PISA 2025 Results (Volume I)

PISA 2025 data provide a solid base to analyse students’ engagement in learning, which is fundamental to developing their future-readiness and capacity to adapt to shifting societal demands and a rapidly changing labour market. The attitudes, beliefs and ways of engaging with learning examined in this chapter are outcomes in their own right; they reflect students’ experiences both in and outside of school. The chapter therefore examines whether students are developing the dispositions and strategies that can support lifelong learning and the increasingly essential ability to adapt to changing demands and contexts, while it also analyses the role schools play in this development.

The different attitudes analysed here relate to students’ engagement with learning. Engagement is commonly conceptualised as a multidimensional construct that captures how students participate in, think about and feel connected to school. It is often described in terms of behavioural engagement (involvement, persistence and positive conduct), cognitive engagement (psychological investment, self-regulation and deep processing) and emotional or social engagement (sense of belonging, supportive relationships and valuing of school) (Fredricks, Blumenfeld and Paris, 2004[1]; Appleton, Christenson and Furlong, 2008[2]; Finn and Zimmer, 2012[3]). This multidimensional approach guides the focus of the analyses in this chapter; for example, on students’ motivation or drive to learn, their beliefs about the malleability of their own intelligence and the nature of scientific knowledge, their sense of belonging at school and the value they place on it for their future.

Another key dimension is student agency, which is closely related to cognitive and social engagement with learning. Agency concerns the extent to which students can act purposefully within their learning environment (Reeve, 2012[4]). Rooted in self-determination theory, it refers to students’ capacity to exercise autonomy through choice, volition and meaningful participation in learning (Deci and Ryan, 2000[5]; Reeve and Tseng, 2011[6]). Agency is treated here as a distinct dimension of engagement, wherein students make proactive and constructive contributions to their learning; for example, by verifying and adapting their approaches to learning or to seeking help. Agency does not simply reflect individual disposition; it also includes whether students are given opportunities to exercise choice and take ownership of their learning. In this sense, it helps to illustrate how motivation, beliefs and school experiences interact to shape the depth of students’ engagement with learning (Assor, 2012[7]; Cleary and Zimmerman, 2012[8]; Reeve, 2012[4]).

This framework provides a useful basis for organising the analyses in this chapter across four analytical groupings:

• Students’ drive to learn, captured here through curiosity for learning (or intrinsic motivation) and perseverance, dispositions that can support effort and persistence in the face of challenge.

• Learning-related beliefs and behaviours, including beliefs about one’s own ability to progress (or growth mindset), more general beliefs about knowledge and science, and students’ capacity to adapt their thinking and learning strategies when faced with challenge or uncertainty (or cognitive adaptivity), which stand at the base of student agency.

• Self-regulated learning, which shapes how motivation is translated into agency through strategic thinking and psychological investment in learning.

• Students’ beliefs about the value of school and their sense of belonging at school. Together, these dimensions reflect whether students regard school as worthwhile and experience it as a place where they feel accepted, included and valued, which can strengthen their emotional connection to learning and help sustain their engagement over time.

Finally, the results presented in this chapter should be interpreted with caution, given indicators in this chapter are based on students’ self-reports. Observed cross-country/economy differences should be interpreted carefully, as these differences may reflect variations in response behaviour rather than true underlying differences. Cultural norms, as well as differing interpretations of and reactions to questionnaire items can influence how students understand and answer questions.1

Two closely related dimensions of students’ drive to learn captured by PISA 2025 data are intrinsic motivation, indicated here through students’ curiosity to learn, and perseverance. Intrinsic motivation refers to engaging in learning for its inherent interest, enjoyment or satisfaction (Deci, 2013[9]; Eccles and Wigfield, 2002[10]). Curiosity to learn is closely related to intrinsic motivation, as it helps direct attention and encourages exploration and learning (Hidi and Renninger, 2006[11]; Izard, 2007[12]). In PISA, these two dimensions are treated as closely intertwined in capturing the internal energy and direction that can prompt students to become involved in learning (Cleary and Zimmerman, 2012[8]; Schunk, Pintrich and Meece, 2014[13]).

This drive to learn is expressed behaviourally through perseverance, understood here as a key indicator of sustained engagement with academic tasks, and can serve as a proxy for resilience (Fredricks, Blumenfeld and Paris, 2004[1]; Skinner et al., 2009[14]). Perseverance refers to students’ capacity to remain engaged, maintain effort and continue working through difficulty or complexity (Corno and Mandinach, 1983[15]; Skinner and Belmont, 1993[16]). While curiosity may help initiate engagement, perseverance helps sustain it over time. Examining these dimensions together makes it possible to consider not only what may energise students’ learning, but also how that energy is carried into action and maintained in ways that support progress at school (Frydenberg, 2005[17]).

As shown in Table I.3.1, student curiosity and perseverance are closely related dimensions of the drive to learn. Across many systems, students who report greater curiosity also tend to report greater perseverance, and changes over time in some indicators of curiosity and perseverance suggest that the two may weaken in parallel.

PISA data capture the drive or curiosity to learn through several complementary questions. It asks students whether they like learning new things, if they find learning new things to be boring, whether they like to know how things work, whether they are curious about different things and whether they feel more curious than most people they know.

The PISA 2025 results show that, on average across OECD countries, 69% of students report that they like learning new things (Table I.B1.3.1). This shows students’ self-reported intrinsic motivation to learn: engaging in learning for its inherent interest, enjoyment or satisfaction. Similarly, 73% of students report being curious about many different things, and 74% of students report that they like to know how things work, on average across OECD countries (Table I.B1.3.1).

Between 2022 and 2025, PISA data point to a drop in the share of students reporting several aspects of curiosity: in most systems, students appear to be less curious than they were in 2022. Across the five trends items considered here, four declined in a majority of participating countries and economies. The sharpest decline concerns the percentage of students who disagree or strongly disagree that “learning new things is boring” (60 systems); in no system did PISA 2025 show the opposite trend. As shown in Figure I.3.2, the share of students who disagreed or strongly disagreed with this statement declined by more than 10 percentage points in 30 countries and economies. Only in 11 countries and economies did the share remain unchanged between 2022 and 2025.

The only item for which students’ reports increased between 2022 and 2025 concerns feeling “more curious than most people they know”, which rose in a majority of systems (59) (Figure I.3.3 and Table I.B1.3.2). This share increased by more than 10 percentage points in 16 countries and economies; in most of these systems, fewer than half of students had reported this in 2022. By contrast, it remained unchanged in 10 systems and declined only in Mongolia in Uzbekistan, by 3 and 4 percentage points, respectively.

This opposite trend is noteworthy because, although the item may capture an internalised belief, it is inherently relational. Students are not assessing themselves in isolation, but in comparison with the people around them. Their responses could therefore reflect both their self-concept and the social context in which it is formed, including the behaviours and attitudes they observe among their peers. In this sense, the measure can be interpreted not only as an indication of how students view themselves, but also as an indication of the environment in which they interpret their own curiosity.

Moreover, these contrasting trends – the increase in the share of students who report that they find learning new things to be boring, alongside the increase in the share who report feeling more curious than most people they know – may point to a widening gap between students’ self-perceived orientation towards exploration and learning, and the extent to which their learning environments stimulate and sustain situational interest (Hidi and Renninger, 2006[11]). In other words, students may continue to see themselves as relatively curious while experiencing learning settings that do not sufficiently engage or challenge them, and therefore do not activate that curiosity in practice (Eccles et al., 1999[18]) .

When examining differences within countries and economies, the PISA 2025 results show that girls and socio-economically advantaged students report higher levels of curiosity than boys and disadvantaged students, on average. Differences in students’ orientation towards learning may therefore be embedded in broader inequalities in their school experiences and attitudes.

This pattern is observed across most participating countries and economies. In 19 countries/economies, there are no differences between boys’ and girls’ reports, while only in B-S-J-Z (China), Hong Kong (China), Japan, Korea, Macao (China), Rwanda, Singapore and Chinese Taipei do boys report feeling more curiosity than girls (Table I.B1.3.3). Socio-economically advantaged students also report higher levels of curiosity than their disadvantaged peers, on average across OECD countries and participating countries/economies. In no country/economy is the inverse pattern observed. Only in Jordan, Kenya and Rwanda are there no differences by students’ socio-economic profile (Table I.B1.3.3).

These results also suggest that average levels of motivation at the system level should be interpreted with caution, since they may mask substantial variation between groups of students. More broadly, they point to the importance of considering not only whether students report positive dispositions towards learning, but also how these dispositions are distributed across groups and how they relate to the contexts in which students learn.

Perseverance is another important dimension of PISA analysis of student attitudes and behaviours, concerning what happens once students encounter difficulty, frustration or uncertainty. In this chapter, perseverance is treated as a behavioural expression of the drive to learn and as a proxy for resilience, understood here not as an exceptional individual trait, but as the capacity to sustain effort and remain engaged when learning becomes more demanding. Perseverance adds a behavioural dimension to the analysis and suggests that engagement may depend not only on the interest generated by learning, but also on whether students are able to sustain effort once learning becomes challenging.

In PISA 2025, students were asked about their attitudes and follow-through in schoolwork, including whether they finish what they start, whether they keep working on a task until it is finished, whether they apply additional effort when work becomes challenging, whether they finish tasks they have started even when they become boring, and whether they give up easily when faced with long homework assignments.2

On average across OECD countries, under two-thirds of students (61%) reported that they finish what they start. However, cross-country variation is considerable (Table I.B1.3.9). In addition, 60% of students, on average across OECD countries, reported that they apply additional effort when work becomes challenging, and the share is lower for finishing tasks they have started even when they become boring, at 52%.

These attitudes appear to have weakened since the previous PISA cycle. In 28 systems, the share of students reporting that they keep working on a task until it is finished decreased, ranging from about a 3 percentage-point decline in Mexico to about 13 percentage points in Kosovo. Likewise, as shown in Figure I.3.4, in 32 systems, a smaller share of students reported that they apply additional effort when work becomes challenging than in 2022, with declines ranging from about 2 percentage points in Poland to about 11 percentage points in Bulgaria. In 40 systems, the share of students reporting that they finish what they start also fell, ranging from 2 percentage points in Costa Rica to 12 percentage points in the Dominican Republic (Table I.B1.3.10).

These trends are noteworthy given that, on average across OECD countries, about one-third of students reported that they quit doing homework if it is too long, and this share has increased in almost all systems since 2022: in 65 out of 69 with available data (Tables I.B1.3.9 and I.B1.3.10). Moreover, about half of students reported that they disagree or strongly disagree that they give up easily, and this share has fallen since 2022, decreasing in almost all systems (65).

Particular attention should be paid to the decline in students’ self-reported perseverance, and to the rising share of students who report giving up on schoolwork, because these patterns bear directly on whether students’ engagement with learning can be sustained when they encounter difficulty. Considered alongside the deterioration of curiosity presented earlier in this chapter, they suggest that several dimensions of engagement may be weakening in parallel. More specifically, the PISA results suggest that systems showing a deterioration in certain curiosity-related attitudes also tend to show a deterioration in perseverance-related attitudes over the same period (Figure I.3.5).

While this does not imply any directional relationship between curiosity and perseverance, it does point to a broader weakening in both students’ initial disposition to engage and their willingness to persist once learning becomes demanding. As a behavioural proxy for resilience, lower perseverance, particularly when it coincides with less positive curiosity-related attitudes, may suggest that students’ orientation towards learning is not always matched by learning experiences that help sustain engagement through difficulty.

This close connection between students’ drive to learn and their perseverance when learning becomes demanding can be seen across students’ attitudes. For example, among students in the top quarter of the perseverance index, 86% reported that they like learning new things, on average across OECD countries (Figure I.3.6). By contrast, among students in the bottom quarter of the index, fewer than half did so (47%, on average). In no country/economy did fewer than 71% of the most reportedly perseverant students also report that they like learning new things, and in Colombia, Dushanbe (Tajikistan), Guatemala, the Philippines, Uzbekistan and Viet Nam virtually all students who reported high perseverance did so (Table I.B1.3.6).

Similar to the earlier analysis of curiosity, on average across OECD countries, and in a majority of participating countries and economies, girls report higher levels of perseverance than boys. In 21 participating countries and economies, however, there is no statistically significant difference, while in only 9 systems do boys report higher levels of perseverance than girls (Table I.B1.3.11). Similarly, in virtually all countries and economies, socio‑economically advantaged students report higher levels of perseverance than their disadvantaged peers, as is also the case on average across OECD countries. The only exceptions are Croatia and Rwanda, where there is no difference by students’ socio-economic profile (Table I.B1.3.11).

Overall, these results suggest that perseverance is an important behavioural dimension of students’ drive to learn, but that it should be analysed in close relation to their curiosity. The average decline in perseverance, alongside the deterioration in some indicators of curiosity and the increase in students reporting that they give up more easily, points to a broader weakening in several dimensions of engagement across OECD countries.

At the same time, perseverance is not distributed evenly across groups of students. This suggests that students’ capacity to remain engaged when schoolwork becomes difficult should not be understood simply as a matter of individual character. Rather, it forms part of a broader pattern in which students’ motivation, behaviour and school experiences are differently aligned across students and education systems.

This section examines three dimensions of students’ cognitive engagement with learning: growth mindset, beliefs about the nature of science (or epistemic beliefs) and adaptability when facing unusual or changing circumstances. These are not fixed individual attributes but learning-related beliefs and practices that are shaped and expressed in schools and beyond. Their relevance lies in that they are related to the ways in which students interpret learning, respond to challenge and act within learning situations (Cleary and Zimmerman, 2012[8]; Wolters and Taylor, 2012[19]).

The first two dimensions concern learning-related beliefs that can orient students’ engagement. A growth mindset refers to the belief that intellectual abilities can be developed through effort, engagement with challenge and response to feedback (Dweck, 2006[20]; Dweck, 2013[21]; Dweck and Leggett, 1988[22]), while beliefs about the nature of science concern students’ views about the nature and justification of knowledge and, therefore, whether learning is approached as the passive reception of information or as the active construction of meaning (Kitchner, 1983[23]). The last dimension, adaptability, concerns the flexibility to respond to changing environments and new demands (Martin et al., 2012[24]).

Each of these cognitive dimensions is examined in relation to the motivational and behavioural dimensions discussed in the previous section, showing how these dimensions align – or fail to align – with the broader architecture of student engagement. As shown in Table I.3.1, across many systems, growth mindset, the beliefs about the nature of science and cognitive adaptability are positively related both to one another and to curiosity and perseverance. These relationships are not equally strong or consistent, however. The association between students’ beliefs about the nature of science and cognitive adaptability is more uneven across systems, while those involving growth mindset and student drive appear to be more strongly shaped by students’ academic performance.

A useful starting point for examining the cognitive dimension of engagement is students’ growth mindset3. As first conceptualised by Carol Dweck (2006[20]), students differ in the extent to which they endorse the belief that intelligence can be developed. At one end of this continuum, students are more likely to see ability as malleable and improvable through effort, persistence, help seeking and support; at the other, they are more likely to view intelligence as largely innate and unchangeable. These beliefs are best understood as different ways of interpreting effort, challenge and feedback. In this chapter, growth mindset is treated as one element in the configuration through which student agency is oriented, insofar as it may shape whether students see learning difficulties as signals to withdraw or as situations in which effort, support and strategic action may make a difference.

PISA data show that, across OECD countries, about 69% of students reported a growth mindset. Only in 17 countries/economies, fewer than half of students did so. Over 80% of students reported a growth mindset in Estonia and France (Table I.B1.3.16).

Growth mindset is not reported equally across groups of students: gender differences are generally small, whereas socio-economic differences are larger and more consistent across countries and economies. For example, on average across OECD countries, girls report a growth mindset slightly more often than boys (70% compared with 68%). Girls report a growth mindset more frequently than boys in 32 countries and economies; in 40 countries and economies there is no difference; and in 15 countries and economies boys report it more frequently than girls (Figure I.3.7 and Table I.B1.3.17).

Socio-economic differences are observed in the majority of education systems. On average, 73% of socio-economically advantaged students report a growth mindset, compared with 65% of disadvantaged students. This pattern is observed in 71 countries and economies, while in only 16 are there no differences between the two groups (Table I.B1.3.17). These differences form part of the broader pattern of how students’ orientations towards learning are distributed across education systems, and they provide an indication of how equitably such attitudes are fostered across groups of students.

Growth mindsets also relate to students’ broader orientation towards learning. As shown in Figure I.3.8, PISA 2025 data suggest that, on average across OECD countries and in a large group of participating countries and economies, growth mindset is positively associated with curiosity to learn, after accounting for gender and students’ and schools’ socio-economic profile (Table I.B1.3.7). For example, the share of students reporting a growth mindset is higher among those who agree or strongly agree with the statement “I like learning new things” than among those who do not, on average across OECD countries. The average gap is over 6 percentage points and exceeds 19 percentage points in B-S-J-Z (China) (Table I.B1.3.20).

Once student performance in science is taken into account, however, the relationship between growth mindset and this form of curiosity becomes much less straightforward: it disappears or turns negative in 20 systems (Table I.B1.3.8). This indicates that the positive association observed on average is partly structured by the fact that both growth mindset and intrinsic motivation tend to be reported more often among higher-performing students (Tables I.B1.3.5 and I.B1.3.19).

This is aligned with previous PISA analyses that have shown that other students’ attitudes (like curiosity or perseverance) do not relate to growth-mindset beliefs in the same way, but depend on their level of academic performance (OECD, 2025[25]). For lower-performing students in particular, the belief that improvement comes through effort and feedback may not always be matched by learning experiences that make such improvement visible or attainable. This may help explain why the relationship between growth mindset and curiosity to learn weakens once students are compared at similar levels of science performance.

The next question, therefore, is whether this alignment extends to perseverance, i.e. students’ willingness to sustain effort when schoolwork becomes difficult and demanding. This is particularly relevant given that students with a growth mindset are more likely to view setbacks as part of the learning process and as opportunities to refine their approach and improve.

PISA data show that, on average across OECD countries and in the largest group of participating countries and economies, growth mindset is positively related to perseverance, even after accounting for students’ and schools’ socio-economic profile (Figure I.3.9 and Table I.B1.3.15). Students who are more likely to believe that abilities can be developed through effort and feedback also tend to report greater persistence when schoolwork becomes difficult. On average across OECD countries, the share of students with a growth mindset who reported they apply additional effort when work becomes challenging is higher (72%) than those who did not (67%) (Table I.B1.3.20). This average gap of 5 percentage points reaches about 14 percentage points in Macao (China), and 23 percentage points in B‑S‑J‑Z (China).

This positive relationship complements the earlier findings on students’ curiosity to learn. This does not imply any directional relationship between these dimensions. It does, however, suggest that beliefs about improvement and sustained effort often align within students’ broader engagement with learning.

This pattern is not observed in all systems, however. In some contexts, the relationship between growth mindset and perseverance is negative, and many of these are countries and economies at the lower end of average performance (Table I.B1.3.15).

Analyses that also account for science performance confirm that the relationship weakens or turns negative in some systems. Similar to curiosity, this suggests that growth mindset does not carry the same meaning at all levels of achievement. For perseverance, however, the interpretation may be somewhat different. As previous analysis with PISA data have suggested (OECD, 2025[25]), at different levels of achievement, persistence may reflect different things: for top-performing students, for example, a strategic and improvement-oriented response to challenge; for lower-performing students, perseverance may reflect the greater effort required simply to maintain similar performance, not necessarily to improve performance (OECD, 2025[25]).

Together, these findings suggest that growth mindset is best understood as one element within a broader configuration of engagement rather than as a stand-alone belief. In many contexts, students who report a growth mindset also tend to report greater curiosity and greater perseverance. At the same time, these relationships are not uniform across and within systems and are less consistent once science performance is taken into account. For this reason, growth mindset should not be treated as a fixed label or as a sufficient marker of cognitive engagement on its own. Rather, it offers one window into how students understand learning, effort and the possibility of improvement within the contexts in which they learn (see Box I.3.1).

Another important dimension explored in PISA 2025 concerns students’ beliefs about the nature of science4: what they think about scientific knowledge itself, and about what it means to learn scientifically, to understand, validate and justify what is known. These beliefs matter because they shape how students position themselves epistemically, whether knowledge is approached as something to be received, or as something to be examined, tested and revised. Across OECD countries, a large majority of students endorse key beliefs about the nature of science, particularly those related to evidence, verification and the revisability of knowledge. These beliefs also stand in opposition to the view that common sense should be relied on more than scientific studies. This becomes crucial in a world where much information circulates outside of traditional information sources, through social media and other platforms (Egan et al., 2026[27]).

PISA 2025 data show that, on average across OECD countries and in many participating countries and economies, over 81% of students endorse key beliefs about the nature of science. These include the view that “one important part of broad science is doing experiments to come up with ideas about how things work”, that “good answers are based on evidence from many different experiments”, that “information from one source should be checked from a range of other reliable sources before it is accepted”, that “there are some questions that scientists cannot answer”, and that “new discoveries can change what scientists think is true”. In no system do fewer than 60% of students report agreeing with these beliefs, with two notable exceptions: in Kenya and Rwanda, about half of students disagree that “there are some questions that scientists cannot answer”. If such a large number of students believe that there are no questions that science cannot answer, it may point to a more limited understanding of scientific knowledge as grounded in hypothesis, evidence and verification, and as open to revision in light of new findings (Table I.B1.3.30).

A closely related question concerns students’ belief that “it is important to rely more on common sense and less on scientific studies” (43% of students agree or strongly agree with this statement, on average across OECD countries) (Table I.B1.3.30). This position contrasts with the view that “information from one source should be checked from a range of other reliable sources before it is accepted”, thus affirmative responses to these two items are negatively correlated (system-level correlation r = -0.42). More broadly, students who value common sense more than scientific studies are generally less likely to endorse the aforementioned wider set of beliefs about the nature of science. Even so, in a number of systems, a sizeable share of students both agree that “information from one source should be checked from a range of other reliable sources before it is accepted” and agree that “it is important to rely more on common sense and less on scientific studies”. As shown in Box I.3.2., in many of these systems, students who report this combination of beliefs are also likely to show positive learning outcomes.

Endorsing scientific epistemic values is positively aligned with students’ curiosity to learn at the system level (Figure I.3.13). This positive relationship is also observed within countries and economies (Table I.B1.3.7). On average across OECD countries, the share of students reporting they like to learn new things is 26 percentage points higher among those whose responses endorse scientific beliefs than among those who do not (Table I.B1.3.6).

Student responses that indicate perseverance also tend to align with support of scientific beliefs. Across all participating countries and economies, students’ beliefs about the nature of science are positively associated with perseverance, after accounting for student and school socio-economic profile (Table I.B1.3.15). This indicates a link between students’ understanding of knowledge and their willingness to sustain effort when learning becomes difficult.

These relationships suggest that the cognitive and motivational dimensions of engagement are closely connected: students who value learning as an inherently worthwhile activity also tend to hold more developed views about the nature of scientific knowledge. This alignment matters because it links students’ drive to inquire with an understanding of knowledge as something to be examined, justified and open to revision, rather than simply received as fixed fact. Furthermore, students who are more likely to view knowledge as evidence-based, examinable and open to revision also tend to be those who approach learning with greater interest and sustained effort. Therefore, if knowledge is approached this way, rather than as fixed and given, then it is reasonable to expect that this orientation may also align with beliefs about one’s own ability to develop and progress through effort and learning. This provides a natural transition to students’ growth mindset.

PISA data show that in 60 countries/economies, students who hold stronger beliefs about the nature of science are more likely to report a growth mindset. Only in 9 is the opposite pattern observed (Table I.B1.3.25). This reinforces this broader alignment across cognitive dimensions of engagement. Beliefs about one’s own capacity to improve and beliefs about how knowledge is formed often move in the same direction and are likely complementary: one concerns how students think about learning and ability, while the other concerns how they think about knowledge itself. Their positive relationship therefore points to a broader orientation in which knowledge (and therefore learning) is approached as something active, revisable and open to development, rather than as the passive reception of fixed truths.

Overall, these results suggest that scientific beliefs appear to sit at the intersection of the cognitive, motivational and behavioural dimensions of engagement: they are linked not only to how students think about scientific knowledge, but also to how they approach learning more broadly. Such beliefs about the nature of science could therefore provide an important bridge between students’ internal beliefs and the ways in which these beliefs may be reflected in their motivation and learning behaviour.

Cognitive adaptability reflects students’ capacity to adjust their thinking and behaviour when faced with unfamiliar situations, uncertainty or complexity. In learning contexts, this ability to respond to novelty and revise one’s approach is increasingly recognised as a core competence for effective learning in changing environments (Hilton and Pellegrino, 2012[28]). From this perspective, adaptability is a way of describing students’ confidence to manage learning when routine strategies are no longer enough. It is therefore a relevant form of resilience, because it shows whether students can remain engaged by adjusting their thinking and actions in the face of complexity or change.

PISA 2025 data show that cognitive adaptability is moderately widespread across OECD countries. At the same time, these attitudes vary considerably across countries and economies, indicating that students’ confidence in adapting to challenge and change is far from uniform. Developing cognitive adaptability is of crucial importance for students’ success in life; they will need this ability throughout their lifetimes to adjust and adapt their thinking and actions in order to meet rapidly changing societal demands.

Across OECD countries, about half of students believe that they can adapt to situations that take them out of their comfort zones. For example, about half of students (50%) reported that the statement “I can handle unusual situations” is "mostly like me" or "very much like me", on average (Table I.B1.3.40). The share is 54% when looking to the statement “I can adapt to different situations even when under stress or pressure”, and about 55% on “when encountering difficult situations with other people, I can find a way to resolve the situation”. The only one of the cognitive adaptability statements that slightly less than half of students feel describes them is: “I can adapt easily to a new culture” (48%, on average) (Table I.B1.3.40).

There is, however, important variation across countries/economies, as shown in Figure I.3.15 (online).5

Moreover, as shown in Figure I.3.15, boys tend to report higher cognitive adaptability than girls, on average across OECD countries and in 31 countries and economies. However, in 33 systems there are no differences between boys and girls, and in another 16, it is girls who reported higher cognitive adaptability than boys. Moreover, what is clearer is the divide between socio-economically advantaged and disadvantaged students. The former reported higher cognitive adaptability than the latter, on average and in almost all countries and economies (Table I.B1.3.42).

Cognitive adaptability can be read as the point at which students’ motivation to learn is tested under changing or uncertain conditions. It is positively related to motivational aspects like curiosity (Table I.B1.3.7). For example, students who are intrinsically motivated to learn (i.e. agree that they like to learn new things), are more likely to report so when they have cognitive adaptability (Table I.B1.3.8). About 78% of students with high cognitive adaptability (in the top quarter of the index) reported this sort of intrinsic motivation, while about 60% of those with low cognitive adaptability did (in the bottom quarter) (Table I.B1.3.6). Students who are more interested in learning and open to new ideas also tend to be better able to adjust when they face unfamiliar situations or unexpected challenges. In this sense, adaptability shows how students’ internal drive to learn may remain active even when learning requires flexibility rather than just routine effort.

Cognitive adaptability is associated with perseverance. As shown in Figure I.3.16, systems with more students reporting cognitive adaptability tend to be those with students reporting higher perseverance (Table I.B1.3.15). The same positive relationship is observed within countries and economies. For example, on average across OECD countries, 69% of students who reported the most cognitive adaptability also reported they apply additional effort when work becomes challenging, while only about half did so among students with low cognitive adaptability (Table I.B1.3.14).

If perseverance reflects sustained effort in the face of challenge, then growth mindset and epistemic beliefs help capture the cognitive orientation within which such adjustment becomes meaningful: learning is approached as something open to development, and knowledge as something that must be examined, tested and revised. However, PISA data show that cognitive adaptability is more closely aligned with students’ beliefs about learning and improvement than with their beliefs about the nature of knowledge itself.

Students who report more cognitive adaptability are more likely to report a growth mindset on average and in almost all countries and economies (Table I.B1.3.25). For example, 75% of students in the top quarter of the index of cognitive adaptability reported a growth mindset whereas about 63% of students in the bottom quarter did (Table I.B1.3.20). The difference between the two is over 30 percentage points in B-S-J-Z (China) and Lebanon. Only in Lithuania and Poland there is no significant difference between the two.

On the other hand, the beliefs that robust and verifiable scientific knowledge is relevant for the validation of true beliefs is not directly associated with these forms of cognitive flexibility in all systems. The specific idea that information from one source should be checked against a range of other reliable sources before it is accepted is a notion that is positively associated with cognitive ability on average across OECD countries and in several systems (29). For example, in Germany, 86% of students with high cognitive adaptability (in the top quarter of the index) reported that they agree that information from one source should be checked against a range of other reliable sources before it is accepted, while 75% of those with low cognitive adaptability did. The gap is smaller but still significant in B-S-J-Z (China),Canada*, Kazakhstan and Macao (China) (between 2 and 3 percentage points of difference). Yet, this association is not consistently found across all countries and economies (Tables I.B1.3.34 and I.B1.3.35).

Therefore, adaptability appears to be more closely connected to how students think about challenge, effort and the possibility of improvement than to how they think about the nature of knowledge itself. This is consistent with the idea that adapting to unfamiliar situations or working through difficulty depends in part on viewing learning as something that can be developed. This distinction helps clarify the kind of cognitive engagement that cognitive adaptability appears to reflect.

The findings analysed here show that cognitive adaptability is best understood as a practical, challenge-oriented dimension of student agency. Students who reported greater adaptability also tended, in many contexts, to report stronger curiosity, higher perseverance and a stronger growth mindset. Cognitive adaptability is thus more closely connected to how students respond to challenge, effort and the possibility of improvement. In this sense, it helps show whether students can remain engaged by adjusting their thinking and behaviour when learning becomes uncertain, demanding or unfamiliar.

The analysis now turns to agency, which refers to the ways in which students actively shape their own learning. In this section, agency is examined through two related dimensions: self-regulated learning help seeking and goal setting.

These dimensions show whether students do more than hold positive beliefs about learning: they indicate whether students seek support when needed, organise their work purposefully and autonomously, adjust strategies and respond to feedback. Examining these forms of agency is therefore an important next step, because they provide a more concrete view of how cognitive engagement is translated into action (Pintrich, 2004[29]; Zimmerman, 2000[30]).

These practices are understood through the lens of student agency, meaning their capacity to act purposefully in relation to their learning (Reeve, 2012[4]; Reeve and Tseng, 2011[6]). These two interrelated dimensions of agency – help seeking and goal setting – are examined in relation to the motivational and behavioural factors discussed in the previous section.

As shown in Table I.3.1, these dimensions of agency tend to align with curiosity and perseverance, although the strength and consistency of these relationships vary. In particular, goal setting appears to be more strongly and consistently aligned, and importantly, the relationship with curiosity appears to be strengthened where teacher support is higher. An inversed pattern is observed with other effort-oriented attitudes, such as growth mindset, which is more strongly and consistently associated with help seeking.

One important dimension of self-regulated learning is the ability to recognise when one’s current understanding or strategy is not sufficient, and to act on that recognition. While perseverance matters when students face difficulty, the expectation should not be that they manage every challenge entirely on their own. Agency does not mean solitary effort; it also includes knowing when and how to seek feedback, guidance or support (Zimmerman, 2000[30]). In this sense, help seeking is itself an expression of autonomy (Deci and Ryan, 2000[5]; Reeve, 2012[4]); it reflects a student’s capacity to monitor their learning, identify what they need, and take purposeful steps to move forward. As such, seeking support can be understood as an important form of agency, and one that is closely aligned with both growth mindset and perseverance (Cleary and Zimmerman, 2012[8]; Dweck, 2006[20]).

PISA 2025 data suggest that help seeking is a common and broadly shared form of self-regulated learning across OECD countries. About 77% of students across OECD countries reported that they ask for help when they have trouble learning and about the same share reported they ask their teachers for help if they need it, although there is important variation across countries as shown in Figure I.3.o5 (online) (Table I.B1.3.44).

But support is not only vertical, it can also come from one’s peers. How students feel about asking their classmates for help gives a sense of students’ ability to collaborate and support each other. About 82% of students report asking their classmates when they do not understand something, on average across OECD countries (Figure I.3.o4 online and Table I.B1.3.44).

Finally, the combination of effort and seeking support suggests that a student tends to ask for support once they have already tried to sort things out on their own. This combination of perseverance and feedback seeking is one important externalised aspect of students with a growth mindset (Blackwell, Trzesniewski and Dweck, 2007[31]; Dweck, 2006[20]). Across OECD countries, about 78% of students report only asking for help after trying hard to find the solution on their own, on average (Table I.B1.3.44).6

PISA data show that girls report greater help seeking than boys, on average (Figure I.3.17). This is also the case in a majority of countries and economies (55). Socio-economically advantaged students also tend to report this to a larger extent than their more disadvantaged peers, on average and across countries and economies (59) (Table I.B1.3.45).

Given the proactive nature of such self-regulated learning attitudes, it is to be expected that such attitudes are closely related to students’ curiosity (Table I.B1.3.7). For example, students who engage in such activities are more likely to report liking to learn new things (Table I.B1.3.8). While among students with high self-regulating learning attitudes of help seeking (in the top quarter of the index), 78% of them reported such intrinsic drive to learn; among students with low self-regulated learning attitudes (in the bottom quarter of the index), about 59% did (Figure I.3.18). This is consistent across almost all countries and economies (Table I.B1.3.6).

These positive relationships suggest that students who are drawn to learning may be more likely to treat uncertainty or incomplete understanding as something to work through, and help seeking is one way of doing so. In this sense, help seeking does not appear as a sign of withdrawal from learning, but as one way in which interest in learning is actively sustained. Together with the earlier findings, this pattern indicates that students’ internal drive to learn is often connected not only to enjoyment and exploration, but also to a readiness to act on difficulties in constructive ways.

Likewise, students who show these self-regulated help-seeking attitudes are also most likely to report greater perseverance (Table I.B1.3.15). For example, on average, the share of students who report applying additional effort when work becomes challenging is significantly higher among those who report high self-regulated learning (in the top quarter of the index), than among those who report low self-regulated learning (in the bottom quarter of the index), by an average 24 percentage-point difference. This is consistent across almost all countries and economies (Table I.B1.3.14).

This strong positive relationship with perseverance adds an important nuance to this picture: seeking support is not a sign that students lack persistence, but rather one way in which persistence is expressed. Students who look for feedback or guidance when needed also tend to be those who continue working through difficulty.

This makes growth mindset a natural next step, since it captures the belief that effort, feedback and adjustment can lead to improvement. Having a growth mindset is one of those attitudes that align with students’ help- and feedback seeking. A growth mindset is an attitude that does not work in isolation and cannot be reduced to effort alone. It is also about identifying challenging situations that require adequate feedback (OECD, 2025[25]). PISA data show that the relationship between self-regulated attitudes like the ones analysed here relate positively with students reporting a growth mindset. On average and across countries and economies, students who engage in these self-regulated help-seeking attitudes are more likely to report a growth mindset (Table I.B1.3.25).

Perhaps the clearest expression of how these attitudes complement one another, is the willingness to respond to challenge by first trying to work through it and then seeking feedback if needed. As shown in Figure I.3.19, PISA data show that students with a growth mindset are more likely to report only asking for help after trying hard to find the solution on their own (Table I.B1.3.25). Indeed, on average and in most countries and economies (53), more students with a growth mindset report this, showing both perseverance and the flexibility to ask for feedback when needed (Table I.B1.3.48).

Moreover, students who report stronger self-regulated help-seeking attitudes are also more likely to hold more developed beliefs about the value of rigorous scientific knowledge and evidence-based information. This is an important link, as navigating one’s own learning with autonomy also requires an understanding of how facts are established and how they differ from unsubstantiated information. In this sense, student agency is not only reflected in how learners organise, adapt or seek for help, but also in how they recognise and value reliable forms of knowledge. For example, students with stronger self-regulated help-seeking learning attitudes are more likely to report that they agree that information from one source should be checked against a range of other reliable sources before it is accepted, on average and across almost all countries and economies (Table I.B1.3.35).

Across OECD countries, 85% of students who report high self-regulated help seeking (in the top quarter of the index) report they agree that information from one source should be checked from a range of other reliable sources before it is accepted, on average. On the opposite end, 78% of students who report low self-regulated help seeking agree with this statement. While in 13 education systems the share of student with high self-regulated learning attitudes who agree with source checking is over 90%, in Armenia, Paraguay, Bulgaria, El Salvador, Guatemala and the Dominican Republic (in ascending order), more than 30% of such students disagree with this basic notion of source checking (Table I.B1.3.34).

These results suggest that students who are more likely to seek feedback or guidance when needed also tend to be more likely to view abilities as developable and to value knowledge that is evidence-based and open to scrutiny. Help seeking may not be simply a practical strategy, but part of a wider orientation towards learning in which difficulties are approached as manageable, improvement is seen as possible, and knowledge is understood as something that must be examined and justified. Help seeking thus appears not only as a self-regulated practice, but also as part of a wider cognitive orientation towards learning.

Goal setting represents one of the clearest expressions of student agency in learning. It refers not only to setting objectives, but to organising learning in ways that allow students to monitor progress, identify gaps in understanding, and adjust their strategies when needed. Planning, checking one’s own understanding, making summaries, and revising one’s approach are all part of this process. In this sense, goal setting lies at the centre of self-regulated learning, as it is embedded in a broader cycle of planning, monitoring and adjusting learning strategies through which students actively direct their work rather than simply responding to immediate demands (Pintrich, 2004[29]; Cleary and Zimmerman, 2012[8]; Zimmerman, 2000[30]).

Goal setting is not a widespread dimension of student agency. Across OECD countries, slightly over one-third of students (37%) reported planning their approach to learning often or very often (Figure I.3.20). Also, slightly over one-third (35%) reported making summaries of the most important points without using their documents in order to identify gaps in their knowledge. Slightly more students, almost half (49%) do often or very often ask themselves questions about the material to check if they have understood everything and 44% change their approach to studying if they notice it is not successful, on average (Table I.B1.3.49).7

As shown in Figure I.3.20, girls tend to report higher goal setting attitudes than boys. In no country/economy do PISA data show the inverse and in 12 systems, there are no differences between boys and girls. Likewise, when looking at students’ socio-economic status, on average and across participating countries and economies, more advantaged students report more goal setting attitudes than their disadvantaged peers. Only in Kenya and Peru, there are no differences by students’ socio-economic status (Table I.B1.3.50).

Importantly, the individual drive to learn can be a factor behind these forms of agency. PISA data show that curiosity is consistently positively related to goal setting, and that students who report using these planning and self-verification techniques most often are also those who report being intrinsically motivated to learn (liking to learn new things), twice as much on average across the OECD, after accounting for students’ and schools’ socio-economic profile (Tables I.B1.3.7 and I.B1.3.8). As shown in Figure I.3.21, on average across the OECD, 83% of students with more goal-setting attitudes reported liking to learn new things, while the share is 57% for students with less goal-setting attitudes. The gap with respect to students with low goal-setting attitudes is quite large, at 26 percentage points.

This positive relationship suggests that students’ internal drive to learn is more likely to be translated into organised study behaviour when it is accompanied by a clearer sense of direction. In this sense, self-regulated goal setting could therefore be read as one of the mechanisms through which students’ motivation becomes organised rather than remaining diffuse or episodic.

The relationship between the internal concept of motivation and proactively taking action to lead one’s own learning can be also dependent on other factors. PISA 2025 data suggest that teachers have a role to play. For example, the relationship between this form of student agency and reporting a curiosity to learn is stronger among students who report receiving support from their teachers. As shown in Figure I.3.22, on average and in 38 systems, the relationship between student motivation and agency is strengthened by high teacher support, after accounting for gender, and students' and schools' socio-economic profile (Table I.B1.3.52).

That teacher support may positively moderate this relationship suggests that motivation alone does not equally translate into goal setting across contexts. In some systems, where students perceive greater teacher support, their motivation could be more closely aligned with organised and self-regulated study behaviour. This reinforces the notion that agency should not be understood as operating in isolation from context, but as something that may be strengthened when students are supported in turning motivation into purposeful action.

Moreover, as these attitudes can be considered a proxy for student agency in learning, is not surprising that PISA 2025 data show that they are consistently related to student perseverance, after accounting for students’ and schools’ socio-economic profile (Table I.B1.3.15). This gives self-regulated goal setting a distinctive place as a more strategic side of agency: it is the point at which interest and effort can be channelled into a planned course of action. Indeed, these relationships suggest that students’ engagement is strongest not simply when they want to learn or when they keep trying.

What, then, does this form of agency imply about students’ beliefs about their own abilities? PISA data show that, on average across OECD countries, students who report this type of agency are more likely to report a growth mindset, after accounting for students’ and schools’ socio-economic profile. However, this relationship is far from universal. In 22 participating systems, the inverse is true and students who report this sort of agency are also likely to report a fixed mindset. Yet, in over one-third of countries and economies, there is no direct relationship between the two (Table I.B1.3.25).

However, when examining individual practices, different patterns emerge. For example, across participating countries and economies, students who report that they often ask themselves questions about the materials they study to check if they have understood everything, also are more likely to report a growth mindset. Only in five systems, the opposite is true (Table I.B1.3.25). In contrast, when students report that they often make a summary of the most important points without using their documents in order to identify knowledge gaps, the association with growth mindset is largely non-existent (as on average) and in 28 systems, such students are more likely to report a fixed mindset (Table I.B1.3.25).

Most likely, self-regulated goal setting does not represent a single form of agency with the same meaning across students. Some behaviours may reflect more proactive and future-oriented forms of self-regulation, which are more closely aligned with the belief that abilities can improve through effort. Others may reflect more immediate checking or corrective strategies, which can be used for different reasons and not necessarily as part of a broader growth-oriented view of learning. The mixed pattern does not necessarily weaken the importance of goal setting, but rather shows that its different components may be connected to students’ beliefs about improvement in distinct ways.

Overall, these findings suggest that self-regulated goal setting is a key strategic dimension of student agency. It is positively related to curiosity and perseverance, indicating that students’ motivation is more likely to take organised form when it is translated into planning, monitoring and adjustment. The relationship with student drive to learn is also stronger where teacher support is higher, underlining that agency depends partly on the conditions in which students learn. At the same time, the mixed relationship with growth mindset shows that goal setting is not a single, uniform form of engagement: its different practices may carry different meanings for different students. Goal setting is therefore best understood as a set of self-regulated behaviours through which students organise, monitor and adapt their learning, but whose meaning may vary across students and situations.

This section shifts the focus to the affective dimension of the student experience (Fredricks, Blumenfeld and Paris, 2004[1]), specifically exploring social engagement through the lens of students’ sense of belonging at school and their views on the value of school. Belonging encompasses the extent to which students feel personally accepted, respected and included within the school community, contrasting directly with feelings of loneliness, social isolation or feeling like an outsider (Osterman, 2000[33]; Voelkl, 1996[34]). While much of the literature highlights the overt struggles of students who exhibit both low academic performance and report a low sense of belonging, it is equally critical to recognise that emotional disengagement can occur silently among students with average or even high academic performance (Janosz et al., 2008[35]). A student may fulfil the procedural requirements of schooling, complete their work, and maintain average grades, yet still feel disconnected from school (Finn, 1989[36]; Janosz et al., 2000[37]).

Although a student's sense of value of school and belonging may not be a direct outcome of academic instruction alone – stemming instead from the broader social climate, peer relationships and interpersonal support – it is linked to the motivational and cognitive dimensions discussed in previous sections of this chapter (Pianta, 1999[38]; Wentzel, 2004[39]). When students feel school has value, they belong and are making friends easily, this emotional security satisfies a fundamental need for relatedness, acting as an affective catalyst that nurtures intrinsic and instrumental motivation, curiosity and the development of a growth mindset (Deci and Ryan, 2000[5]; Furrer and Skinner, 2003[40]).

As shown in Table I.3.1, students’ sense of belonging at school is positively and consistently aligned with curiosity and perseverance, and these relationships are strengthened by teacher support. It is also linked to the dimensions of agency, goal setting and help seeking, as well as to students’ beliefs about the nature of science and growth mindset. Sense of belonging and school value also tend to align with one another, whereas school value and curiosity often move in opposite directions.

A sense of belonging at school captures one of the most basic affective dimensions of student engagement: whether students experience school as a place where they feel accepted, included and socially connected. In practice, this concerns whether students feel that they belong at school, whether they can make friends easily, and whether they are generally free from feeling like outsiders, awkward or lonely in the school environment. In this sense, belonging is not simply a social outcome that sits alongside learning. It reflects the emotional conditions under which students experience school and, therefore, the extent to which the school environment provides a sense of interpersonal security and recognition.

PISA 2025 data show that a large majority of students do report feeling they belong at school, and it is a place where they make friends. On average across OECD countries, most students agree they feel like they belong at school (76%), other students seem to like them (83%) and that they make friends easily at school (80%). A majority also disagrees that they feel like an outsider (or left out of things) at school (85%), feel awkward and out of place (81%) and feel lonely at school (87%) (Table I.B1.3.62). Across countries and economies there is some variation, as shown in Figure I.3.24.

Boys tend to report significantly higher sense of belonging at school than girls, on average across OECD countries and in most countries and economies (63) (Figure I.3.24). Only in Dushanbe (Tajikistan), Kazakhstan, the Palestinian Authority and Uzbekistan, girls report a higher sense of belonging at school than boys. Likewise, socio-economically advantaged students report a higher sense of belonging at school than their disadvantaged peers, on average and in most countries and economies. Kenya is the only country where the opposite is true.

Despite this persistent gender and socio-economic differences, trends over time point to a broad improvement in students’ sense of belonging. Between 2022 and 2025, students’ sense of belonging at school increased in most participating countries and economies with comparable data for both cycles (Table I.B1.3.63). Moreover, in many systems (39), more students reported that they feel they belong at school than students did in 2022, and also that they make friends easily at school (64) (Table I.B1.3.64). Likewise, the share of students who do not feel lonely at school (i.e. who disagreed or strongly disagreed with that statement) increased in most systems (48).

This broad improvement is particularly noteworthy given the close relationship between sense of belonging and other positive attitudes towards learning. PISA data show a positive relationship between students’ intrinsic motivation to learn (i.e. when they report that they like to learn new things) and their sense of belonging at school, on average and across all countries and economies, after accounting for gender and students’ and schools’ socio-economic background (Table I.B1.3.68). Likewise, there is a positive relationship between sense of belonging and student curiosity across all participating countries and economies, after accounting for gender and students’ and schools’ socio-economic background (Table I.B1.3.68).

Moreover, students who report a high sense of belonging at school also are more likely to believe school has a use, for example to think that school has taught them things that can be useful in a job, on average and across most countries and economies. This is one dimension of the value of school that will be analysed next. However, due to its close relationship with students’ sense of belonging at school, it is interesting to note that, on average, 71% of students who report a high sense of belonging (in the top quarter of the index) believe school has taught them things that can be useful in a job, while 60% of students with low sense of belonging (in the bottom quarter) do (Figure I.3.25 and Table I.B1.3.67). In B-S-J-Z (China) and the Ukrainian regions (17 of 27), the difference is over 20 percentage points.

PISA data show a positive relationship between sense of belonging and student perseverance across all participating countries and economies, after accounting for gender and students’ and schools’ socio-economic background (Table I.B1.3.68). About 68% of students with high sense of belonging (in the top quarter of the index) report they apply additional effort when work becomes challenging, while about 51% who report low sense of belonging do. The difference between the two can be as large as 29 percentage points, as in B-S-J-Z (China) and Kenya (Figure I.3.26 and Table I.B1.3.67).

Feeling socially connected at school is aligned not only with wanting to learn, but also with continuing to engage when schoolwork becomes difficult. Yet, belonging does not operate in a vacuum. Whether feeling connected to school is actually reflected in greater curiosity and stronger perseverance may also depend on the extent to which students experience their teachers as supportive.

Interestingly, on average and in a large number of countries and economies (40 and 37), the relationship between students’ motivational and resilient attitudes, and sense of belonging at school is moderated by their teachers’ support, which includes students’ perception of teachers’ attitudes in the classroom (whether the teacher shows an interest in every student’s learning, gives extra help when students need it, helps students with their learning, continues teaching until the students understand and gives students an opportunity to express opinions) (Figure I.3.27 and Table I.B1.3.69). Both curiosity and perseverance are more strongly related to sense of belonging in school when students report high teacher support with respect to their peers who report low teacher support (Table I.B1.3.69).

The fact that teacher support strengthens the relationship between belonging and curiosity, and between belonging and perseverance, suggests that emotional connection to school is more closely linked to active engagement where students also feel supported by teachers in their learning environment. Where students feel both that they belong and that teachers are available, encouraging and responsive, they are not only more likely to be interested in learning, but also more likely to sustain effort when learning becomes difficult. Thus, engagement should not be understood as something students must generate on their own. A sense of belonging appears to matter most where the school environment, and especially teacher support, helps make that belonging meaningful for learning as well as for social inclusion.

Whether students also hold beliefs that make challenge itself more manageable is also relevant. This is where growth mindset becomes important. Students with a stronger sense of belonging are more likely to report having a growth mindset, on average and in most countries and economies (Table I.B1.3.67). In some systems, however, the relationship could be driven by student performance in science (Table I.B1.3.68). On average, among students with high sense of belonging (those in the top quarter of the index) 73% reported a growth mindset, while 64% of those who reported a low sense of belonging (in the bottom quarter of the index) did so. Yet, the gap can be as large as 32 percentage points in B-S-J-Z (China).

Moreover, a similar positive relationship is also observed for the other two dimensions of cognitive engagement showing the positive alignment between these forms of engagement: students who feel that they belong at school are likely to be those who report more rigorous beliefs about the nature of science, as well as those who adapt when learning becomes demanding or unfamiliar (Table I.B1.3.68).

Finally, sense of belonging also relates positively with the two different dimensions of student agency discussed earlier: self-regulated help seeking and goal setting (Table I.B1.3.68). The positive relationship between sense of belonging and these dimensions of student agency suggests that social engagement is closely connected to how students manage their learning in practice. Students who feel that they belong at school are more likely to report seeking support when needed and setting goals and organising their work. This indicates that belonging is not only associated with feeling socially included, but also with forms of agency through which students actively navigate their learning. In this sense, belonging may be understood as part of the relational and emotional foundation on which students are better able to ask for help, regulate their work and respond flexibly to challenge.

These findings suggest that sense of belonging at school is not a peripheral aspect of school life, but one of the emotional conditions through which engagement with learning is sustained. Its positive relationships with motivation, perseverance, growth mindset and the different forms of student agency examined in this chapter indicate that feeling accepted, included and socially at ease at school is closely aligned with how students approach learning more broadly. This reinforces an important point raised throughout these analyses. What matters is not the image of a particularly resilient or self-driven student, but whether schools provide environments in which students can feel socially recognised, emotionally secure and able to engage. In this sense, belonging helps shift the focus from individual dispositions alone to the relational conditions that make engagement more possible for all students.

While the intrinsic motivation analysed so far is closely linked to curiosity and refers to engaging in learning for its inherent interest, enjoyment or satisfaction, instrumental motivation refers to the perceived importance or future usefulness of academic work and schooling (Deci, 2013[9]; Eccles and Wigfield, 2002[10]). The analysis now turns to students’ views of the value of school, as one expression of instrumental motivation to learn. In PISA, this dimension is examined through whether students believe that school teaches things that are useful for future employment or adult life, or whether they see school as a waste of time.

PISA data show that, across OECD countries, 67% of students believe that school has taught them things that could be useful in a job. Cross-country differences remain substantial, however (Table I.B1.3.70). Moreover, 51% of students, on average across OECD countries, reported that school has prepare them for adult life, while 24% reported that school has been a waste of time (Table I.B1.3.70).

Taken together, students’ views of school’s usefulness remain positive in many systems, but are far from uniform and, in a sizeable number of cases, coexist with much more sceptical views of school’s value.

Trends over time present a mixed picture. As shown in Figure I.3.28, in 11 systems the share of students reporting that school has prepare them for adult life decreased between 2022 and 2025. The decrease was largest – by more than 5 percentage points – in Austria, Czechia, Macao (China) and Poland where fewer than half of students reported disagreeing with this statement in 2025. In 36 systems, however, the opposite pattern is observed: the share of students who disagreed with this negative view of school is larger in 2025.

A similar pattern can be seen for the statement that school has been a waste of time. In Costa Rica, Czechia, Denmark, Germany, Latvia, Macao (China), the Netherlands* and Chinese Taipei, the share of students reporting this increased slightly, by between 2 and 4 percentage points, between 2022 and 2025. Even in these systems, however, at least 65% of students reported disagreeing or strongly disagreeing with the statement in 2025. In 40 countries and economies, the share of students disagreeing or strongly disagreeing that school has been a waste of time increased over the same period (Table I.B1.3.71). These trends suggest that students’ instrumental views of school did not move uniformly between the two PISA cycles: in many systems they became more positive, but in others a more sceptical view of school’s usefulness also gained ground.

As with other indicators, average levels can conceal meaningful differences across groups of students. On average across OECD countries, girls reported more often than boys that school has taught them things which could be useful in a job, with an average gap of 7 percentage points. This pattern is relatively consistent across participating countries and economies, with gaps ranging from about 15 to 17 percentage points in Jordan, Kosovo and the Palestinian Authority to less than 4 percentage points in Australia, the Ukrainian regions (17 of 27) and the United Kingdom. In eight participating systems, there are no differences between boys’ and girls’ reports (Table I.B1.3.73). Socio-economic differences are less consistent. Although socio-economically advantaged students are more likely, on average across OECD countries, to report this instrumental view of school, the gap is much smaller, at 2 percentage points. A larger share of advantaged students reported this view in 24 countries and economies, while in 15 systems the opposite pattern is observed and in 48 systems, there are no differences (Table I.B1.3.73). These results suggest that students’ views of school’s usefulness are not distributed evenly across groups, but that gender differences are more systematic than socio-economic differences.

Students’ views of the value of school also connect to whether motivation is sustained when learning becomes demanding. Across participating countries and economies, students who see school as useful for their future – both for adult life and for a job – also tend to report greater perseverance, on average. Less positive views of school are generally associated with weaker perseverance (Tables I.B1.3.15 and I.B1.3.76).

This places instrumental motivation alongside curiosity and perseverance as part of a broader picture of students’ motivation to learn. Curiosity captures students’ interest in learning, while instrumental motivation captures whether they see school as worthwhile and useful for their future. Perseverance adds a behavioural dimension by indicating whether these motivations are sustained when learning becomes challenging. In this respect, engagement depends not only on the interest generated by learning, but also on whether students see the effort involved in schooling as worth sustaining.

Students’ views of the value of school are also linked to growth mindset. On average across OECD countries and in most participating countries and economies, the share of students reporting a growth mindset is considerably lower among those who agree that school has done little to prepare them for adult life after they leave school than among those who disagree (Figure I.3.29). A similar pattern is observed for students who believe that school has been a waste of time: on average, the share of students reporting a growth mindset is 14 percentage points higher among those who disagree with this statement than among those who agree. (Table I.B1.3.20).

Interestingly, the relationship is much weaker with students’ cognitive adaptability. For example, only in B‑S‑J‑Z (China), Finland, Hong Kong (China), Kenya, Korea, Mongolia and New Zealand* students who report greater cognitive adaptability are more likely to believe that school has taught them things that could be useful in a job. In another 14 systems, the opposite is true and, in the majority, there is no relationship (Table I.B1.3.78). Students’ views of school’s usefulness are more closely related to certain attitudes related to effort and adaptability, like a growth mindset, but less so with overall adaptability.

The relationship with self-regulated help seeking is positive and consistent. On average and across countries and economies, students who report stronger help-seeking attitudes are more likely to report that school has taught them things that could be useful in a job, with the sole exception of the Dominican Republic, where the opposite is true (Table I.B1.3.78). Indeed, on average, 71% of students with high self-regulated learning attitudes report this belief, compared to about 58% of those with low self-regulated learning attitudes (Table I.B1.3.77).

Goal setting is also strongly aligned with this instrumental view of school. Among students who report stronger goal-setting attitudes, the share of those who consider that school has taught them things that could be useful in a job is considerably higher than among students who report less of this form of agency. The average gap is 15 percentage points and it is consistently observed across all participating countries and economies, with the sole exception of Indonesia (Table I.B1.3.77).

Overall, these findings suggest that students’ views of the value of school are an important part of student engagement, complementing the earlier analysis of curiosity and perseverance. They are also closely linked to whether students sustain effort, hold more growth-oriented beliefs, and engage in more organised and self-regulated ways of learning. At the same time, these relationships are not uniform across all dimensions, and are notably weaker or less consistent for cognitive adaptability. Students’ views of the value of school also vary considerably across systems, show uneven movement over time, and are not distributed equally across groups of students. This matters because students may continue to participate and meet school expectations without equally strong beliefs that school is worthwhile or worth investing in. The next section builds on this possibility by examining how academic functioning and motivational engagement do not always align in the same way across groups of students.

The findings presented in this chapter suggest that, across education systems, some students may reach the baseline performance level, while not believing themselves to be strongly engaged in motivational terms. They may attend school and meet at least minimum academic expectations, yet report weaker curiosity, lower perseverance or less positive views about the value of school. Their attendance and participation in school may resemble compliance, understood here as meeting expectations without necessarily making a deeper personal investment in learning (Saeed and Zyngier, 2012[41]). This interpretation is consistent with the idea that school participation may be supported by motives that are not fully internalised (Deci and Ryan, 2000[5]), and that acceptable academic functioning does not necessarily coincide with strong interest in learning or with a strong sense that schoolwork is useful and worthwhile (Eccles and Wigfield, 2024[42]).

PISA data show that in general, academic performance, curiosity and perseverance go together (Tables I.B1.3.5 and I.B1.3.13), but these attitudes of engagement are not necessarily widespread, even among students who reached the baseline level of performance. For example, over one-third of students performing at baseline proficiency in science disagree or strongly disagree that they like learning new things, and almost half do so about applying additional effort when work becomes challenging. In several systems, the share of such students reporting these dimensions of curiosity or perseverance are half or less. This is particularly relevant considering that in science, the average student in OECD countries performs at the upper end of such baseline proficiency (482 score points on average, which is within the Level 2 range of 410 to 484 score points).

When it comes to students’ perceived value of school, its relationship with performance is less clear. For example, as shown in Figure I. 3.30, the share of students who believe that what school has taught can be useful for adult life after school is often similar across proficiency levels and in many countries and economies even reversed. On average across OECD countries, about half of students report this positive belief about school, independently of where they stand in the proficiency scale (Table I.B1.3.75).

In other words, not all students reaching baseline proficiency level report motivational engagement. Performance alone may therefore give an incomplete picture of how students relate to learning and school. This is particularly important in light of the declines observed between 2022 and 2025 in some indicators of students’ curiosity and perseverance, as well as the average declines in reading and mathematics performance reported in Chapter 2.

For example, higher curiosity does not automatically coincide with a stronger sense of school’s practical value. Students are more likely to perceive schooling as valuable when they are less curious rather than more curious on average across OECD countries and in about half of countries and economies. The opposite is observed only in seven countries and economies (Figure I.3.31 and Table I.B1.3.76).

However, this kind of mismatch is not observed between the value of school and perseverance. Students who disagree that school has done little to prepare them for adult life when they leave school tend also to be those who report higher perseverance (Table I.B1.3.76). It is possible that perseverance may be sustained more readily than curiosity where students see effort as serving goals, expectations or future benefits: the practical value of school may not always spark curiosity, but it may still help sustain effort.

Student academic compliance without engagement is most visible among students who perform at the proficiency baseline Level 2. However, PISA 2025 data show that this may also apply to a non-negligible share of students performing at Level 3, as shown in Figure I.3.32. On average across OECD countries, about 36% of students performing at Level 2 report disagreeing they like to learn new things, and 29% of those performing at Level 3 do so (Table I.B1.3.5). Similarly, about 44% of students performing at Level 2 report disagreeing they apply additional effort when work becomes challenging, and 38% among those performing at Level 3 do (Table I.B1.3.13).

Moreover, on average across OECD countries, about 35% of students performing at Level 2 in science report that they do not believe that school has taught them things that could be useful in a job, and about 32% of those performing at Level 3 do so (Table I.B1.3.75). Likewise, about 27% of students performing at the baseline Level 2 in science report that school has been a waste of time, compared with about 21% of those performing at Level 3 (Table I.B1.3.75). Almost half of students performing at the baseline Level 2 (49%) report that school has done little to prepare them for adult life when they leave school, as do about half of those performing at Level 3 (50%) (Table I.B1.3.75).

These results do not imply that the same students report all of these attitudes at once. Rather, they indicate that an important share of students who reach at least baseline levels of performance do so without strongly expressing the attitudes that may help sustain engagement more broadly. In this sense, academic compliance can be understood as participation that is sufficiently effective to meet school expectations, but not necessarily accompanied by a correspondingly strong investment in learning itself. For some students, acceptable performance may coexist with weaker curiosity; for others, it may coexist with less positive views about the usefulness of school, even where perseverance or other motivations remain more intact.

A multidimensional perspective is therefore particularly relevant in light of the performance trends discussed in Chapter 2. Between 2022 and 2025, performance declined across the performance distribution in reading and mathematics, and while it remained stable for science, the performance gap has widened and the mean scores of the lowest-performing students are the lowest reported so far by PISA. In this context, students whose scores remain around the average, or who continue to meet baseline proficiency expectations, may still appear academically compliant when assessed on performance alone, even when they do not report interest in learning, sense of purpose and willingness to persist. Such students sit at the limit where the loss of academic competencies can quickly deteriorate their future perspectives. This is what makes the notion of compliance analytically useful in this chapter: it highlights a possible gap between observable academic functioning and the motivational conditions that help sustain engagement over time, particularly among students who are at most academic risk.

The analyses developed in this chapter have treated students’ motivation, beliefs, agency and sense of belonging as outcomes of schooling in their own right, not simply as instruments for improving academic results. To complete the picture, the final section examines the extent to which these dimensions of engagement are also associated with science performance in PISA.

Across many systems, the different dimensions of engagement discussed in this chapter are positively associated with science performance, but not all in the same way or with the same consistency. As shown in Table I.3.1, some, such as curiosity and students’ beliefs about the nature of science, show clearer and more direct alignment with stronger academic outcomes, while others, such as goal setting, display more differentiated or item-specific patterns. The dimensions of agency are also important because some of them strengthen the extent to which motivation and perseverance are associated with performance. More broadly, these patterns suggest that stronger academic outcomes and stronger engagement often go together, although average levels of engagement and the strength of their relationship with performance vary across education systems.

PISA 2025 data show that curiosity to learn is positively related to science performance, after accounting for students’ and schools’ socio-economic profile. This positive association is observed in all participating countries and economies in PISA 2025 with available data. Its strength, however, varies across systems. Countries and economies with the highest average levels of curiosity are not necessarily those where curiosity is most strongly associated with science performance (Table I.B1.3.4).

Conversely, in countries and economies where the association between curiosity and performance is strongest, students’ reported curiosity often lies closer to the OECD average, as in Australia, Iceland, the Philippines and Qatar, or below it, as in the Dominican Republic, Kenya, Lebanon, Luxembourg and Rwanda. This suggests that the relationship between students’ drive to learn and academic outcomes depends not only on to what extent students report being curious, but also on the extent to which that curiosity is channelled into academically relevant forms of engagement within each education system (Tables I.B1.3.1 and I.B1.3.4).

A useful complement to this picture is to examine the relationship between perseverance and academic performance. This provides an indication of how closely sustained effort and academic outcomes are aligned across students. PISA 2025 data show that students with higher levels of perseverance tend to outperform those with lower levels. This positive association is observed in all participating countries and economies in PISA 2025 with available data, with the sole exceptions of B-S-J-Z (China) and Chinese Taipei (Table I.B1.3.12).

As with curiosity, however, the strength of this association varies across education systems. Countries and economies with the highest average levels of perseverance are not necessarily those where perseverance is most strongly associated with science performance. In many of the countries and economies where perseverance is most strongly associated with performance, however, average perseverance remains below the OECD average, as in Denmark, the Dominican Republic, Greece, Norway* and Rwanda(Tables I.B1.3.9 and I.B1.3.12). This suggests that stronger association between perseverance and performance does not necessarily coincide with higher average levels of perseverance. Rather, it may indicate that in some systems, perseverance is less widely shared but more strongly differentiates performance across students. More broadly, the pattern suggests that average levels of engagement and the extent to which engagement is linked to academic performance should be considered separately, as the two do not necessarily correlate.

As in previous PISA cycles, reporting a growth mindset is positively associated with academic performance. PISA 2025 data show that students reporting a growth mindset score, on average, 28 points higher in science than students who do not report this belief, after accounting for students’ and schools’ socio-economic profile (Table I.B1.3.18).

However, previous PISA analyses have shown that the score gaps associated with growth mindset are not uniform across proficiency levels: students who scored at proficiency Level 2, 3 or 4 (i.e. mid-performers) show the largest and most consistent gaps with respect to low and top performers (OECD, 2025[25]) .8

Moreover, students who hold more developed beliefs about the nature of science also tend to show stronger academic outcomes. PISA data show that science performance tends to be stronger where students show greater recognition that scientific knowledge is evidence-based, testable and open to scrutiny. This positive relationship is observed consistently across all countries/economies participating in PISA 2025 (Table I.B1.3.33). Likewise, the association with performance in science is similar when looking at students’ responses to the question on whether one should rely more on common sense and less on scientific studies. Among all participating countries/economies with available information, there is a positive association between students’ disagreement with this statement and performance in science (Table I.B1.3.33).

Cognitive adaptability also shows a positive average relationship with science performance, although moderate (Table I.B1.3.43).

The dimensions of student agency examined in this chapter – self-regulated help seeking and goal setting – are related to academic performance, but not in the same way.

Self-regulated help seeking is associated with science performance in a majority of countries and economies, after accounting for students' and schools' socio-economic profile (63 out of 82) (Table I.B1.3.46). This dimension of agency also matters because it strengthens the relationship between important aspects of engagement and academic performance. For example, the positive relationship between students’ liking to learn new things and their science performance is stronger among students who report high self-regulated help seeking than among those who report low self-regulation, by 13 score points on average and in 42 countries and economies; in no system is the opposite true (Table I.B1.3.47).

This suggests that agency matters not only because it is directly associated with performance, but also because curiosity is more closely aligned with stronger academic outcomes where students are better able to seek support.

Goal setting presents a more mixed picture. On average across the OECD, and in two-thirds of systems, goal setting is positively, but weakly, related to science performance. However, in 27 countries and economies, there is no relationship with performance in science, and in Czechia, Estonia, Italy and Latvia, the relationship is negative (Table I.B1.3.51). This weaker and less consistent pattern reflects the fact that the questions measuring goal setting do not all carry the same meaning. For example, in countries where the relationship is negative, there is a strong negative association between performance in science and “students often making a summary of the most important points without using their documents in order to identify knowledge gaps”. By contrast, in three of these four countries, there is a positive association between science performance and “often asking oneself questions about study materials in order to check understanding” (Table I.B1.3.51).

It is possible that in a number of systems, low-performing students are encouraged to use specific techniques to verify their understanding and learning needs, such as making summaries of the most important points without using reference documents. This suggests that goal setting may not be best understood as a single marker of stronger engagement, but as a set of practices whose meaning depends partly on how and why they are used. Some may be more common among higher-performing students because they reflect a proactive and structured way of managing learning. Others may be more visible among lower-performing students because they are used in a more compensatory way to check understanding, identify gaps or revise strategies more frequently.

Students’ sense of belonging at school and their views of the value of school are both positively associated with science performance, but not in the same way. The index of sense of belonging at school is moderately but positively associated with science performance, on average and in most countries and economies (46), after accounting for students’ and schools’ socio-economic background (Table I.B1.3.66). However, there is no relationship in 31 systems, and the association is negative in Finland, Italy, the Netherlands*, Norway* and Chinese Taipei.

Students’ views of the value of school are also positively associated with science performance. In line with earlier analyses using PISA data (OECD, 2024[43]), students who believe that school has taught them things which could be useful in a job tend to outperform those who believe the opposite in all participating countries and economies, with the sole exceptions of Germany, Hong Kong (China), Kyrgyzstan, Macao (China), Chinese Taipei and Türkiye (Table I.B1.3.74). Similarly, students who do not believe that school has been a waste of time tend to outperform those who do in science, on average across OECD countries. This pattern is observed in all participating countries and economies in PISA 2025, with the sole exception of Türkiye, after accounting for students’ and schools’ socio-economic profile (Table I.B1.3.74).

These relationships suggest that students’ experience of school matters for academic performance not only in terms of whether they feel that they belong, but also whether they see school as worthwhile and relevant. At the same time, the two dimensions do not operate identically. Together with the earlier analyses of curiosity, perseverance and agency, these findings help shift attention from students’ internal orientations alone to the extent to which school is experienced as both socially meaningful and worth investing in.

[2] Appleton, J., S. Christenson and M. Furlong (2008), “Student engagement with school: Critical conceptual and methodological issues of the construct”, Psychology in the Schools, Vol. 45/5, pp. 369-386, https://doi.org/10.1002/pits.20303.

[7] Assor, A. (2012), “Allowing Choice and Nurturing an Inner Compass: Educational Practices Supporting Students’ Need for Autonomy”, in Handbook of Research on Student Engagement, Springer US, Boston, MA, https://doi.org/10.1007/978-1-4614-2018-7_20.

[31] Blackwell, L., K. Trzesniewski and C. Dweck (2007), “Implicit Theories of Intelligence Predict Achievement Across an Adolescent Transition: A Longitudinal Study and an Intervention”, Child Development, Vol. 78/1, pp. 246-263, https://doi.org/10.1111/j.1467-8624.2007.00995.x.

[8] Cleary, T. and B. Zimmerman (2012), “A Cyclical Self-Regulatory Account of Student Engagement: Theoretical Foundations and Applications”, in Handbook of Research on Student Engagement, Springer US, Boston, MA, https://doi.org/10.1007/978-1-4614-2018-7_11.

[15] Corno, L. and E. Mandinach (1983), “The role of cognitive engagement in classroom learning and motivation”, Educational Psychologist, Vol. 18/2, pp. 88-108, https://doi.org/10.1080/00461528309529266.

[9] Deci, E. (2013), Intrinsic Motivation and Self-Determination in Human Behavior, Springer Science & Business Media, https://doi.org/10.1007/978-1-4899-2271-7.

[5] Deci, E. and R. Ryan (2000), “The “What” and “Why” of Goal Pursuits: Human Needs and the Self-Determination of Behavior”, Psychological Inquiry, Vol. 11/4, pp. 227-268, https://doi.org/10.1207/s15327965pli1104_01.

[21] Dweck, C. (2013), Self-theories: Their Role in Motivation, Personality, and Development, Psychology Press, London, https://doi.org/10.4324/9781315783048.

[20] Dweck, C. (2006), Mindset: The New Psychology of Success, Random House, New York.

[22] Dweck, C. and E. Leggett (1988), “A social-cognitive approach to motivation and personality.”, Psychological Review, Vol. 95/2, pp. 256-273, https://doi.org/10.1037/0033-295x.95.2.256.

[18] Eccles, J. et al. (1999), Development During Adolescence: The Impact of Stage-Environment Fit on Young Adolescents’ Experiences in Schools and in Families, Routledge.

[42] Eccles, J. and A. Wigfield (2024), “The Development, Testing, and Refinement of Eccles, Wigfield, and Colleagues’ Situated Expectancy-Value Model of Achievement Performance and Choice”, Educational Psychology Review, Vol. 36/2, https://doi.org/10.1007/s10648-024-09888-9.

[10] Eccles, J. and A. Wigfield (2002), “Motivational Beliefs, Values, and Goals”, Annual Review of Psychology, Vol. 53/1, pp. 109-132, https://doi.org/10.1146/annurev.psych.53.100901.135153.

[27] Egan, J. et al. (2026), Reuters Institute Digital News Report 2026, Reuters Institute / University of Oxford, https://doi.org/10.60625/risj-4vgf-s811.

[36] Finn, J. (1989), “Withdrawing From School”, Review of Educational Research, Vol. 59/2, pp. 117-142, https://doi.org/10.3102/00346543059002117.

[3] Finn, J. and K. Zimmer (2012), “Student Engagement: What Is It? Why Does It Matter?”, in Handbook of Research on Student Engagement, Springer US, Boston, MA, https://doi.org/10.1007/978-1-4614-2018-7_5.

[1] Fredricks, J., P. Blumenfeld and A. Paris (2004), “School Engagement: Potential of the Concept, State of the Evidence”, Review of Educational Research, Vol. 74/1, pp. 59-109, https://doi.org/10.3102/00346543074001059.

[17] Frydenberg, E. (2005), Schooling issue digest: Student motivation and engagement, Australian Government Department of Education: Canberra, Australia.

[40] Furrer, C. and E. Skinner (2003), “Sense of relatedness as a factor in children’s academic engagement and performance.”, Journal of Educational Psychology, Vol. 95/1, pp. 148-162, https://doi.org/10.1037/0022-0663.95.1.148.

[11] Hidi, S. and K. Renninger (2006), “The Four-Phase Model of Interest Development”, Educational Psychologist, Vol. 41/2, pp. 111-127, https://doi.org/10.1207/s15326985ep4102_4.

[28] Hilton, M. and J. Pellegrino (2012), Education for life and work: Developing transferable knowledge and skills in the 21st century, National Academies Press.

[12] Izard, C. (2007), “Basic Emotions, Natural Kinds, Emotion Schemas, and a New Paradigm”, Perspectives on Psychological Science, Vol. 2/3, pp. 260-280, https://doi.org/10.1111/j.1745-6916.2007.00044.x.

[35] Janosz, M. et al. (2008), “School Engagement Trajectories and Their Differential Predictive Relations to Dropout”, Journal of Social Issues, Vol. 64/1, pp. 21-40, https://doi.org/10.1111/j.1540-4560.2008.00546.x.

[37] Janosz, M. et al. (2000), “Predicting different types of school dropouts: A typological approach with two longitudinal samples.”, Journal of Educational Psychology, Vol. 92/1, pp. 171-190, https://doi.org/10.1037/0022-0663.92.1.171.

[23] Kitchner, K. (1983), “Cognition, Metacognition, and Epistemic Cognition”, Human Development, Vol. 26/4, pp. 222-232, https://doi.org/10.1159/000272885.

[24] Martin, A. et al. (2012), “Adaptability: Conceptual and Empirical Perspectives on Responses to Change, Novelty and Uncertainty”, Australian Journal of Guidance and Counselling, Vol. 22/1, pp. 58-81, https://doi.org/10.1017/jgc.2012.8.

[25] OECD (2025), “Mindsets, attitudes and learning: Exploring their connections among students”, OECD Education Policy Perspectives, No. 133, OECD Publishing, Paris, https://doi.org/10.1787/8b1756bc-en.

[26] OECD (2025), Results from TALIS 2024: The State of Teaching, TALIS, OECD Publishing, Paris, https://doi.org/10.1787/90df6235-en.

[43] OECD (2024), PISA 2022 Results (Volume V): Learning Strategies and Attitudes for Life, PISA, OECD Publishing, Paris, https://doi.org/10.1787/c2e44201-en.

[33] Osterman, K. (2000), “Students’ Need for Belonging in the School Community”, Review of Educational Research, Vol. 70/3, pp. 323-367, https://doi.org/10.3102/00346543070003323.

[38] Pianta, R. (1999), Enhancing Relationships between Children and Teachers, American Psychological Association, Washington, https://doi.org/10.1037/10314-000.

[29] Pintrich, P. (2004), “A Conceptual Framework for Assessing Motivation and Self-Regulated Learning in College Students”, Educational Psychology Review, Vol. 16/4, pp. 385-407, https://doi.org/10.1007/s10648-004-0006-x.

[46] Rammstedt, B. and O. John (2007), “Measuring personality in one minute or less: A 10-item short version of the Big Five Inventory in English and German”, Journal of Research in Personality, Vol. 41/1, pp. 203-212, https://doi.org/10.1016/j.jrp.2006.02.001.

[45] Rammstedt, B. and T. Rammsayer (2002), “Gender differences in self‐estimated intelligence and their relation to gender‐role orientation”, European Journal of Personality, Vol. 16/5, pp. 369-382, https://doi.org/10.1002/per.454.

[4] Reeve, J. (2012), “A Self-determination Theory Perspective on Student Engagement”, in Handbook of Research on Student Engagement, Springer US, Boston, MA, https://doi.org/10.1007/978-1-4614-2018-7_7.

[6] Reeve, J. and C. Tseng (2011), “Agency as a fourth aspect of students’ engagement during learning activities”, Contemporary Educational Psychology, Vol. 36/4, pp. 257-267, https://doi.org/10.1016/j.cedpsych.2011.05.002.

[44] Robins, R. et al. (2001), “Personality Correlates of Self-Esteem”, Journal of Research in Personality, Vol. 35/4, pp. 463-482, https://doi.org/10.1006/jrpe.2001.2324.

[41] Saeed, S. and D. Zyngier (2012), “How Motivation Influences Student Engagement: A Qualitative Case Study”, Journal of Education and Learning, Vol. 1/2, https://doi.org/10.5539/jel.v1n2p252.

[13] Schunk, D., P. Pintrich and J. Meece (2014), Motivation in Education: Theory, Research, and Applications (4th ed.), Pearson Education.

[16] Skinner, E. and M. Belmont (1993), “Motivation in the classroom: Reciprocal effects of teacher behavior and student engagement across the school year”, Journal of educational psychology, Vol. 4/85, pp. 571-581.

[14] Skinner, E. et al. (2009), Engagement and Disaffection as Organizational Constructs in the Dynamics of Motivational Development, Routledge/Taylor & Francis Group.

[34] Voelkl, K. (1996), “Measuring Students’ Identification with School”, Educational and Psychological Measurement, Vol. 56/5, pp. 760-770, https://doi.org/10.1177/0013164496056005003.

[32] Wang, M. et al. (2026), “Using a Robust Moderated Mediation Approach to Understand Associations Between Classroom Quality, Mindset Beliefs, Engagement, and Student Achievement in Math and Science”, American Educational Research Journal, https://doi.org/10.3102/00028312261435545.

[39] Wentzel, K. (2004), “Understanding classroom competence: The role of social-motivational and self-processes”, in Advances in Child Development and Behavior, Advances in Child Development and Behavior Volume 32, Elsevier, https://doi.org/10.1016/s0065-2407(04)80008-9.

[19] Wolters, C. and D. Taylor (2012), “A Self-regulated Learning Perspective on Student Engagement”, in Handbook of Research on Student Engagement, Springer US, Boston, MA, https://doi.org/10.1007/978-1-4614-2018-7_30.

[30] Zimmerman, B. (2000), “Attaining Self-Regulation”, in Handbook of Self-Regulation, Elsevier, https://doi.org/10.1016/b978-012109890-2/50031-7.