). In Albania, France, Singapore and Uzbekistan, more than 60% of students reported that they can confidently solve these problems but in Thailand and Viet Nam, less than 40% did. This result is concerning for students’ lifelong learning opportunities as it suggests that students who are less confident applying their mathematical skills to real-life contexts may struggle. In other words, they will feel less confident with all the various forms of mathematics they may need as adults in daily life.
Students who reported more mathematics self-efficacy tend to have higher mathematics scores
In line with previous PISA cycles, there is a positive association in PISA 2022 between mathematics self-efficacy and mathematics performance even after accounting for students’ and schools’ socio-economic profile. Specifically, a one-unit increase in the index of mathematics self-efficacy is associated with an increase of 28 points in mathematics on average across OECD countries. This means that students who feel more confident solving mathematics tasks perform better in mathematics (see Table V.B1.4.3). This association is larger in Australia*, Korea, Macao (China), New Zealand*, Singapore and the United Kingdom*, with an increase of more than 35 score points in mathematics. On the contrary, this association is smaller in Cambodia, the Dominican Republic, Guatemala and the Philippines, with an increase lower than 10 score points.
Low-achieving students feel less confident in some top-performing countries and economies
Among countries and economies that show the widest gaps in self-efficacy between low and skilled performers are some top-performing systems. These include Australia*, Austria, Canada*, Japan, Korea, Macao (China), New Zealand*, Singapore, Sweden and Chinese Taipei (see Table V.B1.4.4) – systems that tend to show the highest levels of mathematics self-efficacy overall. Low achievers in top-performing systems might feel particularly insecure about their mathematics skills. This could also be related to the low shares of students who like mathematics in top-performing countries and economies (Box V.4.1). Given the strong association between self-efficacy and performance, it is important to help low achievers feel confident about solving mathematics tasks. This could help improve their performance.
Liking mathematics might improve students’ readiness for lifelong learning
High shares of students liking mathematics reflects positive learning experiences. This motivates students to put more effort into and feel optimistic about overcoming challenges and can have a positive impact on students’ lifelong learning attitudes towards mathematics. Across OECD countries on average, 39% of students agreed or strongly agreed that mathematics is one of their favourite subjects. Singapore and Uzbekistan have the highest shares, with, respectively, 65% and 69% of students reporting mathematics as one of their favourite subjects (see Table V.B1.4.30).
In all PISA-participating countries and economies, liking mathematics is positively associated with mathematics self-efficacy even when accounting for students’ and schools’ socio-economic profile and mathematics performance (see Table V.B1.4.39). This means that when students like mathematics they might feel more confident about their mathematics abilities. This can be important, especially for students who feel less confident and underperform in mathematics. Not surprisingly, fewer girls than boys and fewer disadvantaged than advantaged students reported that mathematics was one of their favourite subjects in almost all countries and economies. The differences are, respectively, an average of 11 and 13 percentage points across OECD countries (see Table V.B1.4.30).
Students in top-performing countries did not report mathematics to be one of their favourite subjects but skilled-performing students reported it more than their low-performing peers
While the expectation would be that top-performing countries and economies in mathematics have a larger share of students who consider mathematics to be one of their favourite subjects, there is no clear country-level association between liking mathematics and performance. In most top-performing countries and economies in mathematics, the overall share of students who consider mathematics to be one of their favourite subjects is only around the OECD average. The exceptions are the Netherlands*, which has one of the lowest shares (29%), Hong Kong (China)* and Singapore, which both have much higher shares (more than 50%). Surprisingly, other education systems with large shares of students who consider mathematics to be one of their favourite subjects do not necessarily perform well in mathematics (see Table I.B1.2.1 [Volume I – PISA 2022 Results] and Figure V.4.1).
However, PISA 2022 also shows there is a strong relationship within countries/economies between performance and liking mathematics. Specifically, students in OECD countries agreeing or strongly agreeing that mathematics is a favourite subject is associated with an increase in performance of 42 points on average, after accounting for students’ and schools’ socio-economic profile. Korea presents by far the strongest association, with an increase of 63 points in mathematics. The lowest associations (below 20 points) were found in the Dominican Republic, El Salvador, Jamaica*, the Philippines and Thailand (Table V.B1.3.15).
This result is confirmed when looking at the shares of students within countries who like mathematics and their performance. In all countries and economies, more skilled than low performers agreed or strongly agreed that mathematics is one of their favourite subjects. On average across OECD countries, 51% of skilled performers and 27% of low performers reported so (see Figure V.4.1). Some countries that are top performers in mathematics but show, overall, low shares of students liking mathematics are also characterised by larger differences between skilled and low performers reporting to like mathematics. This is the case, for example, in Denmark* and Korea. It is important, especially in these countries, to support low performers and provide them with more support in mathematics class as they might feel more pressure to perform well and discouraged in their learning.
Percentage of students agreeing or strongly agreeing that mathematics is one of their favourite subject, by level of performance in mathematics
Note: Only countries and economies with available data are shown.
Countries and economies are ranked in descending order of the percentage-point difference relative to mathematics performance (skilled performers minus low performers).
Source: OECD, PISA 2022 Database, Table V.B1.4.30. See Table V.4.1 for StatLink at the end of this chapter.
Teacher support can help increase students’ chances of liking mathematics
Liking mathematics is also influenced by other factors, notably curricula and pedagogy. Singapore, for example, emphasises mathematical reasoning and problem-solving in mathematics curricula and allows students to choose among different options and levels: this may be why more students say that mathematics is one of their favourite subjects (Box V.4.2). Additionally, teacher support can play a key role in helping students enjoy learning.
PISA 2022 Volume II shows how students with supportive teachers suffer less from anxiety and perform better. PISA 2022 results also show that students with supportive teachers are more likely to agree or strongly agree that mathematics is one of their favourite subjects (see Table V.B1.4.41) in almost all countries and economies. This is particularly salient as there is discussion that teacher support has deteriorated since 2012 in many countries and economies. Countries with low teacher support, such as Austria, Croatia, Czechia, Hungary, the Netherlands*, Poland and Slovenia, also show low shares of students liking mathematics. On the contrary, countries with highly supportive teachers such as Albania, Kazakhstan and Saudi Arabia have high shares of students liking mathematics (see Tables V.B1.4.30 and V.B1.4.37).
Source: OECD (2023[7]), PISA 2022 Results (Volume II): Learning During – and From – Disruption, https://doi.org/10.1787/a97db61c-en; OECD (Forthcoming[8]), Mathematics for Life and Work, A Comparative Perspective on Mathematics to Inform Upper Secondary Reform in England.
Students who are more confident ask more questions and apply more proactive study behaviours than those who are less confident
Students who are more confident in mathematics reported applying different types of learning strategies more frequently (see Table V.B1.4.7). They especially reported using more strategies related to controlling their own learning, such as asking questions when not understanding the mathematics material being taught. This not only requires confidence but is essential to improving one’s understanding and learning outcomes. In Albania, Baku (Azerbaijan), Iceland, Korea, Mongolia and Norway, the difference between the shares of confident and less confident students reporting that they asked questions is more than 40 percentage points compared to an OECD average of 29 percentage points. In Costa Rica and Guatemala this difference is below 20 percentage points.
Students who are more confident also demonstrate more use of proactive study behaviours, which includes things like connecting new material with previous lessons. This is another key association for lifelong learning: confident students tend to trust their knowledge more and might find it easier to build on it with new information. While this difference is an average of 33 percentage points across OECD countries, it is a difference of more than 50 percentage points in Albania and Korea. In Austria, Belgium and Japan, it is less than 20 percentage points.
Learning strategies, especially those related to proactive behaviours and problem-solving, relate positively to mathematics self-efficacy
Figure V.4.2 shows that mathematics self-efficacy is positively associated with all learning strategies even when accounting for students’ and schools’ socio-economic profile and students’ mathematics performance. While we cannot attribute causality from these analyses, one way to interpret these results is that when students use specific learning strategies and teachers encourage them to use them, they might become more confident solving mathematics tasks for the same level of performance. Learning strategies that show the largest associations are those that require proactive learning behaviours such as connecting new and prior knowledge and those related to problem-solving such as finding new ways to solve problems. Students asking questions when they do not understand something has a strong association with mathematics self-efficacy. This suggests that not only are these students more likely to ask questions, but they are more confident. Asking questions helps students become more confident in mathematics tasks.
Interestingly, some education systems show larger associations between mathematics self-efficacy and all or most learning strategies. This is the case, for example, in Albania, Baku (Azerbaijan), the Dominican Republic, Israel, Jordan, North Macedonia, Saudi Arabia and Uzbekistan. These associations are smaller in Belgium, Brunei Darussalam, Italy, Singapore, Spain, Switzerland and Viet Nam (see Table V.B1.4.9).
Change in mathematics self-efficacy when students reported the following learning strategies; OECD average
1. The socio-economic profile is measured by the PISA index of economic, social and cultural status (ESCS).
2. Students doing the corresponding statement more than half of the time.
3. Students agree or strongly agree with the corresponding statement.
4. Students doing the corresponding statement more than half of the lessons.
Note: All differences are statistically significant (see Annex A3).
Items are ranked by kind of learning strategies and then in descending order of the change in mathematics self-efficacy index, after accounting for students' and schools' socio-economic profile and performance in mathematics.
Source: OECD, PISA 2022 Database, Table V.B1.4.9. See Table V.4.1 for StatLink at the end of this chapter.
Students who are more confident feel more motivated and enjoy learning more than those who are less confident
Students who are more confident in their capacity to solve mathematics problems also reported feeling more motivated (see Figure V.4.3). Responses to “I like schoolwork that is challenging” and “I love learning new things in school” revealed the biggest difference between confident and less confident students. This suggests that students who are more confident enjoy learning and challenging themselves more than those less so.
Australia*, Canada*, Denmark*, Finland, Hong Kong (China)*, Ireland*, Macao (China), New Zealand*, Norway and the United Kingdom* show wider differences between the shares of confident and less confident students liking challenging schoolwork. These differences range between 40 and 48 percentage points compared to 30 percentage points on average across OECD countries. On the contrary, Albania, Brunei Darussalam, Colombia, Georgia, Serbia, the Slovak Republic, Slovenia and Uruguay show smaller differences between confident and less confident students, ranging between 16 and 20 percentage points (see Figure V.4.3b [available online] and Table V.B1.4.8).
Percentage of students agreeing or strongly agreeing with the following motivations, by the bottom and top quarters of the index of mathematics self-efficacy; OECD average
Notes: Students who are less confident (more confident) are those in the bottom (top) quarter of the index of mathematics self-efficacy in their own country/economy.
Differences between students in top and those in bottom quarters of the index of mathematics self-efficacy are all statistically significant (see Annex A3).
Items are ranked in descending order of the percentage-point difference related to confidence in mathematics (more confident minus less confident students).
Source: OECD, PISA 2022 Database, Table V.B1.4.8. See Table V.4.1 for StatLink at the end of this chapter.
Intrinsic motivations have an especially positive relationship with confidence
Mathematics self-efficacy is positively associated with all types of motivation even after accounting for students’ and schools’ socio-economic profile, and mathematics performance (see Table V.B1.4.10). But the largest association is between mathematics self-efficacy and liking schoolwork that is challenging. When students are motivated to learn and challenge themselves, they are more likely to have confidence in their abilities to succeed. Hong Kong (China)* and Uzbekistan show the largest associations between self-efficacy and liking schoolwork that is challenging. Students who like challenging schoolwork in these countries and economies are particularly likely to feel confident about their capacity to solve mathematics tasks.
While students who are less confident in their mathematic abilities need more support and encouragement to take on new and challenging learning, most students with low levels of confidence still want to do well in school. This is a strong driver of using learning strategies (see Chapter 2). Students with low levels of confidence also show high levels of instrumental or extrinsic motivation as they think school teaches them things that might be useful for a job (see Table V.B1.4.10). These results suggest that even if students find learning less enjoyable, they still see its value for instrumental purposes and feel motivated to do well at school. This finding has important implications for lifelong learning: supporting students to become more confident could help them enjoy learning as a process in itself rather than just as a means to an end. Learning for the sake of learning is the kind of motivation that is more related to learning strategies for lifelong learning than learning to be able to get a job (see Chapter 3). While instrumental motivation is a good driver for learning in school, it is less likely to motivate them as adults if they do not enjoy learning. As discussed in Box V.4.1 and Box V.4.2, one way to improve students’ enjoyment of learning mathematics is through curricula and pedagogy. These include providing different options and levels to learn mathematics and ensuring teachers’ support.
Schools having different options and levels in mathematics caters to different student needs and aspirations, and keeps them studying mathematics in upper secondary education
Mathematics is still a compulsory subject in most education systems. This ensures that all students develop the numeracy skills they will need as adults. But as students get older and develop their interests and preferences, it becomes more challenging to keep them motivated and in school. Schools in most countries, especially at the upper secondary level, have found a solution by allowing students to choose among a variety of different course levels and options in mathematics. This caters to students’ wider set of needs and aspirations, and helps them stay motivated and finish their studies.
Singapore provides different mathematics levels as early as Grades 5 and 6. Primary education students who require more time and support can take the Foundation Mathematics syllabus instead of the Standard Mathematics syllabus. The Foundation Mathematics syllabus provides greater focus on basic concepts and skills, and allows students to learn a lower content load with more curriculum time.
At the secondary level, Singapore offers either the G1, G2 or G3 (G stands for General) Mathematics syllabus from Grade 7 to Grade 10, based on their Mathematics performance in primary school. Students have the flexibility to adjust their subject levels, based on their strengths, interests and learning needs. For example, G1 Mathematics caters to students who studied Foundation Mathematics in primary school. From Grade 9, students who are interested and are performing well at their current level may opt to study at a higher level; that is, either G2 Additional Mathematics or G3 Additional Mathematics.
Together, these 5 syllabi cater to students’ different needs, interests and abilities, with the following goals:
ensure that all students will achieve a level of mastery of mathematics that enables them to function effectively in everyday life
for those who have the interest and ability, to learn more mathematics so that they can pursue mathematics or mathematics-related courses of study in the next stage of education.
High-quality education and career guidance ensure that all students understand their options and are supported in making choices that will be decisive for their future
A high degree of choice, however, comes with some risks. Young people are often not aware of the high-stakes nature of deciding what they take in upper secondary education. They do not think about how it shapes their future education and work pathways. Students and their parents often make decisions based on perceptions or parents’ own experiences rather than objective information about the labour market outcomes associated with different options. Giving greater choice to students should go hand-in-hand with high-quality student guidance that informs students about future opportunities and labour market needs.
To help students better navigate these different options and decide which education and career pathways they want to pursue, Singapore provides Education and Career Guidance (ECG) to students across different levels of education:
Secondary schools and post-secondary education institutions are resourced with ECG counsellors who help students identify their strengths and interests through education and career counselling, workshops and student development experiences (e.g. ECG talks/fairs, talks by industry professionals and learning journeys to industries). ECG counsellors ensure that all students, regardless of their background and performance, receive tailored advice and support to make informed decisions about their education and career plans. For students with greater ECG needs (e.g. those from disadvantaged socio-economic status), targeted support is provided by ECG counsellors in consultation with school personnel overseeing student development. They also work with the school team to help teachers and other school personnel be able to provide basic advice on ECG for their students.
The “MySkillsFuture” is an online portal with information and tools to explore various education and career pathways for all students from primary to tertiary education.
The ECG Centre at the Ministry of Education provides education and career counselling services for students during the period when national examination results are released. Students can book an ECG counselling appointment, which can be conducted online or via phone.
ECG resources are made available online for students, teachers and their parents to better understand possible education and career pathways. Examples include a publication on the World of Work and the skills in demand, and a guide for parents to support their children in making choices at key moments in education. Stories of individuals and how they discovered their personal interests, and educational and career pathways are also part of these resources.
Source: Ministry of Education of Singapore (2024[9]), 2020 G1,G2 & G3 Mathematics Syllabuses and 2020 G2 & G3 Additional Mathematics Syllabus, https://www.moe.gov.sg/secondary/schools-offering-full-sbb/syllabus; Ministry of Education of Singapore (2024[10]), Overview of Education and Career Guidance, https://www.moe.gov.sg/education-in-sg/our-programmes/education-and-career-guidance/overview; OECD (Forthcoming[8]), Mathematics for Life and Work, A Comparative Perspective on Mathematics to Inform Upper Secondary Reform in England; Stronati (2024[11]), “Managing choice, coherence, and specialisation in upper secondary education”, https://doi.org/10.1787/4a278519-en.
Mathematics can make many students feel worried, stressed and, even, helpless (OECD, 2015[12]). Mathematics anxiety is a major challenge for students' well-being but also for their lifelong learning. PISA 2022 results show that students with high levels of self-efficacy are more likely to use learning strategies and feel motivated. Results show the opposite effect for those who reported lower levels of mathematics self-efficacy and experience higher levels of mathematics anxiety. The same result is found at the country/economy level. Most education systems that show the lowest levels of self-efficacy, such as some Latin American countries (Argentina, Brazil, Chile, Costa Rica, the Dominican Republic, Guatemala and Mexico) and some Southeast Asian countries (Cambodia, Malaysia, and the Philippines), also show the highest levels of mathematics anxiety (see Tables V.B1.4.1 and V.B1.4.12).
Mathematics anxiety is negatively associated with all socio-emotional skills measured in PISA 2022, especially with stress resistance3 and emotional control4, even when accounting for students’ and schools’ socio-economic profile and mathematics performance (see Table V.B1.4.22). This means that students who have more control over their emotions and who can manage stress better experience less mathematics anxiety. Students who are particularly anxious about mathematics find it difficult to work through mathematical tasks because they focus on managing the stress associated with the process rather than solving problems (Ashcraft, Kirk and Hopko, 1998[13]). These students also tend to avoid challenging themselves because they are worried about failing. These negative associations with mathematics are detrimental for lifelong learning. They discourage students from taking on problems that involve mathematics, which they are sure to encounter in contexts outside school. Overall, anxious students will avoid challenging situations, which are, nonetheless, important opportunities for evolving and developing new skills.
A sharp rise in students’ mathematics anxiety can hinder students’ opportunities for lifelong learning
Students in OECD countries feel particularly anxious about their results and performance in mathematics5. On average, 65% of students worry about getting poor marks in mathematics and 55% feel anxious about failing in mathematics. Furthermore, around 40% of students reported feeling nervous, helpless or anxious while solving mathematics problems or doing homework, meaning that students are anxious simply dealing with the subject and not just because they are worrying about failing (see Table V.B1.4.12). These shares are even higher in Brazil, Brunei Darussalam, El Salvador, Indonesia, Malaysia, the Philippines and Thailand. Lastly, PISA shows there has been a sharp rise in mathematics anxiety from 2012 to 2022 in most PISA-participating countries and economies (Box V.4.3).
The association between anxiety and mathematics can be detrimental to lifelong learning. Students who develop negative feelings towards mathematics at schools may be less likely to opt for further education that includes mathematics. They may avoid reskilling opportunities that involve mathematics as well. Reducing students’ mathematics anxiety has become a key policy challenge for improving students’ well-being, performance and their readiness to keep learning throughout their lives.
Additionally, low performers (those who perform below Level 2 in mathematics) show higher levels of mathematics anxiety than skilled students (those who perform at Level 3 or above in mathematics) (see Table V.B1.4.15) (see Box V.2.1 Chapter 2). This suggests that while anxiety is an obstacle to lifelong learning for all learners, it is even more so for those who also struggle with basic skills. Skilled students who have a solid foundation and strong mathematical skills will be able to build on those and be less likely to experience high levels of anxiety about mathematics. Those who lack a strong foundation in mathematics, however, will struggle to engage with lifelong learning and be more likely to experience anxiety and develop an adverse attitude towards mathematics.
Students feel more anxious doing mathematics now than 10 years ago
Compared to 2012, students in 2022 reported higher levels of mathematics anxiety in most PISA-participating countries and economies. While anxiety levels have risen sharply, especially in most European countries and most Latin American countries, it fell significantly from 2012 levels in Korea. More students also reported feeling helpless or tense doing mathematics problems or homework on average across OECD countries than in 2012. On the contrary, there was only a slight increase in students worrying about their marks and no change in the share of students worrying that it would be difficult for them in mathematics classes (see Table V.B1.4.23). This result is worrying. Students are developing increasingly negative attitudes about learning mathematics. This may impact not only their performance but their readiness for lifelong learning. This finding also suggests that young people’s well-being has deteriorated, and policies are needed to support students’ mental health.
Iceland and Türkiye had the largest increases in the shares of students reporting that they felt nervous doing mathematics problems between 2012 and 2022. Korea, Singapore and Thailand are the only countries that show a significant drop between 2012 and 2022, with Korea showing an impressive drop of 11 percentage points (see Figure V.4.4).
Change between 2012 and 2022 in the percentage of students agreeing or strongly agreeing that they get very nervous when doing mathematics problems
Notes: Only countries and economies with available data are shown.
Statistically significant differences are shown in a darker tone (see Annex A3).
OECD average-36 refers to the average across OECD countries, excluding Israel and Luxembourg.
Countries and economies are ranked in descending order of the percentage-point difference of students agreeing or strongly agreeing that they get very nervous when doing mathematics problems between 2012 and 2022.
Source: OECD, PISA 2022 Database, Table V.B1.4.23. See Table V.4.1 for StatLink at the end of this chapter.
Students worry about the future and feel anxious about their learning
Since 2003, PISA has recorded an increase in students’ levels of anxiety. This trend could be related to students increasingly feeling more uncertain and afraid of the future. PISA 2022 asked students a series of questions about their feelings about the future and how prepared they are for it. Almost 50% of students worry they are not prepared for life after they finish compulsory education, on average across OECD countries. This share is higher in Argentina, Brunei Darussalam, Singapore and Chinese Taipei, exceeding 70%. In Hungary, Morocco and Palestinian Authority, it is lower than 30% (see Table V.B1.4.31).
An average of almost 45% of students worry they won't have enough money to do what they would like to do after the final year of compulsory education. This share is higher for in Brazil, Brunei Darussalam, Chile, Colombia, Costa Rica and Peru, exceeding 65% (see Table V.B1.4.33). While less than 35% of students reported feeling pressure from family to follow a specific path after finishing compulsory education, in Brunei Darussalam, the Philippines and Uzbekistan, almost 60% reported this (see Table V.B1.4.32).
All these different anxieties about the future have a large positive association with mathematics anxiety, meaning that students who worry about the future can also suffer from greater mathematics anxiety. And, while all these different anxieties about the future relate negatively to mathematics performance, those who worry about not being prepared for life after school show the largest association with mathematics anxiety (see Table V.B1.3.10). In terms of performance, feeling pressure from parents has the largest impact, with a decrease of 14 points in mathematics performance after accounting for students’ and schools’ socio-economic profile (see Table V.B1.4.31). Because students’ performance and mental health are so related to their fears about the future, it is important that students have access to career guidance and counselling as well as psychological and emotional support.
Source: (OECD, 2004[14]); (OECD, 2015[12]); (OECD, 2019[15]).
Reducing mathematics anxiety among girls and disadvantaged students can help improve their performance and readiness for lifelong learning
Since 2003, PISA has shown a negative association between mathematics anxiety and mathematics performance in every education system that has participated in PISA (OECD, 2004[14]). Specifically, PISA 2022 finds that a one-point increase in the index of mathematics anxiety is associated with a decrease in mathematics achievement of 18 score points after accounting for students’ and schools’ socio-economic profile on average across OECD countries (see Table V.B1.4.14) (OECD, 2023[16]). In some countries and economies this association is larger, especially in Denmark* where a one-point increase in the index of mathematics anxiety is associated with a decrease of 27 points. In Brunei Darussalam, the Dominican Republic and El Salvador, however, the decrease is of only 6 points.
In most countries and economies, 15-year-old girls reported significantly higher levels of mathematics anxiety than boys. While these differences partly reflect differences in mathematics performance related to gender, the gender gap in anxiety persists even among top-performing students, suggesting that girls feel more anxious than boys even when they perform at similarly high levels (see Table V.B1.4.16). Similar results are found when comparing advantaged and disadvantaged students (see Table V.B1.4.13). This suggests that focusing solely on performance is not the solution for reducing students' anxiety. Neither is it an effective way to tackle gender and socio-economic gaps. Rather, schools might work to make sure that girls and disadvantaged students do not perceive learning outcomes, like performance in mathematics, as inherently difficult or beyond their capabilities. This is important not just for lifelong learning but learning in general.
Students who are more anxious tend to ask less questions and demonstrate a less proactive study behaviour than those who are not
Students who are more anxious reported using learning strategies less often than those who are less anxious (see Table V.B1.4.18). In particular, anxious students are less likely to ask questions when they do not understand material covered in class than their less anxious peers. And, more anxious students use fewer proactive behaviour study techniques, such as connecting new material to previous mathematics lessons. These differences are, respectively, of 20 and 16 percentage points, on average across OECD countries. The gap is larger in non-OECD countries and economies, especially in Baku (Azerbaijan), Bulgaria, Morocco, Saudi Arabia and Uzbekistan, where the differences between the shares of anxious and less anxious students reporting these learning strategies exceed 30 percentage points.
Learning strategies, especially those related to proactive behaviour, relate negatively to mathematics anxiety
The negative association between mathematics anxiety and performance is partly related to the negative effect of anxiety on cognitive resource activation. This means that when students are anxious about mathematics, they cannot focus on solving mathematics problems because they are distracted by worrying about these tasks and their ability to solve them. One way to help students improve their performance is to help them develop learning strategies that can reduce anxiety.
PISA data suggest that specific proactive mathematics study behaviours, which include connecting new and prior knowledge, have large negative associations with mathematics anxiety (see Figure V.4.5). This means that students who use certain proactive learning techniques do not suffer as much from mathematics anxiety. Similarly, learning strategies introduced by teachers to encourage cognitive activation, problem-solving and creativity are linked with lower levels of anxiety. This is true even after accounting for students’ and schools’ socio-economic profile, and students’ mathematics performance. Fostering and supporting the use of learning strategies in the classroom, like new ways to solve problems, and connecting new and prior knowledge can help students cope with mathematics anxiety, and, in turn, improve their lifelong learning opportunities.
In contrast, learning strategies related to controlling one’s learning, such as checking homework or making sure there are no mistakes, and those related to critical thinking have very small or almost no association with anxiety. While causality cannot be attributed from these analyses, one way to interpret these results is that independent learning behaviours such as checking for correct or incorrect answers might not always provide students with the external validation they need to reduce their anxiety.
It is important to mention that while most of the associations between anxiety and learning strategies are quite small for OECD countries, they become larger in some non-OECD education systems, especially in Albania, Baku (Azerbaijan), Jordan, Saudi Arabia and Uzbekistan (see Figure V.4.5b [available online] and Table V.B1.4.20).
Change in mathematics anxiety when students reported the following learning strategies; OECD average
1. The socio-economic profile is measured by the PISA index of economic, social and cultural status (ESCS).
2. Students doing the corresponding statement more than half of the time.
3. Students agree or strongly agree with the corresponding statement.
4. Students doing the corresponding statement more than half of the lessons.
Notes: Differences that are statistically significant are shown in a darker tone (see Annex A3).
Differences before accounting for students' and schools' socio-economic profile, and mathematics performance are all statistically significant (see Annex A3).
Items are ranked by kind of learning strategies and then in descending order of the change in mathematics anxiety index, after accounting for students' and schools' socio-economic profile and performance in mathematics.
Source: OECD, PISA 2022 Database, Table V.B1.4.20. See Table V.4.1 for StatLink at the end of this chapter.
Students who are more anxious find less enjoyment in learning than those who are not
Responses to “I love schoolwork that is challenging” and “I love learning new things in school” uncovered the greatest gap in motivation (22 and 10 percentage points on average across OECD countries, respectively) between more and less anxious students (see Figure V.4.6). Students who are more anxious about mathematics are less likely to enjoy learning new and challenging things as they worry about their grades and failing. On the contrary, students feel motivated to do well in school despite their anxiety. This could be because the desire to do well in school can sometimes trigger worries about failing.
Compared to the OECD average, Italy shows smaller differences between anxious and less anxious students enjoying learning new things and challenging schoolwork. On the contrary, Denmark*, Finland, Macao (China) and Norway show much larger differences (see Table V.B1.4.19).
Percentage of students agreeing or strongly agreeing with the following motivations, by the bottom and top quarters of the index of mathematics anxiety; OECD average
Notes: Students who are less anxious (more anxious) are those in the bottom (top) quarter of the index of mathematics anxiety in their own country/economy.
Differences between students in top and those in bottom quarters of the index of mathematics anxiety are all statistically significant (see Annex A3).
Items are ranked in descending order of the percentage-point difference related to mathematics anxiety (less anxious minus more anxious students).
Source: OECD, PISA 2022 Database, Table V.B1.4.19. See Table V.4.1 for StatLink at the end of this chapter.
Mathematics anxiety relates negatively to student motivation
Mathematics anxiety is negatively associated with all types of motivation, even when accounting for students’ and schools’ socio-economic profile (see Table V.B1.4.21). This means that motivation might help students feel less anxious. On average across OECD countries, this association is large with intrinsic motivations such as “I like schoolwork that is challenging”. This suggests that being motivated to learn could help reduce anxiety. On the contrary, while there is a negative association between anxiety and wanting to do well in mathematics class, this is not significant when accounting for mathematics performance. This means that wanting to do well in mathematics class is not a sufficient enough motivation for reducing mathematics anxiety. Interestingly, in 11 education systems the association remains significant and positive, suggesting that wanting to do well in school could contribute to students’ anxiety in these systems. This is especially true in Italy and Korea, where this positive association is larger than in other systems.
Finally, anxiety is also negatively associated with the instrumental motivation of making an effort at school because the knowledge may be useful in a job. While it is not possible to attribute causality from these analyses, one way to interpret this result is that understanding the value of school for finding a job in the future might help students feel less anxious. This finding is important as schools and teachers can help students build their instrumental motivation with school guidance and career exploration (see Chapter 6). This result also holds when accounting for mathematics performance.
Teachers’ attitudes towards students can help reduce their anxiety towards mathematics and increase their use of learning strategies
Reducing mathematics anxiety and tackling students' negative associations with mathematics while they are still in school can help learners embrace lifelong learning, especially lifelong learning that involves mathematics. Teachers can significantly enhance or diminish their students' self-beliefs, and this extends to anxiety. Teachers can go a long way towards alleviating their students’ mathematics anxiety by using specific teaching techniques and support strategies, and, as found in PISA 2022 Volume II (OECD, 2023[7]), building positive relationships with students.6
The index of quality of student-teacher relationships is negatively associated with mathematics anxiety in almost all countries and economies (see Table V.B1.4.29). Feeling intimidated by teachers at school has the largest positive association with students’ anxiety towards mathematics (see Figure V.4.7). Put another way: the better students get on with their teachers the less mathematics anxiety they suffer from. A one-point increase in the index of quality of student-teacher relationship is associated with a 0.13 decrease in the index of mathematics anxiety after accounting for students’ performance in mathematics on average across OECD countries. Every PISA 2022 test item on the quality of teacher-student relationship has a strong association with mathematics anxiety across OECD countries.
Students who have more positive relationships with their teachers also reported using all learning strategies more than those who had less positive relationships and they are also more motivated (see Tables V.B1.4.25 and V.B1.4.26). In countries and economies where the quality of student-teacher relationships particularly impacts students’ anxiety levels, such as in El Salvador, Georgia, Kazakhstan and the Ukrainian regions (18 of 27) healthy relationships between teachers and their students are crucial. It can directly galvanise students to use learning strategies and indirectly decrease students’ anxiety (see Figure V.4.7b [available online] and Table V.B1.4.29).
Change in mathematics anxiety when students agree or strongly agree with the following statements; OECD average
1. The socio-economic profile is measured by the PISA index of economic, social and cultural status (ESCS).
Note: All differences are statistically significant (see Annex A3).
Items are ranked in descending order of the change in mathematics anxiety, after accounting for students’ and schools’ socio-economic profile, and mathematics performance.
Source: OECD, PISA 2022 Database, Table V.B1.4.29. See Table V.4.1 for StatLink at the end of this chapter.
References
[13] Ashcraft, M., E. Kirk and D. Hopko (1998), “On the cognitive consequences of mathematics anxiety”, The development of mathematical skills, pp. pp. 175–196.
[1] Bandura (1997), Self-efficacy: The exercise of control., W H Freeman/Times Books/ Henry Holt & Co.
[4] Bandura, A. (1977), “Self-efficacy: Toward a unifying theory of behavioral change”, Psychological Review, Vol. Vol. 84/2, pp. pp. 191-215, https://doi.org/10.1016/0146-6402(78)90002-4.
[2] Bong, M. and E. Skaalvik (2003), “Academic Self-Concept and Self-Efficacy: How Different Are They Really?”, Educational Psychology Review, Vol. 15, pp. 1–40, https://doi.org/10.1023/A:1021302408382.
[9] Ministry of Education of Singapore (2024), 2020 G1,G2 & G3 Mathematics Syllabuses and 2020 G2 & G3 Additional Mathematics Syllabuses, https://www.moe.gov.sg/secondary/schools-offering-full-sbb/syllabus.
[10] Ministry of Education of Singapore (2024), Overview of Education and Career Guidance, https://www.moe.gov.sg/education-in-sg/our-programmes/education-and-career-guidance/overview.
[16] OECD (2023), PISA 2022 Results (Volume I): The State of Learning and Equity in Education, PISA, OECD Publishing, https://doi.org/10.1787/53f23881-en.
[7] OECD (2023), PISA 2022 Results (Volume II): Learning During – and From – Disruption, https://doi.org/10.1787/a97db61c-en.
[15] OECD (2019), PISA 2018 Volume III: What school life means for students’ lives, https://www.oecd.org/publications/pisa-2018-results-volume-iii-acd78851-en.htm (accessed on 6 December 2021).
[17] OECD (2016), PISA 2015 Results (Volume I): Excellence and Equity in Education, OECD Publishing, https://doi.org/10.1787/9789264266490-en.
[12] OECD (2015), Does Math Make You Anxious?“, PISA in Focus, No. 48, https://doi.org/10.1787/5js6b2579tnx-en.
[14] OECD (2004), Learning for Tomorrow’s World: First Results from PISA 2003, PISA, OECD Publishing, https://doi.org/10.1787/9789264006416-en.
[8] OECD (Forthcoming), Mathematics for Life and Work, A Comparative Perspective on Mathematics to Inform Upper Secondary Reform in England.
[11] Stronati, C. (2024), Managing choice, coherence, and specialisation in upper secondary education, https://doi.org/10.1787/4a278519-en.
[5] Urdan, T. (ed.) (2010), Self-efficacy in educational settings: Recent research and emerging directions, Emerald, United Kingdom, https://doi.org/10.1108/S0749-7423(2010)000016A004.
[3] Wang, M., J. Eccles and S. Kenny (2013), “Not lack of ability but more choice: Individual and gender differences in choice of careers in science, technology, engineering, and mathematics. ,”, Psychological Science, Vol. 24(5), pp. 770–775, https://doi.org/10.1177/0956797612458937.
[6] Wigfield, K. (ed.) (2009), Self-efficacy theory, Taylor Francis, New York.
← 1. Students’ ratings of their agreement with statements about a range of behaviours indicative of persistence, such as “I keep working on a task until it is finished” or “I give up after making mistakes”, were scaled into the index of persistence. Each of the 10 items included in this scale have five response options: “Strongly disagree”, “Disagree”, “Neither agree nor disagree”, “Agree” and “Strongly agree”.
← 2. As in previous cycles of PISA, in 2022 students were asked to rate how confident they felt about having to do a range of formal and applied mathematics tasks. The tasks include: working out from a train timetable how long it would take to get from one place to another; calculating how much more expensive a computer would be after adding tax; calculating how many square metres of tiles are needed to cover a floor; understanding scientific tables presented in an article; solving an equation like 6x² + 5 = 29; finding the actual distance between two places on a map with a 1:10,000 scale; solving an equation like 2(x+3) = (x + 3) (x - 3); calculating the power consumption of an electronic appliance per week; solving an equation like 3x+5=17. Responses from these items were used to create the index of mathematics self-efficacy. The index of mathematics self-efficacy used in this chapter is for formal and applied mathematics. PISA also provides the index of mathematics self-efficacy for mathematical reasoning and 21st-century mathematics, which is constructed by asking students how they feel about having to do a range of mathematical reasoning and 21st-century mathematics tasks. This index is not used in this chapter but is covered in Chapter 8.
← 3. Students’ ratings of their agreement with statements about a range of behaviours indicative of stress resistance, such as “I remain calm under stress” or “I get nervous easily”, were scaled into the index of stress resistance. Note that this scale used a within-construct matrix sampling design. Each of the 10 items included in this scale have five response options: “Strongly disagree”, “Disagree”, “Neither agree nor disagree”, “Agree”, “Strongly agree”.
← 4. Students’ ratings of their agreement with statements about a range of behaviours indicative of emotional control, such as “I keep my emotions under control” or “I get mad easily”, were scaled into the index of emotional control. Each of the 10 items included in this scale have five response options: “Strongly disagree”, “Disagree”, “Neither agree nor disagree”, “Agree”, “Strongly agree”.
← 5. PISA 2022 asked students to report whether they agreed or strongly agreed with a series of statements about experiencing anxiety about mathematics. These responses were then combined to construct the index of mathematics anxiety.
← 6. Students who participated in PISA 2022 were asked to rate their agreement with a series of statements about the quality of their student-teacher relationships. The items were scaled into the index of “Quality of student-teacher relationships”.