The use of digital resources plays an increasingly prominent role in students’ lives, both at home and in the classroom. Ensuring that digital devices and learning materials – if they are used – are of high quality and employed effectively is important for students’ educational achievement. Using data from the OECD Programme for International Student Assessment (PISA) in 2022, this chapter analyses how digital resources are used in Swedish schools and how they are related to students’ mathematics performance. 12 indices and six individual items are examined in depth, comparing Sweden’s results to those of other OECD and Nordic countries. Associations with mathematics performance are explored using bivariate and multivariate analyses, disaggregated by gender, socio-economic status, migrant background and school location. PISA data show that students in Sweden make extensive use of digital technologies for learning. They express a high level of satisfaction with the availability of digital resources and their teachers’ digital skills. Compared with other OECD countries, the use of digital resources in Sweden is also associated with better student performance, though not in mathematics lessons. The PISA data point to challenges related to a greater prevalence of digital distractions as well as socio-economic inequities in students’ digital literacy and the quality of digital resources. Strategies that could be considered to address these challenges are presented at the end of this chapter and further developed in Chapter 5.
Education in Sweden
Abstract
Digital technologies play an increasingly important role in OECD education systems. Investments in digital resources have surged over the past decade and the COVID-19 pandemic has further accelerated their integration in schools and classrooms (OECD, 2023[1]). There is a broad consensus that the use of digital technology in schools should not be an end in itself but ultimately serve students’ learning needs. If used effectively, digital technologies can enhance educators’ ability to meet all students’ learning needs by enabling more engaging and individualised forms of instruction and assessment that are responsive to students’ needs, abilities and learning styles (Ganimian, Vegas and Hess, 2020[2]). Proponents also highlight the potential of technology to foster student engagement, promote collaboration, strengthen students’ digital competency and encourage self-regulated learning (OECD, 2023[1]). Notwithstanding their potential benefits, however, education systems are far from exploiting the full potential of digital technologies and the challenges and risks associated with their use have been extensively documented (Bulman and Fairlie, 2016[3]; OECD, 2021[4]; OECD, 2023[1]; OECD, 2019[5]).
Despite a growing body of rigorous research on the use of digital technologies in schools, generalisable evidence concerning their impact on students’ educational development is limited, given the highly context-dependent and diverse uses of digital resources in schools. While a detailed discussion of recent evidence on effective digital learning practices is beyond the scope of this working paper, research suggests that the effective integration of digital tools depends on the careful design of tasks, their alignment with learning objectives, and the selection of suitable tools (Forsström et al., 2025[6]). More specifically, an in-depth OECD literature review finds that effective digital integration is characterised by structured tasks aligned with curriculum standards, the thoughtful integration of tools linked to learning goals, robust teacher support for students and their learning process, a balanced use of digital and analogue methods, personalised learning opportunities, collaboration, and classroom management that builds on relational aspects such as trust, mutual respect, and positive teacher-student relationships (Forsström et al., 2025[7]).
Without an enabling policy environment and sufficient professional capacity at all levels of the education system, investments in education technology are unlikely to pay off and have the desired effect on students’ learning outcomes. Many education systems have therefore turned to strengthening teachers’ capacity for digital education through professional learning, guidance and resources (OECD, 2025[8]). To take effective measures to improve the use of digital resources in schools, policy makers need a detailed understanding of the role that digital technologies play in students’ education and their relationship with students’ learning outcomes. The 2022 Programme for International Student Assessment (PISA) presents a unique opportunity to engage in this analysis based on comprehensive and representative international data on the use of digital technologies for teaching and learning after the COVID-19 pandemic. In doing so, this chapter seeks to support policy makers’ efforts to promote a more effective use of digital resources in schools and to ensure that all schools, teachers and students are in a position to benefit from the digital transformation (OECD, 2023[9]).
Data sources and limitations
PISA collects data on digital education through both student and school questionnaires, offering a comprehensive perspective on students’ use of digital resources and schools’ digital education policies and practices. Student-reported data is derived from both the main PISA 2022 Student Questionnaire (OECD, 2021[10]) and an optional questionnaire on students’ familiarity with Information and Communications Technology (ICT) (the ICT Familiarity Questionnaire), which Sweden administered as one of 53 countries/economies participating in PISA 2022 (OECD, 2021[11]). This data provides insights into students’ use of digital devices at home and at school as well as their experience of digital learning practices. It also allows us to investigate the association between students’ use of digital resources and their performance in mathematics.
This student-level data is combined with principal-reported data from the PISA 2022 School Questionnaire (OECD, 2021[12]), which provides important contextual information on schools’ digital resources, their preparedness for digital learning and their policies concerning the use of digital devices. By combining students’ perspectives with school-level information, the chapter provides a nuanced picture of digital education in Swedish schools. It also complements previous analyses carried out by the Swedish National Agency for Education (Skolverket), which focussed on student-level data (Skolverket, 2024[13]).
While PISA data provides rich insights into the use of digital technology in Swedish schools, it also has methodological limitations that policy makers need to be aware of when interpreting the data presented in this report. Most importantly, as discussed in the introduction (Chapter 2) and Chapter 4 of this report, the associations presented throughout the chapter reflect statistical correlations and should not be interpreted as causal. Second, participation in PISA is limited to 15-year-old students. All results presented in this chapter therefore reflect the learning experience of students in this age group and may not be representative of those of younger or older students. Third, most of the data on the use of digital resources presented in this chapter is reported by students or school leaders and requires cautious interpretation due to the subjective nature of self-reported data and potential cultural differences in responses. Finally, unlike some of the questionnaire items on school climate analysed in Chapter 4, many of the PISA questions on digital education were newly introduced or modified in 2022. As a result, the scope for the analysis of trends in digital education and comparisons across PISA cycles is limited (See Chapter 4 for a detailed discussion).
Main indicators
This analysis draws on data from the PISA 2022 student and school questionnaires to examine key dimensions of digital education in Swedish schools. A total of 12 indices and six additional variables were analysed, covering a broad range of factors related to students’ use of digital resources and schools’ digital education policies and practices. Specifically, 11 indices were constructed from student questionnaire data and one from the school questionnaire, alongside three student-reported and three principal-reported separate indicators:
Indices from the PISA Student Questionnaires:
Index of ICT availability at school (ICTSCH)
Index of ICT availability outside school (ICTHOME)
Index of the quality of access to ICT (ICTQUAL)
Index of subject-related ICT use during lessons (ICTSUBJ)
Index of ICT use in enquiry-based learning activities (ICTENQ)
Index of support or feedback via ICT (ICTFEED)
Index of ICT use for school activities outside of the classroom (ICTOUT)
Index of students’ self-efficacy in digital competencies (ICTEFFIC)
Index of students’ practices regarding online information (ICTINFO)
Index of frequency of ICT activity (leisure) – Weekday (ICTWKDY)
Index of frequency of ICT activity (leisure) – Weekend (ICTWKEND)
Additional items from the PISA Student Questionnaire:
Time spent using digital resources for learning or leisure per day (ST326)
Distractions by digital resources in mathematics lessons (ST273)
Feelings and behaviours related to the use of digital devices (ST322)
Indices from the PISA School Questionnaire (principal-reported):
Index of schools’ preparedness for digital learning (DIGPREP)
Additional items from the PISA School Questionnaire (principal-reported):
A lack of or inadequate digital resources hindering the capacity to provide instruction (SC017)
Professional development on the integration of digital resources for mathematics teachers (SC184)
School policies on how to use digital devices in mathematics instruction (SC189)
The following sections explore the relationship between students’ performance in PISA 2022 and a range of indicators related to digital education. The indicators cover various aspects of students’ use of digital resources, including the intensity of students’ digital device use both inside and outside of school, students’ access to high-quality digital resources, their self-rated digital competency, teachers’ preparedness to use digital tools, the way digital resources are used in the classroom and schools’ policies concerning digital devices. Taken together, these indicators provide rich insights into the role of digital resources in Swedish schools and how they relate to students’ performance.
For each index or item, average results for Sweden were compared to the OECD average and to those of three Nordic peers: Finland, Norway and Denmark*,1 where possible.2 To better understand how digital resources are associated with students’ learning outcomes, the analysis proceeds in several steps. First, bivariate regressions were conducted to examine the linear relationship between each digital education index/item and students’ mathematics performance. All PISA indices are normalised across OECD countries, such that 0 corresponds to the OECD average and 1 corresponds to the standard deviation across OECD countries. To account for potential non-linear relationships between students’ performance and selected variables, separate estimates of students’ mathematics scores were obtained for independent variables where curvilinear relationships were expected (for example, previous PISA analyses demonstrated that spending both very little and very much time using digital resources for learning is associated with lower performance (OECD, 2023[14]; Skolverket, 2024[13])).
The associations between the selected variables and students’ performance were estimated for the entire population and for subgroups of students (girls, boys, socio-economically advantaged and disadvantaged students, immigrant and non-immigrant students, and students attending schools in rural area or villages, towns and cities). Throughout the report, socio-economically disadvantaged (or advantaged) students are defined as those in the bottom (or top) quarter of the PISA index of economic, social and cultural status (ESCS) in their respective countries. Rural area or villages (referred to as “rural” in figures and tables) are defined as having fewer than 3 000 inhabitants; towns have between 3 000 and 100 000 inhabitants; and cities have over 100 000 inhabitants. The bivariate regressions describe the simple association between each index/item and students’ performance, without controlling for other variables.
In a second step, multivariate analyses were conducted to account for students’ and schools’ socio-economic profile, as measured by the PISA index of economic, social and cultural status (ESCS). The results of the analysis indicate whether the observed associations between the use of digital resources and students’ performance persist independent of students’ socio-economic context. All associations described in this chapter should be interpreted as correlational, not causal. The statistical significance of all results is clearly indicated.
Students’ use of digital education resources and performance in PISA 2022 (Student Questionnaire)
Data from the PISA 2022 Student Questionnaire provide a deep insight into Swedish students’ use of digital resources both at school and at home. They capture students’ use of digital resources for a variety of purposes, both in private and in the classroom, how they feel when using digital devices, as well as their subjective assessment of the quality of their schools’ digital resources and their teachers’ digital skills. However, the interpretation requires caution due to its self-reported nature and potential biases arising, for example, from memory decay, social desirability (the tendency to respond in a manner that feels more socially or culturally acceptable) or reference-group bias (arising from different standards to which students compare their situation, e.g. when evaluating whether their schools’ digital resources are adequate) (OECD, 2023[14]).
Availability of different types of digital resources at school and at home
The availability of digital infrastructure in schools and students’ homes has increased dramatically over the past decade and is now widespread across OECD countries. Between 2012 and 2022, PISA data indicates that the computer-to-student ratio in the schools of 15-year-olds has increased significantly in 21 of 37 OECD countries (rising by 0.13 to reach 0.81 on average). Sweden has seen an even more dramatic increase (by 0.34 since 2012 to reach 0.96 in 2022). Most of this expansion of digital devices in Sweden took place between 2012 and 2015 (OECD, 2023[14]). While students’ access to digital resources is not sufficient to guarantee their effective use, it is an important precondition for the digital transformation of schooling. The following sub-sections therefore examine students’ access to different types of digital equipment both at school and at home, as well as its relationship with students’ performance in Sweden.
International comparison and differences between student groups (resources at school)
The great majority of Swedish students reported having access to a broad range of digital resources at school. As is the case across OECD countries, more than 90% of Swedish 15-year-olds reported that a desktop or laptop, a smartphone, Internet access, educational software and a learning management system are available at their school (Table 3.1).
Proportion of students reporting that the following digital resources are available to them at school
Source: OECD (2023[15]), PISA 2022 Database, https://www.oecd.org/en/data/datasets/pisa-2022-database.html.
The range of digital resources that students reported having access to at school is summarised by the PISA index of ICT availability at school (OECD, 2024[16]).3 The mean index of ICT availability at school in Sweden is -0.16, slightly below the OECD average and the scores for Denmark (0.12) and Finland (0.14). This indicates that students in Sweden have access to a somewhat narrower range of digital resources at school on average. This is largely explained by a slightly lower proportion of students in Sweden reporting not having access to tablet devices or e-book readers (81.7% compared to the OECD average of 88.5%), smartphones (90.0% compared to the OECD average of 94.9%) and desktops or laptops (93.8% compared to the OECD average of 96.1%). The reported availability of digital resources does not significantly differ by students’ socio-economic status, gender or school location. However, immigrant students reported having access to a more limited range of digital resources at school than non-immigrant students (scoring -0.27 vs. -0.13 in the index).
It should be noted that the index of ICT availability at school does not capture the frequency with which students use digital resources or the purposes they use them for. In fact, despite a slightly lower availability of some digital resources, students in Sweden reported a significantly more frequent use of digital devices than their peers in other OECD countries (see Table 3.2). For example, 71.7% of students in Sweden reported using a desktop or laptop computer at school “almost every day” or more, compared to only 35.9% across OECD countries (and compared to 89.3% in Denmark and 42.2% in Finland). Likewise, 41.7% of students in Sweden reported using educational software, games or apps, other learning tools “almost every day” or more at school, compared to only 25.0% across OECD countries (and compared to 56.4% in Denmark and 19.9% in Finland). Even for tablet devices and e-books, which are less frequently available in Sweden than in other countries, 21.4% of students reported using them at school “almost every day” or more, which is above the OECD average of 18.3% and more than in Denmark (17.7%) or Finland (16.1%).
Proportion of students reporting that they use the following digital resources “every day or almost every day” or “several times a day” at school
Note: *Across this report, caution is required when interpreting estimates for Denmark and other countries marked with an asterisk in figures and tables because one or more PISA sampling standards were not met (see Reader’s Guide in OECD (2023[17]), PISA 2022 Results (Volume I): The State of Learning and Equity in Education, https://doi.org/10.1787/53f23881-en, Annexes A2 and A4).
Source: OECD (2023[15]), PISA 2022 Database, https://www.oecd.org/en/data/datasets/pisa-2022-database.html.
Association with students’ performance (resources at school)
Overall, the range of digital resources available to students in Sweden is not statistically significantly associated their mathematics performance in PISA (Figure 3.1). By contrast, a one-unit increase in the index of ICT availability at school is associated with small increases in performance on average across OECD countries (3 points), in Denmark (4 points) and in Finland (3 points). However, the availability of resources in Sweden is associated with moderately higher performance among girls (6 points) and among students attending schools in cities (5 points) (Figure 3.1).
Difference in mathematics performance associated with a one-unit increase in the index (student-reported), by student and school characteristics in Sweden
Note: Statistically significant differences are shown in a blue tone. Each bar represents the coefficient from a bivariate regression of mathematics performance on the index, estimated separately for each subgroup. No control variables are included in these models. Results reflect associations, not causal effects.
Source: OECD (2023[15]), PISA 2022 Database, https://www.oecd.org/en/data/datasets/pisa-2022-database.html.
Once students’ and schools’ socio-economic profile is accounted for, the association between the range of digital resources available to students and their mathematics performance turns statistically significant across the student population. After accounting for socio-economic context, a one-unit increase in the index (corresponding to one standard deviation across OECD countries) is associated with a 3-point increase in mathematics performance (Annex Table 3.A.1), i.e. slightly more than the bivariate estimate shown in Figure 3.1 (2 points). The index explains only 0.1% of the variance in mathematics performance (before accounting for students’ and schools’ socio-economic profile), which is below the OECD average of 0.3%. By comparison, on average across OECD countries, 15% of the variance in mathematics performance was explained by students’ socio-economic status in 2022 (OECD, 2023[17]).
International comparison and differences between student groups (resources at home)
In addition to the availability of digital resources at school, PISA 2022 asked students about their access to various types of digital resources at home. The results indicate that – in line with the OECD average and other Nordic countries – well over 90% of students in Sweden have access to computers, smartphones, tablets, education software and learning management systems or learning platforms at home. The range of digital resources that students reported having access to at home is summarised by the PISA index of ICT availability outside school (OECD, 2024[16]).4 The mean index in Sweden is -0.04, very close to the OECD average and only slightly below the scores for Denmark (0.11) and Finland (0.04). This suggests that the overall availability of digital devices in Swedish students’ homes is broadly in line with other OECD countries.
In most OECD countries, advantaged students reported having access to a greater range of digital resources at home than disadvantaged students (on average, they score 0.07 vs. -0.06 in the index of ICT availability outside school). By contrast, there are no significant socio-economic differences between the digital resources available to advantaged and disadvantaged students in Sweden (-0.02 vs. -0.05). However, boys reported having access to fewer digital resources at home than girls (-0.08 vs. 0.01) and immigrant students have access to fewer resources than non-immigrant students (-0.13 vs. -0.01). Both are in line with the average gaps observed across OECD countries.
Difference in mathematics performance associated with a one-unit increase in the index (student-reported), by student and school characteristics in Sweden
Note: Statistically significant differences are shown in a blue tone. Each bar represents the coefficient from a bivariate regression of mathematics performance on the index, estimated separately for each subgroup. No control variables are included in these models. Results reflect associations, not causal effects.
Source: OECD (2023[15]), PISA 2022 Database, https://www.oecd.org/en/data/datasets/pisa-2022-database.html.
Association with students’ performance (resources at home)
Unlike the range of digital resources available at school, the availability of digital resources at home, as measured by the PISA index of ICT availability outside school, is significantly associated with the mathematics performance of most student groups in Sweden. Across the student population, a one-unit increase in the index is associated with a 6-point increase in mathematics performance (similar to the 5-point increase observed on average across OECD countries). This positive association was observed across most subgroups. One exception are students in rural areas or villages for whom the association is not significant. Likewise, advantaged students benefit significantly more from the availability of ICT resources at home than disadvantaged students for whom the index is not significantly associated with higher performance (Figure 3.2). The average association between the availability of digital resources at home and students’ performance remains significant and of similar magnitude (5-points per one-unit increase in the index of ICT availability outside school) after accounting for students’ and schools’ socio-economic profile (Annex Table 3.A.2).
Conclusion
Sweden is a highly digitalised country with widespread access to digital devices both in schools and in students’ homes. In PISA 2022, students reported that some digital resources are nearly universally available at school and at home, though the range of digital devices present in schools is slightly below the OECD average. The distribution of digital resources also appears to be equitable with few socio-economic disparities in the reported availability of devices at school or at home (although there is some evidence of socio-economic inequities in the quality of these devices, as discussed further below). The only exception are immigrant students, who reported having access to a slightly more limited range of digital resources at school than non-immigrant students. The availability of digital devices in Swedish schools is associated with slightly higher levels of mathematics performance once students’ and schools’ socio-economic profile is accounted for. Having access to a wider range of digital resources at home is associated with higher levels of performance for advantaged, but not for disadvantaged students. This relatively small but statistically significant increase in performance associated with students’ access to digital resources is consistent with the notion that access to digital resources is an important, if not sufficient, condition for effective teaching and learning – particularly in highly digitalised education systems like Sweden.
Access to and quality of ICT resources at school (students’ reports)
The presence of digital devices in schools alone does not guarantee students’ and teachers’ ability to integrate them in the learning process effectively. To assess whether the ICT resources in schools are both accessible and of sufficient quality, PISA asked students whether they were satisfied with the accessibility of digital resources (e.g. whether there are enough devices with Internet access for every student), the quality of digital resources (e.g. whether the resources function properly) and their teachers’ skills to use these digital devices during instruction. These responses are summarised in the PISA index of the quality of access to ICT (OECD, 2024[16]).5
International comparison and differences between student groups
In PISA 2022, Swedish students reported a high level of satisfaction with the availability and quality of digital resources in their schools, scoring 0.42 in the index of the quality of access to ICT on average. This is significantly above the OECD average and the scores in other Nordic countries, including Denmark (0.25) and Finland (0.20) (OECD, 2024[18]). Sweden has a lead over the OECD average across all dimensions related to the access to and quality of ICT resources at school. The proportion of Swedish students who agreed or strongly agreed with statements about the quality or availability of ICT resources at their school is consistently 9-19% above the OECD average for all but one of the items comprising the index (see Figure 3.3). For example, 85% of Swedish students agreed that there are enough digital resources for every student at their school, compared to 71% on average across OECD countries. Likewise, 69% agreed that their schools’ Internet speed is sufficient and 80% agreed that digital resources function properly at their school, compared to 54% and 71% on average across OECD countries respectively. Furthermore, 79% of Swedish students agreed that their teachers have the necessary skills to use digital devices during instruction, compared to 69% on average across OECD countries (Figure 3.3).
Percentage of students who agreed or strongly agreed with the following statements
Source: OECD (2024[18]), PISA 2022 Results (Volume V): Learning Strategies and Attitudes for Life, https://doi.org/10.1787/c2e44201-en, Table V.B1.10.1.
Although many Swedish students are satisfied with the quality and accessibility of digital resources in their schools on average, there are significant socio-economic discrepancies. Advantaged students reported significantly better access to high-quality digital resources, scoring 0.53 on the combined index, compared to disadvantaged students, who score 0.29. This gap (0.24) is larger than that observed in Denmark (0.19), Finland (0.07) and on average across OECD countries (0.13). By contrast, no significant gaps are observed between boys and girls or between immigrant and non-immigrant students in Sweden.
Association with students’ performance
Students who reported being satisfied with the availability and quality of digital resources at their school (as measured by the PISA index of the quality of access to ICT) performed significantly better in PISA. Across the Swedish student population, a one-unit increase in the index is associated with a 15-point increase in mathematics performance. This association is stronger than the one observed on average across OECD countries (7 points), in line with that observed in Denmark (17 points) and stronger than that observed in Finland (4 points). This association is strong and significant across different subgroups in Sweden, holding for students regardless of their gender, socio-economic status or school location (Figure 3.4). There are no statistically significant differences in the strength of the association with students’ performance across subgroups (though gender differences were observed for individual items, as shown in Table 3.3).
As discussed above, advantaged students in Sweden are more satisfied with the quality and accessibility of ICT resources at their schools than disadvantaged students. However, a multivariate analysis shows that the observed relationship between the quality of access to ICT and students’ performance is only partly explained by socio-economic context. After accounting for students’ and schools’ socio-economic profile, a one-unit increase in the index of the quality of access to ICT remains associated with a statistically significant 12-point increase in mathematics performance in Sweden. This change is smaller but directionally consistent with the bivariate estimate presented above (15 points) (Annex Table 3.A.3). A similar reduction in the strength of the association could be observed on average across OECD countries (from 7 to 4 points), in Denmark (from 17 to 13 points) and in Finland (from 4 to 3 points). This underlines that students’ access to high-quality ICT at school is an important correlate of performance, regardless of students’ socio-economic status.
Difference in mathematics performance associated with a one-unit increase in the index (student-reported), by student and school characteristics in Sweden
Note: Statistically significant differences are shown in a blue tone. Each bar represents the coefficient from a bivariate regression of mathematics performance on the index, estimated separately for each subgroup. No control variables are included in these models. Results reflect associations, not causal effects.
Source: OECD (2023[15]), PISA 2022 Database, https://www.oecd.org/en/data/datasets/pisa-2022-database.html.
Students’ satisfaction with their schools’ digital resources is associated with higher performance and there are no significant differences in the strength of this association between different student groups when considering the overall index of the quality of access to ICT. When looking at its constituent items, however, some heterogeneity is apparent, particularly between boys and girls. Notably, boys’ mathematics performance improves by a significantly greater margin than girls’ if they reported that there are enough digital resources for every student or enough digital devices with access to the Internet at their schools. Similarly, though girls perform slightly better if they reported that digital resources are functioning properly at their school and are easily accessible within the classroom, the difference is even greater for boys (Table 3.3).
Differences in mathematics performance associated with students agreeing or strongly agreeing with the following statements in Sweden
Note: The estimated differences in mathematics performance are based on separate bivariate regressions for each subgroup. Statistically significant differences in performance (or differences between subgroups) are highlighted in bold. Results reflect associations, not causal effects
Source: OECD (2023[15]), PISA 2022 Database, https://www.oecd.org/en/data/datasets/pisa-2022-database.html.
Insights from international practice
OECD countries have developed a range of strategies to address inequities in students’ access to digital resources. While many countries already account for school characteristics in their general funding mechanisms, some have added targeted programmes designed to strengthen the digital capacity of disadvantaged schools (OECD, 2023[1]; OECD, 2020[19]). Two international examples of such funding mechanisms, in France and in Ireland, are described in Box 3.1.
Ireland: Targeted funding for digital devices in disadvantaged schools
Ireland has committed to providing primary and post-primary schools with a total of EUR 200 million to purchase digital equipment between 2021 and 2027 via the ICT Infrastructure Grant Scheme. Schools can use the funding to purchase desktop PCs, laptops or other digital devices for students and teachers. The amount of the ICT Grant is based on a flat rate lump sum (EUR 2 000) and a per capita amount, which is higher for students in special education needs schools and in the 1 200 schools that are part of the government’s DEIS (Delivering Equality of Opportunity in Schools) programme due to their high concentration of students at risk of educational disadvantage. Schools have discretion over how they implement the provision of digital devices for students. Some operate a BYOD (Bring Your Own Device) policy; others implement a laptop loan system or have one device per student. Schools may supplement the ICT Grant with additional funding from their general operating funding (the capitation grant) and may ask parents to make voluntary contributions. Schools are encouraged to specify how they intend to utilise the ICT Grant in their Digital Learning Plans. The submission of the Plan may become mandatory in the future (Department of Education, 2024[20]).
France: Targeted funding to strengthen the digital capacity of rural schools
Since 2017, France has aimed to strengthen digital capacity in rural schools with the targeted programme Innovative Digital Schools and Rurality Programme (Programme Écoles numériques innovantes et ruralité). Initially endowed with EUR 20 million, and with another EUR 15 million added in 2020, the programme supported around 7 000 schools with digital equipment to promote learning, enrich relationships with families and reinforce the attractiveness of rural schools and territories (Ministère de l’Éducation Nationale et de la Jeunesse, 2018[21]). This programme complements previous efforts to build capacity in rural schools, for example by fostering collaboration and by regularly letting teachers visit rural schools to promote the use of educational materials and digital equipment in the classroom (Équipe mobile académique de liaison et d’animation, EMALA) (Echazarra and Radinger, 2019[22]).
Source: OECD (2023[1]), Shaping Digital Education: Enabling Factors for Quality, Equity and Efficiency, https://doi.org/10.1787/bac4dc9f-en, p.177; Department of Education (2024[20]), Grant Scheme for ICT Infrastructure (Circular 0039/2024).
Conclusion
Swedish students are, on average, highly satisfied with the availability and quality of digital devices in their schools. Their average satisfaction exceeds both the OECD average and students’ reports in other Nordic countries, including Denmark and Finland. While there are no significant differences in the ICT access and quality reported by boys and girls or immigrant and non-immigrant students, advantaged students in Sweden reported having better access to quality ICT resources than disadvantaged students (scoring 0.53 vs. 0.29 on the relevant PISA index). This socio-economic gap is larger than in most OECD and Nordic countries. Swedish students who reported better access to high-quality digital resources at school also score higher in mathematics (+15 points per one-unit increase in the index). This positive association holds across all student groups and after accounting for socio-economic context. In light of the inequities observed above, the uneven quality of digital resources could therefore contribute to widening socio-economic inequities in students’ performance. Furthermore, boys’ performance appears to be more strongly linked to their schools having sufficient, well-functioning and easily accessible digital resources than girls’.
Time spent using digital resources for learning and leisure
The time that students spend using digital resources (for both leisure and learning activities) varies widely across education systems (OECD, 2023[14]). Sweden, together with other Nordic countries, is among the education systems whose students spend the most time engaging with digital resources for learning activities at school (OECD, 2023[14]). On a typical weekday, Swedish students reported spending 3.0 hours using digital resources for learning activities at school. This is significantly above the OECD average of 2.0 and – among OECD countries – only surpassed by students in Denmark (3.8 hours) and Norway (3.1 hours). In addition, Swedish students reported spending 1.5 hours per day before and after school and 1.6 hours per day on weekends using digital resources for learning activities (Figure 3.5). These times were closer to the OECD average.
Based on students' reports
Source: OECD (2023[14]), PISA 2022 Results (Volume II): Learning During – and From – Disruption, https://doi.org/10.1787/a97db61c-en, Table II.B1.5.62.
As discussed in the introduction, there is no generalisable evidence concerning the impact of digital resources on students’ educational development, given its highly context-dependent and variable nature. The use of digital technologies has been credited for its potential to enable more engaging and individualised forms of instruction and assessment that are responsive to students’ needs, abilities and learning styles (OECD, 2023[1]). Others have expressed concerns about the effect that the use of digital tools might have on children’s emotional well-being, their mental health and brain development (OECD, 2017[23]). The available evidence does suggest that not all uses of digital technologies lead to positive learning outcomes, that promoting access to digital equipment alone is insufficient (Bulman and Fairlie, 2016[3]; OECD, 2022[24]; OECD, 2019[5]) and that the association between the use of digital education technology and students’ performance varies across different tools and use cases (Skolverket, 2024[13]). Seizing the potential of digital technologies appears to depend on the types of technology used as well as how and with which pedagogical intent they are employed in teaching and learning (OECD, 2023[1]). The following sub-sections explore how the time that students spend using digital resources for both learning and leisure is associated with their mathematics performance in Sweden.
Time spent using digital resources for learning at school
In most OECD countries, the time that students reported spending using digital resources for learning at school has a curvilinear relationship with their performance. That is, while students who spend a moderate number of hours per day using digital resources for learning activities at school score higher in mathematics than students who spend no time on such devices (even after accounting for students’ and schools’ socio-economic profile), longer periods of use tend to be associated with diminishing student performance. On average across OECD countries, students who reported spending between 5 and 7 hours per day using digital resources for learning at school (7.8% of students), for example, score 12 points lower than students who spend between 3 and 5 hours per day. (After accounting for students’ and schools’ socio-economic profile, the former group of students scores 10 points lower). Students who spend over 7 hours per day using digital resources for learning activities at school score even lower (OECD, 2023[14]).
Mathematics performance associated with the time (per day) that students reported spending using digital resources for learning at school
Note: Statistically significant changes from one category to the next are shown in a darker tone. Results are based on bivariate regressions of mathematics performance on the variable. No control variables are included in these models. Results reflect associations, not causal effects.
Source: OECD (2023[15]), PISA 2022 Database, https://www.oecd.org/en/data/datasets/pisa-2022-database.html.
Compared to other OECD countries, the time that students reported spending using digital resources for learning at school in Sweden is more positively related to performance. Notably, strong increases in performance are observed for students who reported spending up to 4 hours using digital resources for learning per day and they significantly diminish only for students who reported spending more than 7 hours per day (Figure 3.6). Even after accounting for students’ and schools’ socio-economic profile, students who spend up to 2 hours score 18 points higher than those reporting spending up to 1 hour and students who spend 3 to 5 hours score 8 points higher than those spending 3 hours or less (OECD, 2023[14]). This is in line with previous analyses of Swedish PISA data, which indicated that the use of digital resources tends to be associated with higher student performance (Skolverket, 2024, pp. 85, Table 17[13]), whereas the mere availability of resources is not (see Figure 3.1).
The time that students spend using digital resources for learning at school varies significantly (Figure 3.7). On average across OECD countries, 44.2% of students reported spending either none or no more than one hour on a typical day to learn using digital resources at school. This share was significantly lower in Sweden (21.7%), as well as the other Nordic countries, Denmark (13.6%), Finland (27.1%) and Norway (17.8%). In Sweden, 44.1% of students reported spending between 2 and 5 hours per day using digital resources for learning – similar to the share observed in Denmark (45.8%) and Norway (53.1%), but significantly more than in Finland (29.4%) or on average across OECD countries (23.7%). The share of students spending more than 5 hours per day using digital resources for learning at school in Sweden (18.5%) – while above the OECD average (11.6%) – was significantly smaller than that in Denmark (33.3%) and in line the shares observed in Finland (18.3%) and Norway (17.0%) (OECD, 2023[14]).
Percentage of students who reported spending the following hours per day
Source: OECD (2023[14]), PISA 2022 Results (Volume II): Learning During – and From – Disruption, https://doi.org/10.1787/a97db61c-en, Table II.B1.5.62.
There are some notable differences in the time that different student groups reported spending using digital resources for learning at school. On average, disadvantaged students in Sweden spend 0.5 hours less per day learning with digital resources than advantaged students (2.7 vs. 3.2 hours), and immigrant students spend 0.4 hours less than non-immigrant students (2.6 vs. 3.1 hours) (OECD, 2023[14]). Advantaged students, girls and non-immigrant students are also more likely to report being among the most intensive users, spending more than 5 hours a day using digital resources for learning at school.
Time spent using digital resources for learning before and after school
In addition to the time they spend learning at school, Swedish students reported spending 1.5 hours on digital resources for learning before and after school. This is close to the OECD average and broadly in line with the time spent by students in Denmark (1.4 hours), Finland (1.8) and Norway (1.5) (OECD, 2023[14]). As for the time spent at school, the time spent using digital resources for learning before and after school has a curvilinear association with students’ performance. On average across OECD countries, students who reported using digital resources for learning before and after school for 1 hour or less per day perform better than those who do not at all. Performance further increases for students who report spending up to 2 hours per day before gradually diminishing. While a similar pattern holds in Sweden, the positive association between digital learning outside the school diminishes and turns negative more quickly. While students who spend up to 1 hour score 29 points higher than those who do not, students spending up to 3 hours do worse than those who spend up to 2 hours and students spending up to 4 hours fare worse than those who spend no time at all using digital resources to learn before or after school (Figure 3.8).
Mathematics performance associated with the time (per day) students reported spending using digital resources for learning before and after school
Note: Differences in mathematics performance compared with the reference category "not at all". Statistically significant changes from one category to the next are shown in a darker tone. Results are based on bivariate regressions of mathematics performance on the variable. No control variables are included in these models. Results reflect associations, not causal effects.
Source: OECD (2023[15]), PISA 2022 Database, https://www.oecd.org/en/data/datasets/pisa-2022-database.html.
There are some notable differences in the time that different student groups reported spending using digital resources for learning before and after school. On average, boys in Sweden spend 0.2 hours less per day learning with digital resources than girls (1.4 vs. 1.6 hours), disadvantaged students spend 0.2 hours less than advantaged students (1.4 vs. 1.6 hours) and non-immigrant students spend 0.4 hours less than immigrant students (1.4 vs. 1.8 hours) (OECD, 2023[14]).
Time spent using digital resources for leisure at school
Students not only use digital devices for learning, but also for leisure and schools in many OECD countries have enacted policies to limit the use smartphones and other personal devices at school. This is also the case for Sweden where a 2022 survey of the National Agency for Education found that seven out of ten teachers in Year 7-9 and just over half of the teachers in upper secondary schools reported that they ban mobile phones during their lessons (Skolverket, 2022[25]). PISA 2022 data suggests that the use of smartphones at school is associated with greater levels of distractions and can contribute to a deterioration of the classroom climate (OECD, 2023[14]).
In Sweden, students reported spending 1.1 hours per day using digital resources for leisure at school. This is the same as the OECD average, slightly less than in Finland (1.4 hours) and slightly more than in Denmark (0.7 hours) and Norway (0.8 hours) (see Figure 3.5). It should be noted that the data does not distinguish between students’ use of digital resources during lessons or between lessons, so this may include the time students spend using digital resources during breaks. Although the great majority of Swedish students reported spending no time (28.8%), up to 1 hour (37.8%) or up to 2 hours (15.4%) using digital resources for leisure at school, a significant minority reported spending between 3 and 5 hours (6.7%), between 5 and 7 hours (2.5%) or even more than 7 hours per day (2.0%) using digital resources for leisure at school (OECD, 2023[14]).
As one might expect, the time spent using digital resources at school is more negatively associated with students’ performance when they are used for leisure, rather than learning. On average across OECD countries, students who reported spending up to one hour per day using digital resources for leisure activities at school score 20 points higher in mathematics than students who spend no time. However, the mathematics performance gradually diminishes for students spending more than an hour using digital resources for leisure at school (OECD, 2023[14]). The association between the leisure use of digital resources and students’ performance in Sweden closely tracks that of the OECD average (Figure 3.9). Notably, students spending up to 3 hours per day using digital resources for leisure at school performed 26 points worse in mathematics than those who used them for up to 2 hours and worse than students who did not use digital resources for leisure at school at all. (Even after accounting for students’ and schools’ socio-economic profile, this drop remained significant at 23 points).
Mathematics performance associated with the time (per day) students reported spending for leisure using digital resources at school
Note: Statistically significant changes from one category to the next are shown in a darker tone. Results are based on bivariate regressions of mathematics performance on the variable. No control variables are included in these models. Results reflect associations, not causal effects.
Source: OECD (2023[15]), PISA 2022 Database, https://www.oecd.org/en/data/datasets/pisa-2022-database.html.
Overall, there is little systematic variation in the average time that different student groups use digital resources for leisure at school in Sweden. On average, boys and girls use them a similar amount of time, as do immigrant and non-immigrant students. However, disadvantaged students spend 0.1 more hours per day using digital resources for leisure at school than their advantaged peers (1.2. vs. 1.0 hours) (OECD, 2023[14]). Disadvantaged students were also more likely to report spending more than one hour a day using digital resources than advantaged students (34.9% vs. 29.7%).
Conclusion
Students in Sweden spend significantly more time using digital resources for learning at school than their peers in most other OECD countries. As in most OECD countries, the time spent learning with digital resources at school has a curvilinear relationship with students’ mathematics performance in Sweden, meaning that performance increases with its use up to a certain point before diminishing at higher levels of intensity. Although these findings should not be interpreted as causal evidence, they are consistent with the hypothesis that the moderate use of digital devices is not intrinsically harmful and can even be positively associated with performance, whereas the misuse, excessive use, or abstention from the use of digital resources can be negatively associated with performance (OECD, 2023[14]).
In comparison with the OECD average, the time that Swedish students spend using digital resources for learning is associated with larger increases in performance, up to a longer duration of digital learning time. This may indicate that Swedish schools make more effective use of digital technologies than other education systems. It could also reflect that students or schools that make limited use of digital resources for learning are systematically lower-performing or vice versa. By contrast, the time spent using digital resources for learning before or after school is associated with fewer and more quickly diminishing improvements in performance in Sweden, compared to the OECD average. There are some potential sources of inequities in Swedish students’ engagement with digital resources. Notably, both disadvantaged and immigrant students spend less time using digital resources for learning at school, while the latter spend more time using them before and after school.
As expected, students’ intensive use of digital resources for leisure activities at school is associated with more negative outcomes than their use in learning activities. In this respect, the Swedish PISA results track the OECD average both regarding the intensity of students’ use of digital resources for leisure as well as its association with students’ performance. Disadvantaged students spend more time using digital resources for leisure at school in Sweden. Although the differences are small, they could contribute to performance gaps.
Use of ICT during subject lessons
The previous sections analysed the association between students’ mathematics performance and their use of digital resources for learning broadly defined, i.e. across any type of lesson or learning activity. To get a more fine-grained picture of students’ use of ICT in the classroom, PISA 2022 also asked students how frequently they used digital resources in lessons of specific subjects, namely their test language, mathematics, science and computer science, information technology, informatics or similar lessons.6 Their responses are captured in the PISA index of subject-related ICT use during lessons, which is analysed in the following sub-sections (OECD, 2024[16]).7
International comparison and differences between student groups
Given the significant amount of time that Swedish students reported spending using ICT for learning at school (see Figure 3.5), they score high on the overall index of subject-related ICT use (0.45), significantly exceeding the OECD average and Finland (0.05), but falling short of Denmark (1.32). Both Sweden and Denmark distinguish themselves by an intensive use of ICT in test language and science lessons. 71.7% and 64.7% of Swedish students reported using digital resources in more than half of these lessons respectively. This is less than was reported by students in Denmark (88.4% and 78.0% respectively) but significantly more than the OECD average (29.3% and 33.4% respectively) or in Finland. By contrast, Swedish students make comparatively little use of digital resources in mathematics lessons, where only 28.6% reported using them in more than half of their lessons. This was in line with the OECD average (26.3%), but significantly below Denmark (87.2%) (Figure 3.10).
Percentage of students reporting that they use digital resources in "more than half" or "every or almost every" of the following lessons (among students who report having the subject)
Note: Averages for a given subject are calculated excluding the students who report not having that subject.
Source: OECD (2023[15]), PISA 2022 Database, https://www.oecd.org/en/data/datasets/pisa-2022-database.html.
Association with students’ performance
In general, students who reported making more intensive use of ICT in subject lessons (as measured by the PISA index of subject-related ICT use during lessons) performed slightly better in PISA. In Sweden, a one-unit increase in the index is associated with an 8-point increase in mathematics performance. This is close to the association observed on average across OECD countries (7 points), but significantly weaker than the one observed in Denmark (19 points). In Finland, the association is not statistically significant. The association in Sweden is not observed consistently across the student population. While boys who reported a frequent use of ICT in subject lessons perform significantly better, girls do not. The same holds for disadvantaged, but not for advantaged students (Figure 3.11). Furthermore, the overall statistically significant association between the index of subject-related ICT use and mathematics performance in Sweden disappears once students’ and schools’ socio-economic profile is accounted for in a multivariate analysis (Annex Table 3.A.4). This suggests that the correlation between the use of ICT in subject lessons and students’ performance is, to some extent, driven by advantaged students reporting a more frequent use of ICT in subject lessons.
Difference in mathematics performance associated with a one-unit increase in the index (student-reported), by student and school characteristics
Note: Statistically significant differences are shown in a blue tone. Each bar represents the coefficient from a bivariate regression of mathematics performance on the index, estimated separately for each subgroup. No control variables are included in these models. Results reflect associations, not causal effects.
Source: OECD (2023[15]), PISA 2022 Database, https://www.oecd.org/en/data/datasets/pisa-2022-database.html.
Focussing on the association between students’ mathematics performance and the use of ICT in mathematics lessons (rather than all subjects combined) permits us to examine the link between the use of ICT and students’ performance more directly. In contrast to the general association between ICT use in subject lessons and mathematics performance (which is not significant once socio-economic context is taken into account), the intensive use of ICT in mathematics lessons appears to be negatively associated with students’ performance in Sweden. After accounting for students’ and schools’ socio-economic profile, students in Sweden who reported using digital resources in more than half of their mathematics lessons (which is the case for 28.3% of Swedish students) score 15 points lower in mathematics than those who reported using digital resources less frequently (Figure 3.12).
Students in Finland show an even stronger drop in performance associated with the intensive use of digital resources in mathematics lessons than Swedish students (-23 points after accounting for socio-economic context). Nevertheless, this negative association is not universal across OECD countries. While students who reported intensively using ICT in mathematics lessons tend to fare worse in mathematics on average (-6 points across OECD countries after accounting for socio-economic context), the intensive use of ICT is associated with significantly higher performance in several countries, including in Korea (13 points), Türkiye (23 points) and Denmark (32 points) (Figure 3.12).
Difference in mathematics performance associated with students using digital resources in more than half of mathematics lessons (student-reported)
Note: Only OECD countries with available data are shown. Caution is required when interpreting estimates for countries marked with an asterisk (*) because one or more PISA sampling standards were not met. Statistically significant associations are shown in a darker tone. The socio-economic profile is measured by the PISA index of economic, social and cultural status (ESCS). Results reflect associations, not causal effects.
Source: OECD (2023[15]), PISA 2022 Database, https://www.oecd.org/en/data/datasets/pisa-2022-database.html.
Conclusion
Swedish students reported frequently using ICT during subject lessons, scoring well above the OECD average and Finland, but below Denmark on the relevant PISA indicator. This is consistent with the relatively large amount of time that Swedish students reported spending to learn with digital resources. Students’ ICT use is especially frequent in Swedish language lessons and in science lessons (71.7% and 64.7% of students reported using ICT in over half of their lessons in these subjects), but relatively low in mathematics (28.6%). Once socio-economic context is taken into account, the frequency of students’ ICT use across subject lessons is not significantly linked to mathematics performance in Sweden. However, focussing on mathematics lessons shows that the intensive use of ICT (i.e. in more than half of all lessons) is associated with lower performance, even after accounting for socio-economic context. This contrasts with Denmark, but negative associations of different proportions were also observed in Finland and on average across OECD countries.
While it is not possible to determine the reason for the negative association between ICT use in mathematics lessons and students’ performance, separate analyses indicate that more intensive use of ICT is associated with a number of both positive and negative changes in students’ experience of mathematics lessons across OECD countries. On the one hand, the intensive use of digital resources in mathematics is associated with students reporting increased cognitive activation, increased teacher support, reduced mathematics anxiety, as well as a higher perceived quality of instruction (Boeskens and Echazarra, 2025[26]). On the other hand, spending more time using digital resources for learning in mathematics lessons is associated with a poorer disciplinary climate and a greater likelihood of students reporting that their classmates are being distracted by their own or their peers’ digital devices (Boeskens and Echazarra, 2025[26]).
Use of ICT for enquiry-based learning, support and feedback
Evidence on the use of digital education resources highlights that its efficacy relies on their pedagogical integration in the learning process and on teachers’ ability to use digital resources strategically and purposefully to enhance their practice (Bulman and Fairlie, 2016[3]; Forsström et al., 2025[6]; OECD, 2023[1]). PISA 2022 asked students about two types of pedagogical practices involving the use of digital resources: Enquiry-based learning activities and the provision of feedback or support. The following sections explore these pedagogical uses of digital resources and their association with students’ performance in greater detail.
International comparison and differences between student groups (enquiry)
Enquiry-based learning comprises classroom activities that encourage students to formulate questions, investigate to find answers, and build new understanding as they progress. Research suggests that enquiry-based learning stimulates independent research skills as well as curiosity and critical thinking. By means of their self-directed progression, activities of this type can help students develop proactive learning behaviours (OECD, 2024[18]; Pedaste et al., 2015[27]). Enquiry-based learning does not have to involve digital resources, but ICT can be used to promote certain enquiry-focussed activities. The PISA index of the use of ICT in enquiry-based learning activities captures how often students reported using digital resources for such learning activities, e.g. to create multimedia presentations, to track the progress of their own work or projects, or to analyse data that they have collected (OECD, 2024[16]).8
Overall, Swedish students frequently use ICT for enquiry-based learning activities, scoring 0.38 in the index. This is well above the OECD average and the score of Finland (0.02), but below Denmark (0.62) (OECD, 2024[18]). This general pattern holds across most of the activities included in the index. Only 16% of students in Sweden reported using ICT to create multimedia presentations and 23% to play digital learning games at least once or twice a week. These were also the least frequently reported activities across OECD countries. For all other activities, Swedish students reported a significantly more frequent engagement than their peers across OECD countries, from writing and editing text, to planning and managing their work or collaborating with peers to create digital content (Figure 3.13).
Percentage of students reporting that they use digital resources at least once or twice a week to conduct the following activities
Source: OECD (2024[18]), PISA 2022 Results (Volume V): Learning Strategies and Attitudes for Life, https://doi.org/10.1787/c2e44201-en, Table V.B1.10.4.
Differences in inquiry-based uses of ICT across student groups are broadly in line with those observed on average across OECD countries. While there are no significant differences between girls and boys (0.40 vs. 0.36 in the index), advantaged students reported engaging in enquiry-based activities with ICT more frequently than disadvantaged students (0.48 vs. 0.23). This statistically significant difference (0.25) is comparable to the ones observed in Denmark (0.21), Finland (0.31) and on average across OECD countries (0.26).
Association with students’ performance (enquiry)
Despite the potential benefits of enquiry-based learning, students’ frequent use of ICT in such activities is not associated with higher student performance across OECD countries. No statistically significant improvements are observed in Denmark, Finland or Sweden either (in Sweden, it is only associated with significantly worse performance for advantaged students) (Figure 3.14). In fact, once students’ and schools’ socio-economic profile is accounted for in a multivariate analysis, a one-unit increase in the index is negatively associated with students’ performance both in Sweden (-4 points) and on average across OECD countries (-4 points) (OECD, 2024[18]).
Difference in mathematics performance associated with a one-unit increase in the index (student-reported), by student and school characteristics in Sweden
Note: Statistically significant differences are shown in a blue tone. Each bar represents the coefficient from a bivariate regression of mathematics performance on the index, estimated separately for each subgroup. No control variables are included in these models. Results reflect associations, not causal effects.
Source: OECD (2023[15]), PISA 2022 Database, https://www.oecd.org/en/data/datasets/pisa-2022-database.html.
Some of the individual enquiry-based activities that comprise the index are statistically significantly associated with students’ performance. Notably, students who reported using ICT resources to write or edit text for a school assignment at least once or twice a week fare significantly better in mathematics (45 points) after accounting for students' and schools' socio-economic profile. Likewise, students who reported frequently using ICT to find information about real-world problems or phenomena online score significantly higher (22 points). Nevertheless, negative associations are observed for a number of other activities, such as creating a multimedia presentation (-28 points), playing digital learning games (-32 points), analysing data that students collected themselves (-23 points) or reporting or sharing results from students’ own experiments or investigations (-21 points) (Figure 3.15).
Difference in mathematics performance associated with students reporting that they use digital resources at least about once or twice a week to conduct the following activities in Sweden
Note: Statistically significant differences are shown in a darker tone. The socio-economic profile is measured by the PISA index of economic, social and cultural status (ESCS). Results reflect associations, not causal effects.
Source: OECD (2024[18]), PISA 2022 Results (Volume V): Learning Strategies and Attitudes for Life, https://doi.org/10.1787/c2e44201-en, Table V.B1.10.6.
International comparison and differences between student groups (feedback)
Digital resources promise to provide students with new channels to receive feedback on their work, be it from their teachers, from educational software or from their peers. The PISA index of support or feedback via ICT captures how often students reported reading or listening to feedback sent by their teachers and peers as well as the frequency with which students obtain feedback from learning software or work on digital drill and practice exercises (OECD, 2024[16]).9 Overall, students in Sweden reported making above-average use of ICT to receive support or feedback, scoring 0.24 in the index. This was significantly above the OECD average and Finland (0.04) and in line with the score of Denmark (0.21).
In international comparison, Swedish students receive particularly frequent feedback from their teachers and frequently engage in practice exercises using digital software. 37% of students reported doing the latter at least once or twice a week (compared to the OECD average of 27%, 32% in Denmark and 22% in Finland). 45% of Swedish students reported reading or listening to feedback sent by their teachers regarding their work and academic results at least once or twice a week (compared to the OECD average of 33%, 39% in Denmark and 31% in Finland).
Association with students’ performance (feedback)
As is the case across OECD countries, Swedish students who reported receiving more support or feedback via ICT tend to perform significantly worse in PISA. A one-unit increase in the index is associated with a 12-point reduction in mathematics performance in Sweden (Figure 3.16). This is broadly in line with the average association across OECD countries (-10 points) and those in Denmark (-15 points) and Finland (- 13 points). It is likely that this association reflects that teachers focus on providing feedback to under- performing students who stand to benefit the most from additional support. The negative association between ICT-facilitated feedback and performance holds across student groups. Though the association appears to be stronger for advantaged students, non-immigrant students and students living in cities, the respective differences between the comparator groups are not statistically significant. The negative overall association between the index and students’ performance persists and remains statistically significant after accounting for students’ and schools’ socio-economic profile in a multivariate analysis (strengthening slightly from a 12-point to a 13-point decline in performance per one-unit increase in the index) (Annex Table 3.A.5).
Difference in mathematics performance associated with a one-unit increase in the index (student-reported), by student and school characteristics in Sweden
Note: Statistically significant differences are shown in a blue tone. Each bar represents the coefficient from a bivariate regression of mathematics performance on the index, estimated separately for each subgroup. No control variables are included in these models. Results reflect associations, not causal effects.
Source: OECD (2023[15]), PISA 2022 Database, https://www.oecd.org/en/data/datasets/pisa-2022-database.html.
Conclusion
Swedish students make more frequent use of ICT for enquiry-based learning activities than the OECD average and – as in most countries - advantaged students do so more frequently than disadvantaged students. The association between enquiry-based activities and students’ performance is mixed. While students’ overall engagement in these activities is associated with lower performance (once socio-economic context is accounted for), there is significant variation across different types of activities and no clear pattern to suggest that some types of activities are more beneficial than others. Students in Sweden also make frequent use of digital resources to obtain feedback on their work, be it directly from their teachers or indirectly through digital practice exercises. In contrast to enquiry-based activities, evidence clearly suggests that students who receive frequent feedback tend to perform worse. This is likely explained by the fact that feedback is targeted at under-achieving students and those who are most likely to benefit from further support. It should not be interpreted as evidence of a negative impact of these practices.
Digital distractions in mathematics lessons
Despite improvements in recent years, many students in OECD countries continue to suffer from a disciplinary climate that is not conducive to learning (see Chapter 4 for a detailed discussion). The use of digital devices in lessons risks exacerbating these problems. Students who use digital devices can easily be tempted to multitask, shift their attention to other information or software available on the devices, or use their Internet browsers for non-academic activities when using these devices (Amez and Baert, 2020[28]; Beland and Murphy, 2016[29]; OECD, 2023, p. 186[14]). Evidence from PISA confirms that students who frequently use digital devices in mathematics lessons are more likely to report digital distractions (OECD, 2023[14]). By threatening to distract students and their peers, the use of digital devices thus introduces new challenges for teachers to address, adding to “traditional” disciplinary problems, such as noisy or disorderly classrooms (OECD, 2023[14]). The prevalence of digital distractions and its association with students’ performance in Sweden are analysed in the following sub-sections.
International comparison and differences between student groups
In PISA 2022, more than a quarter of students across OECD countries reported that students in their mathematics classes are getting distracted by using digital resources or by other students’ use of digital resources in most or every lesson (OECD, 2023[14]). In Sweden, the reported level of digital distractions is slightly above the OECD average, as in the other Nordic countries whose students reported using ICT intensively in lessons. 36.9% of Swedish students reported that students get distracted by using digital resources in most or every mathematics lesson. This is above the OECD average of 30.5%, and the averages of Denmark (31.5%) and Norway (31.2%), but below the average in Finland (40.6%). Likewise, 29.2% of Swedish students reported that students get distracted by other students who are using digital resources, in most or every mathematics lesson (above the OECD average of 25.2% and more than in Denmark [25.1%], Finland [23.2%] or Norway [24.6%]) (OECD, 2023[14]).
It should be noted that the PISA questionnaire asked students to report whether students in their lessons were distracted by devices, rather than whether they were distracted themselves. The perceived level of classroom distractions caused by students’ use of their digital resources vary little by students’ gender or socio-economic status. Only immigrant students reported that students in their lessons are suffering slightly less frequently from these distractions than non-immigrant students (32.3% vs. 38.3% reported that students get distracted in most or every mathematics lesson). When it comes to distractions caused by other students’ devices, girls were slightly more likely than boys to frequently encounter this problem (30.6% vs. 27.8%), with no significant differences across other student groups.
Association with students’ performance
Frequent digital distractions are associated with significantly lower performance across OECD countries. Distractions by other students’ use of digital resources is among the disciplinary aspects in the PISA questionnaire that show the strongest association with mathematics performance (OECD, 2023[14]). On average across OECD countries, those who reported that their classmates are getting distracted by their peers’ use of digital resources (e.g. smartphones, websites and apps) in every or most mathematics lessons score 27 points lower in mathematics. In Sweden, this association is slightly less pronounced but remains significant and negative at -16 points. The association in other Nordic countries is also negative, but smaller than on average across OECD countries (Denmark: -13 points; Finland: -18 points; and Norway: -8 points).
After accounting for students’ and schools’ socio-economic profile, the negative association between distractions and performance is diminished but remains statistically significant across OECD countries (from a 28-point to a 19-point reduction), which suggests that the relationship between distractions and performance is partly explained by socio-economic context. In Sweden, the association is largely unchanged after accounting for socio-economic context (diminishing only slightly from a 16-point to a 15-point decrease), which is consistent with the observation that both advantaged and disadvantaged students reported that students in their lessons suffered from distractions to a similar extent (Annex Table 3.A.6a).
By contrast, students getting distracted by their own use of digital resources appears to be less of a problem in the Nordic countries. While those who reported that students in their class get distracted by using digital tools in most or every mathematics lesson score 16 points less on average across OECD countries, the association was not significant in either Sweden, Denmark, Finland or Norway (Figure 3.17). This lack of a statistically significant association was observed across all student groups (i.e. regardless of students’ gender, socio-economic status, immigrant background or school location). It also persisted after accounting for students’ and schools’ socio-economic profile (Annex Table 3.A.6b).
Difference in mathematics performance associated with students reporting that students get distracted in "every or most mathematics lessons"
Note: Statistically significant differences are shown in a blue tone. Each bar represents the coefficient from a bivariate regression of mathematics performance on the variable. No control variables are included in these models. Results reflect associations, not causal effects.
Source: OECD (2023[15]), PISA 2022 Database, https://www.oecd.org/en/data/datasets/pisa-2022-database.html.
Conclusion
Students in Sweden are more frequently distracted by digital devices than the average student in OECD countries, which is not surprising, considering the intensive use of ICT in Swedish schools. This problem is reported by students across social groups and those who report that students in their mathematics lessons are getting distracted by other students’ use of digital resources have a significantly lower mathematics performance. In contrast to the pattern observed in other OECD countries, students getting distracted by their own use of digital resources does not appear to be associated with reduced mathematics performance in Sweden or any of the Nordic countries. Further efforts to reduce distractions caused by other students’ digital devices could thus be an important lever to raise students’ performance. Furthermore, finding effective ways to limit digital distractions may be conducive to students’ well-being. For example, students who perceived the climate in their mathematics lessons to be less disruptive were also less anxious about mathematics in all PISA countries and economies (OECD, 2023[14]).
Students’ use of digital resources for school activities outside of the classroom
Students can use ICT for a range of school-related activities outside the classroom, for example to check their grades and results or to communicate with their teachers. The PISA index of the use of ICT for school activities outside of the classroom captures how frequently students reported using digital resources to engage such activities (OECD, 2024[16]).10 The index and its association with students’ performance are analysed in the following sub-sections.
International comparison and differences between student groups
In PISA 2022, Swedish students reported using digital resources for school activities outside of the classroom at a significantly higher rate than students on average across OECD countries, scoring 0.22 on the relevant PISA index. The level of engagement of Swedish students was also significantly higher than that reported by students in Finland (-0.07), but below the high levels seen in Denmark (0.38). In Sweden, boys reported spending less time on school-related activities outside the classroom than girls (0.15 vs. 0.30 in the index). This difference between boys and girls of -0.15 is similar to the gender gap observed on average across OECD countries (-0.19), in Denmark (-0.09) and in Finland (-0.12). There is also a significant socio-economic gap, with advantaged students reporting that they spend significantly more time on school-related activities outside the classroom than disadvantaged students (0.36 vs. 0.05 in the index). This gap (0.30) is large and significant, similar to the one observed on average across OECD countries (0.33), in Denmark (0.27) and in Finland (0.26). No significant difference in the PISA index is observed between immigrant and non-immigrant students in Sweden.
Association with students’ performance
In contrast to leisure activities, the use of digital resources for school activities outside of the classroom tends to be associated with modest increases in students’ performance (Figure 3.18). In Sweden, a one-unit increase in the index of use of ICT for school activities outside of the classroom is associated with a 6-point increase in mathematics performance across the student population. This is in line with the OECD average of 7 points and Denmark’s 6-point increase. In Finland, the association is not statistically significant. This positive association is observed (though not always statistically significant) across all subgroups, apart from advantaged students. Although the difference in the association between advantaged and disadvantaged students is not statistically significant in Sweden, it is in line with the OECD average, where the performance of advantaged students benefited less from the time spent using ICT for school activities outside the classroom than that of disadvantaged students (2 points vs. 4 points).
A multivariate analysis shows that the relationship between the use of digital resources for school activities outside of the classroom and students’ performance in Sweden is largely explained by socio-economic factors. After accounting for students’ and schools’ socio-economic profile, a one-unit increase in the index is only associated with a 0.3-point decrease in mathematics performance and no longer statistically significant (compared with a significant 6-point increase in the bivariate analysis) (Annex Table 3.A.7). This suggests that, for students in similar socio-economic contexts, a more frequent use of digital resources for school activities outside of the classroom is not associated with any difference in mathematics performance. The same is true on average across OECD countries, where accounting for ESCS reduces the association from 7 points to 0.4.
Difference in mathematics performance associated with a one-unit increase in the index (student-reported), by student and school characteristics in Sweden
Note: Statistically significant differences are shown in a blue tone. Each bar represents the coefficient from a bivariate regression of mathematics performance on the index, estimated separately for each subgroup. No control variables are included in these models. Results reflect associations, not causal effects.
Source: OECD (2023[15]), PISA 2022 Database, https://www.oecd.org/en/data/datasets/pisa-2022-database.html.
Conclusion
Swedish students more frequently use digital resources for school-related activities outside the classroom (e.g. communicating with their teachers, uploading and sharing their work, or browsing the Internet for schoolwork) than the OECD average. In Sweden, girls engage in these activities more frequently than boys and disadvantaged students more frequently use ICT for school-related activities than their disadvantaged peers. (Similar gaps were observed across OECD countries). While engaging in these tasks is associated with modestly higher mathematics performance in Sweden, the association is largely explained by the fact that socio-economically advantaged students make greater use of ICT for school-related activities outside the classroom.
Students’ self-reported competency in using digital resources
Strengthening students’ digital literacy is increasingly important to enable them to successfully navigate the complexities of the digital age and the lifelong learning that will accompany them. Students’ confidence in engaging with digital resources and learning environments may also constrain teachers’ ability to use digital technologies effectively in their instruction. The International Computer and Information Literacy Study (ICILS) – a large-scale international assessment of students’ digital literacy in Grade 8 – provides an insight into Swedish students’ computer and information literacy as well as their computational thinking skills (Fraillon, 2024[30]). Despite Swedish students’ intensive use of digital resources, ICILS 2023 showed that their digital literacy was only average compared to other participating EU/OECD countries. The results also pointed to wide gaps between high- and low-performing students (Skolverket, 2024[31]).
PISA 2022 offers a complementary perspective, providing information on students’ self-rated ability to carry out a range of tasks using digital resources, e.g. searching for relevant information online, writing or editing text for a school assignment, or creating a computer programme. The PISA index of students’ self-efficacy in digital competencies captures this self-rated assessment.11 The following sub-sections analyse Swedish students’ digital literacy, as measured by the index, and its association with mathematics performance.
International comparison and differences between student groups
On average, students in Sweden score -0.08 in the index of self-efficacy in digital competencies. This is slightly below the OECD average and students in Finland (0.02) and significantly below the average student in Denmark (0.18). Two of the most important aspects of digital literacy covered by the index are students’ ability to search for and find relevant information online and their ability to assess the quality of the information found. Both are critical for students' ability to effectively navigate and critically engage with digital content (OECD, 2024[18]).
In Sweden, 60% of students reported being easily able to search for and find relevant information online while 19% could do so with some effort (slightly below the OECD averages of 64% and 21% respectively). Likewise, 48% of Swedish students reported being easily able to assess the quality of the information found online and 31% reported being able do so with some effort (below the OECD averages of 51% and 33% respectively). As with any measure of self-efficacy, it is important to consider that students’ self-assessment is subjective and is likely to be affected by their expectations of what constitutes proficiency.
In Sweden, there are no significant differences in the self-reported digital competency of boys and girls, as is the case across OECD countries (in both Denmark and Finland, boys reported being significantly more competent than girls, with differences of 0.13 and 0.12 in the index, respectively). By contrast, advantaged students in Sweden are significantly more confident in their digital skills than disadvantaged students. The gap between the two groups of students is large (0.31) and in line with similar gaps observed in Denmark (0.29), Finland (0.31) and on average across OECD countries (0.32). While immigrant students feel less confident in their digital abilities than non-immigrant students on average across OECD countries (-0.04) and in other Nordic countries, including Denmark (-0.11) and Finland (-0.09), the gap in Sweden (-0.08) is not statistically significant.
Association with students’ performance
As is the case across OECD countries, higher self-efficacy in digital competencies is associated with increased performance in Sweden. A one-unit increase in the index of self-efficacy is associated with a 9-point increase in mathematics performance. This is in line with the 11-point increase observed in Denmark and on average across OECD countries. In Finland, the association is even stronger (17 points). This positive association is observed across student groups, though less pronounced and not statistically significant for advantaged students and those in rural areas or villages (Figure 3.19).
As discussed above, advantaged students in Sweden are more confident in their digital competency than disadvantaged students. However, a multivariate analysis shows that the observed relationship between digital competency and performance is only partly explained by socio-economic context. After accounting for students’ and schools’ socio-economic profile, a one-unit increase in the index of self-efficacy in digital competencies remains associated with a statistically significant 5-point increase in mathematics performance in Sweden. This relationship is smaller but directionally consistent with the bivariate estimate presented above (9 points) (Annex Table 3.A.8). A similar reduction in the strength of the association can be observed on average across OECD countries (from 11 to 5 points), in Denmark (from 10 to 6 points) and in Finland (17 to 12 points) once socio-economic context is accounted for. This underlines that students’ digital efficacy is an important correlate of performance, particularly for disadvantaged students, who tend to have lower self-efficacy in digital skills.
Difference in mathematics performance associated with a one-unit increase in the index (student-reported), by student and school characteristics in Sweden
Note: Statistically significant differences are shown in a blue tone. Each bar represents the coefficient from a bivariate regression of mathematics performance on the index, estimated separately for each subgroup. No control variables are included in these models. Results reflect associations, not causal effects.
Source: OECD (2023[15]), PISA 2022 Database, https://www.oecd.org/en/data/datasets/pisa-2022-database.html.
Insights from international practice
Developing digital literacy and the ability to distinguish between reliable and unreliable information online is essential in today's digital age (Burns and Gottschalk, 2020[32]). These skills are also an important condition for students to effectively engage with digital education resources both at school and at home. PISA 2022 results indicate that strengthening students’ digital literacy should remain a priority for Sweden and could be a means to raise performance, particularly among disadvantaged students. The role of teachers and schools in developing students’ digital literacy cannot be overstated. Teachers can provide students with the tools they need to critically evaluate online information and develop healthy digital habits (OECD, 2024[18]). This could ensure that students are not only proficient in using digital tools but able to navigate the digital world responsibly and effectively (Paniagua and Istance, 2018[33]). An example of central efforts to strengthen students’ digital literacy through local initiatives in Australia is presented in Box 3.2.
In Australia the Digital Literacy School Grants was a competitive grant programme, introduced in 2016, which supported Australian schools in enhancing digital literacy among students and teachers. The programme funded innovative projects to promote the development of digital skills. Successful projects in the first funding round included a school-based Technology Design Centre for peer-led teacher training, supports for implementing the Australian Digital Technologies curriculum, and the purchase of a humanoid robot to provide additional learning support for students with special education needs.
Source: (2023[1]), Shaping Digital Education: Enabling Factors for Quality, Equity and Efficiency, https://doi.org/10.1787/bac4dc9f-en, p.120; Australian Department of Education (2017[34]), Digital Literacy School Grants Round 1 - Successful Projects, https://www.education.gov.au/australian-curriculum/resources/digital-literacy-school-grants-round-1-successful-projects.
Conclusion
Swedish students rated their digital competency slightly below the OECD average, and below students in Finland and Denmark. This is consistent with results from ICILS 2023, showing that students’ intensive use of digital devices for learning need not go hand in hand with digital literacy. In PISA 2022, advantaged students in Sweden reported significantly higher levels of digital confidence than disadvantaged students, mirroring trends across OECD countries and Nordic countries. Unlike in Denmark and Finland, however, Swedish boys and girls reported similar levels of digital self-efficacy. Consistent with the pattern observed across OECD countries, greater self-efficacy in digital skills is associated with better mathematics performance in Sweden, particularly for disadvantaged students. This suggests that students’ digital literacy remains an important area of improvement in Sweden and a potential means of raising performance, particularly among disadvantaged students.
Students’ online information habits
In addition to students’ reported confidence in their ability to evaluate the quality of online information (see above), PISA asked students about their actual behaviour when engaging in online information, e.g. whether they compare different sources when searching for information online or whether they discuss the accuracy of online information with friends or other students. The PISA index of students’ practices regarding online information captures how many of these practices students reported engaging in (OECD, 2024[16]).12 The index thus complements students’ self-assessments with a more objective measure of their actions.
PISA 2022 data underline the importance of students’ online information habits. On average across OECD countries, students who critically evaluate online information exhibit a higher motivation to learn, particularly an intrinsic motivation driven by curiosity. Students who check the quality, credibility and accuracy of online information; compare different sources; or discuss the accuracy of online information are also more likely to enjoy solving problems creatively, and to enjoy developing and testing hypotheses based on what they observe (OECD, 2024[18]). PISA data also suggest that students who check the quality, credibility and accuracy of online information are more likely to be meticulous students, critical thinkers and proactive learners who make connections between what they learn and what they know (OECD, 2024[18]).
Previous surveys commissioned by the Swedish National Agency for Education (in 2021) suggest that Swedish schools need to strengthen their efforts to develop their students’ digital competency systematically, particularly in areas such as information retrieval and the critical analysis of digital sources (Skolverket, 2022[35]). Results from ICILS 2023 are in line with this assessment. As was the case for all participating countries, a substantial proportion of Swedish students did not reach a basic level (Level 2) of computer and information literacy in the test (with 25% scoring at Level 1 and 14% scoring below Level 1) (Skolverket, 2024[31]; Fraillon, 2024, pp. 142, Table 5.2[30]). However, recent student surveys have presented a more encouraging picture, showing that most Swedish students have been taught about the critical evaluation of online sources by their teachers (Skolverket, 2025[36]).
International comparison and differences between student groups
Despite the sobering results in ICILS, Swedish students score highly in the index of students’ practices regarding online information in PISA 2022 (0.22), significantly above the OECD average, on par with Denmark (0.23) and significantly above Finland (-0.08). This indicates that Swedish students tend to take care to check for the quality, credibility and accuracy of online information (OECD, 2024[18]). As is the case on average across OECD countries, Swedish girls score more highly in the index than boys (0.26 vs. 0.19) and advantaged students score more highly than disadvantaged students (0.32 vs. 0.16). These differences are statistically significant and in the same direction as those observed on average across OECD countries. Immigrant students also score more highly in the index than non-immigrant students (0.33 vs. 0.20), whereas there are no significant differences between these groups on average across OECD countries.
Association with students’ performance
Students who score highly in the PISA index of practices regarding online information tend to perform better in mathematics. On average across OECD countries, a one-unit increase in the index is associated with a 9-point increase in students’ mathematics performance in PISA. A positive association is also observed in Denmark (14 points) and Finland (6 points). In Sweden, smaller positive associations are observed for some subgroups of students – notably girls (5 points) and non-immigrant students (5 points), but the association is not statistically significant for the entire student population (Figure 3.20). After accounting for students’ and schools’ socio-economic profile in a multivariate analysis, the overall association between students’ practices regarding online information and their mathematics performance remains statistically insignificant and is further reduced (from a 3-point to a 1-point increase in performance per one-unit increase in the index) (Annex Table 3.A.9).
Difference in mathematics performance associated with a one-unit increase in the index (student-reported), by student and school characteristics in Sweden
Note: Statistically significant differences are shown in a blue tone. Each bar represents the coefficient from a bivariate regression of mathematics performance on the index, estimated separately for each subgroup. No control variables are included in these models. Results reflect associations, not causal effects.
Source: OECD (2023[15]), PISA 2022 Database, https://www.oecd.org/en/data/datasets/pisa-2022-database.html.
Conclusion
Despite scoring at the EU/OECD average in the ICILS 2023 assessment of computer and information literacy, PISA 2022 data suggest that Swedish students frequently engage in practices to check the quality, credibility and accuracy of online information. They score well above the OECD average on the relevant PISA index (similar to students in Denmark and significantly above students in Finland). Girls, advantaged students, and immigrant students in Sweden all reported more positive online information habits than their respective counterparts. Unlike in many OECD countries, the positive association between strong online information habits and mathematics performance is not statistically significant for the overall Swedish student population, though modest positive associations are seen for girls and non-immigrant students.
Students’ behaviours and feelings when engaging with digital devices
Considering Swedish students’ intensive engagement with digital devices, both at school and at home, it is particularly important to analyse and address the risks it may pose to teenagers’ well-being. Many apps and social media are designed to hold students’ attention and a high frequency of notifications can lead to fragmented sleep or become a source of distractions for students and their peers (OECD, 2025[37]). PISA therefore collected data on restrictions imposed on the use of digital devices at school and asked students whether they adopt self-regulating behaviours, such as turning off notifications. While the evidence linking digital technology use directly to mental health problems is inconclusive, a significant share of students in OECD countries reported experiencing nervousness or anxiety related to the use of digital devices and their deprivation thereof. For at least some adolescents, issues like excessive screen time, “hyperconnectivity”, exposure to inappropriate online content or cyber-bullying are likely to have a negative impact on their well-being (OECD, 2025[37]).
International comparison and differences between student groups
Students in PISA were asked how frequently they engage in a range of self-regulating behaviours (like turning off notifications in class or at night) and whether they feel pressure or nervousness related to their engagement with digital devices. Almost half of students in Sweden (47.6%) reported turning off notifications from social networks and apps on their digital devices during class “all or almost all of the time” (compared to 45.5% on average across OECD countries). Likewise, 41.6% of Swedish students reported turning off notifications on their digital devices when they go to sleep “all or almost all of the time” (compared to 45.5% on average across OECD countries) (Figure 3.21). Nevertheless, more than a fifth of Swedish students reported never turning off their notifications in class (21.6%) or when they go to sleep (30.0%), close to the OECD averages of 25.2% and 30.4%, respectively. Furthermore, 59.8% of Swedish students reported keeping digital device near them to answer messages when they are at home all or almost all of the time (more than the OECD average of 50.7%) and 19.7% reported having their digital device open in class so they can take notes or search for information (compared to the OECD average of 15.6%).
When it comes to negative feelings associated with digital devices, most students in Sweden reported never or almost never feeling pressured to be online and answer messages when they are in class (60.9%, compared to the OECD average of 57.7%). Nevertheless, a small share of students reported feeling this way all or almost all of the time (4.8%, compared to the OECD average of 5.3%). Likewise, most students in Sweden reported never or almost never feeling nervous or anxious when they do not have their digital device near them (47.6.9%, compared to the OECD average of 48.5%). However, again, a small but significant share reported feeling this way all or almost all of the time (9.9%, compared to the OECD average of 9.0%) (Figure 3.21).
Percentage of students who reported feeling or acting the following ways "all or almost all of the time"
Source: OECD (2023[14]), PISA 2022 Results (Volume II): Learning During – and From – Disruption, https://doi.org/10.1787/a97db61c-en, Table II.B1.5.46.
There are some notable socio-economic differences in the way students reported behaving and feeling when using digital devices in Sweden. Notably, girls are significantly more likely than boys to regularly turn off notifications from social networks and apps on their digital devices during class (75% vs. 61%) and when they go to sleep (66% vs. 54%). By contrast, girls were more likely than boys to report frequently feeling nervous or anxious when they do not have their digital device near them (35% vs. 23%). Similarly, disadvantaged students in Sweden are less likely to engage in self-regulating behaviour using digital devices. They are significantly less likely than advantaged students to regularly turn off notifications from social networks and apps on their digital devices during class (65% vs. 73%) and when they go to sleep (54% vs. 67%). Disadvantaged students were also more likely than advantaged students to report frequently feeling pressured to be online and to answer messages when they are in class (21% vs. 13%).
Association with students’ performance
The way students reported engaging with digital devices is strongly associated with their performance. Students who reported that they engage in self-regulating behaviour, like turning off notifications during classes or at night, fare significantly better in mathematics than those who do not. In PISA, Swedish students who reported turning off notifications from social networks and apps on their digital devices during class “more than half of the time” score 32 points higher than those who did not. This is line with the OECD average (34 points) and broadly in line with the associations observed in Denmark (21 points), Finland (28 points) and Norway (39 points). Likewise, students who reported turning off notifications when going to sleep “more than half of the time” score 25 points higher than those who did not. This is broadly in line with the OECD average (29 points) and the associations observed in Denmark (19 points), Finland (29 points) and Norway (32 points) (Figure 3.22).
Students who reported keeping digital device near them to answer messages when they are at home “at least half of the time” score significantly higher in mathematics than those who did not in Sweden (25 points, significantly more than the OECD average of 10 points), although the latter are a minority, accounting for only 9% of all students. Similarly strong associations are observed in Finland (17 points) and Norway (33 points), but not in Denmark. Students who reported often having digital device open in class so they can take notes or search for information do not fare significantly better or worse than students who did not. This is in contrast to a small negative association across OECD countries (-6 points), a stronger positive association in Denmark (17 points) and a significant negative association in Finland (-18 points) (Figure 3.22).
Swedish students’ feelings of anxiety or nervousness related to digital devices are associated with significantly lower performance, as is the case across OECD countries. Students who reported feeling pressured to be online and answer messages when they are in class “at least half of the time” (9.1% of all students) score -58 points worse than those who did not. This negative association is even stronger than the OECD average (-41 points) and similar to those observed in other Nordic countries like Denmark (-56 points), Finland (-52 points) and Norway (-56 points). Likewise, Swedish students who reported feeling nervous or anxious when they do not have their digital device near them “at least half of the time” (17.4% of all students) score -36 points worse than those who did not. This is similar to the associations seen on average across OECD countries (-33 points) and in Nordic peers (Figure 3.22).
Difference in mathematics performance associated with students reporting that they engage in the following behaviours or have the following feelings at least "more than half of the time"
Note: Statistically significant differences are shown in a darker tone. Each bar represents the coefficient from a bivariate regression of mathematics performance on the variable. No control variables are included in these models. Results reflect associations, not causal effects.
Source: OECD (2023[15]), PISA 2022 Database, https://www.oecd.org/en/data/datasets/pisa-2022-database.html.
Given the systematic differences in students' behaviours and feelings when using digital devices and their significant association with mathematics performance, they can be a source of inequities. Another potential source of inequities are group-specific differences in the association of these behaviours and feelings with students’ performance.
Differences in mathematics performance associated with students reporting that they engage in the following behaviours or have the following feelings at least "more than half of the time" in Sweden
Note: The estimated difference in mathematics performance is based on separate bivariate regressions for each subgroup. Statistically significant differences in performance (or differences between subgroups) are highlighted in bold. Results reflect associations, not causal effects.
Source: OECD (2023[15]), PISA 2022 Database, https://www.oecd.org/en/data/datasets/pisa-2022-database.html. Accessed on 19 August 2025.
Broadly speaking, the behaviours described above have a similar association with performance for both boys and girls. Self-regulating behaviours, for example, are associated with better performance regardless of students’ gender. The only notable gender difference was observed for self-reported feelings of nervousness or anxiety when students do not have their digital device near them. Such feelings have a more negative association with performance for boys than for girls. (At the same time, girls more frequently reported having such feelings) (Table 3.4).
There are some notable differences in the association between behaviours and feelings related to digital resources and the performance of advantaged and disadvantaged students. Like girls, advantaged students who reported feeling nervous or anxious when they do not have their digital device near them fare significantly worse in mathematics than disadvantaged students. Similarly, advantaged students benefit significantly more than disadvantaged students from turning off notifications on their digital devices when going to sleep. By contrast, disadvantaged students who reported frequently keeping their device near them to answer messages when they are at home or having them open in class perform significantly better than advantaged students reporting the same behaviours (Table 3.5).
Differences in mathematics performance associated with students reporting that they engage in the following behaviours or have the following feelings at least "more than half of the time" in Sweden
Note: The estimated differences in mathematics performance are based on separate bivariate regressions for each subgroup. Statistically significant differences in performance (or differences between subgroups) are highlighted in bold. Results reflect associations, not causal effects.
Source: OECD (2023[15]), PISA 2022 Database, https://www.oecd.org/en/data/datasets/pisa-2022-database.html.
Insights from international practice
In light of the problems associated with digital distractions in the classroom and students’ engagement with digital devices, an increasing number of OECD countries, including Sweden, have moved to reduce the presence of students’ phones in schools (Skolverket, 2022[35]). At the same time, it is important to ensure that banning mobile phones at school does not inadvertently lead students to engage in unhealthy digital device use at home. PISA data show, for example, that students are less likely to turn off notifications from social networks and apps on their devices before going to sleep when mobile phones are banned at school (OECD, 2023[14]). To encourage students, parents and educators to promote healthy habits when using digital devices, many OECD countries have issued guidelines on children’s screen time or engagement with digital resources. This is also the case in Sweden, where the Public Health Agency has issued recommendations to parents concerning their children’s use of digital devices (Folkhälsomyndigheten, 2025[38]). International recommendations vary, reflecting diverse approaches to balancing the benefits and risks associated with screen time and taking account of the limited evidence to suggest that engagement with digital devices is inherently harmful to child health at any age (OECD, 2025[37]). While some guidelines focus on limiting screen time to promote healthy sleep and physical development in children, others also address how to mitigate the risks of the digital world and promote good practices. Examples of such guidelines from Australia, Germany, Canada and the United States are presented in Box 3.3.
Australia and Germany: Public recommendations on screen time for children
Australia and Germany provide examples of central recommendations focussed on promoting children’s physical activity and development. In Australia, national guidelines for screen time are included in the Australian Department of Health's Physical Activity and Sedentary Behaviour Guidelines, which recommend that children under the age of 2 should avoid all screens except for video-chatting and that children aged 2 to 5 years should be limited to less than one hour per day. For children aged 5 to 17 years, screen time should be limited to no more than two hours of recreational screen time per day, in addition to using screens for educational purposes. In Germany, the German Federal Ministry of Health recommends reducing children’s screen time to a minimum: No screen time for infants and toddlers; as little as possible and a maximum of 30 minutes per day for preschool children; a maximum of 60 minutes per day for primary school children and up to a maximum of two hours per day for adolescents.
Canada: Guidance for families on appropriate screen time practices
The Canadian Pediatric Society takes a broader approach, providing recommendations to minimise screen time for young children but also proposing strategies for families to mitigate associated risks, to be mindful about the use of screen time as a family, and to model healthy screen use as adults (Canadian Pediatric Society, 2022[39]). The recommendations include:
Minimise screen time: Screen time is not recommended for children under 2 years, except for video-chatting with caring adults. For children aged 2 to 5 years, limit screen time to about 1 hour or less per day; limit recreational screen time to no more than two hours per day for older children. More broadly it is recommended to ensure that screen time is not a routine part of childcare for children under 5 years, and maintain screen-free times, particularly during family meals and book-sharing. Avoid screens at least 1 hour before bedtime to prevent stimulation and melatonin suppression.
Mitigate screen time risks by being present and engaged during screen use, co-viewing with children to promote digital media literacy and help them recognise and question advertising and stereotypes; prioritising educational, age-appropriate, and interactive programming, and encourage creative activities using screen devices; monitoring media use by creating playlists or selecting appropriate channels; limiting exposure to advertising and commercial content.
Enhance family mindfulness about screen use by conducting a self-assessment of screen habits and developing a family media plan; prioritising shared family media use over solitary screen time for children; and encouraging older siblings to mentor younger children’s digital use.
United States: Evidence-based guidance on social media and video viewing
In the United States, a report from an expert committee of the National Academies of Sciences, Engineering, and Medicine found a lack of strong evidence that social media causes population-level changes in adolescent health. Instead, it found heterogeneous effects, including risks, benefits and neutral impacts (National Academies of Sciences, 2023[40]). Besides further research on these relationships, the committee proposed platform design changes for greater transparency, expanded education for youth and adults, and measures to address online harassment and bullying.
Recent guidance by the American Psychological Association on healthy teen video viewing highlights that video contents can support personal growth but also pose risks, particularly through platform features like autoplay and algorithm-driven recommendations. Parents are encouraged to set clear rules, engage in discussions about content, and model positive viewing habits to foster critical thinking and healthier choices in teens. The guide also underscores the role of education and policy in promoting media literacy and safeguarding youth from harmful content. Schools can teach video literacy as part of broader media education, while regulators and platforms should enhance moderation and age-appropriate labelling. Collaboration between parents, educators and policymakers is key to ensuring that teens develop a constructive relationship with video content (American Psychological Association, 2024[41]).
Source: Reproduced from OECD (2025[37]), How's Life for Children in the Digital Age?, https://doi.org/10.1787/0854b900-en, Box 5.4.
Conclusion
Swedish students’ self-regulation habits when engaging with digital devices are in line with the OECD average. A little under half of Swedish students reported almost always turning off notifications during class and when going to sleep. However, more than 20% of students never turn off notifications at all, and many (59.8%) keep their devices nearby at home to answer messages. As is the case in most OECD countries, turning off notifications is strongly associated with students’ performance in Sweden and there is a risk of inequities since boys and disadvantaged students are less likely to engage in self-regulating behaviours using digital devices. At the same time, the adoption of self-regulating behaviours appears to be correlated with greater improvements in performance for advantaged students. One explanation for the observed association between students’ behaviour and their performance could be that students who switch off notifications from social networks and apps on their digital devices during class are less likely to report distractions (OECD, 2023[14]).
Although most Swedish students reported rarely feeling pressured to be online, a small but significant minority (5%) reported feeling this way all or almost all of the time. Likewise, less than half of Swedish students reported never or almost never feeling nervous or anxious when they do not have their digital device near them and 10% reported feeling this way all or almost all of the time. While these results are close to the OECD average, they are nevertheless concerning and students who reported having these negative feelings score significantly lower in mathematics. Both girls and disadvantaged students are more likely to report feeling some forms of anxiety or pressure when interacting with digital devices. However, boys’ performance appears to suffer more when they report feeling such anxiety or nervousness.
Students’ engagement in leisure activities using ICT
While Swedish and other students in Nordic countries make significantly more use of digital resources for learning activities at school than their peers in other OECD countries, the time that Swedish students spend using digital resources for learning outside of school as well as for leisure activities is closer to the OECD average (see Figure 3.5). On a typical day, Swedish students reported spending 1.1 hours using digital resources for leisure activities at school (the same as the OECD average), 2.7 hours on leisure activities before and after school (close to the OECD average of 2.6 hours) and 3.8 hours on leisure activities on the weekend (close to the OECD average of 3.9 hours) (OECD, 2023[14]).
While the frequent use of ICT for leisure activities at school is associated with lower mathematics performance in Sweden and across OECD countries (see Figure 3.9 discussed above), the time students reported spending using ICT for leisure before and after school and on weekends is associated with slight increases in mathematics performance (before and after accounting for ESCS) both in Sweden and across OECD countries.13 In Sweden, only the excessive use of ICT for leisure activities (i.e. more than 60 hours per week) is associated with significant reductions in students’ performance (OECD, 2023[14]). ICT use for leisure encompasses a highly heterogeneous set activities, including passive and active practices (such as playing video games), as well as solitary and social activities (such as communicating with friends via social networks). To provide a more granular perspective on students’ ICT use for leisure, the following sub-sections explore Swedish students’ engagement in a selection of such activities and their distinct associations with mathematics performance.
International comparison and differences between student groups
In addition to asking students about the overall time they spend on digital resources for leisure, PISA also asked students how much time they spend on specific leisure activities during a typical weekday and a typical weekend day. Like many students across OECD countries, Swedish students spend a significant part of their leisure time engaged with digital resources. On a typical weekday, Swedish students reported spending 2.4 hours playing video games, 2.7 hours browsing social networks, 2.4 hours browsing the Internet (excluding social networks) for fun, 1.4 hours looking for practical information online, 2.1 hours communicating and sharing digital content on social networks or any communication platform, 1.5 hours reading, listening to or viewing informational materials to learn how to do something (e.g. podcasts), and 1.4 hours creating or editing their own digital content. These times are close to the OECD average and those reported by students in other Nordic countries (Figure 3.23). Students’ reported engagement in digital leisure activities during a typical weekend day is similar, both in terms of its volume and the types of activities students reported engaging in.14
Based on students' reports, in hours
Note: To create measures of the average time spent on leisure activities, students' reposes (“none”, “up to 1 hour”, “more than 1 hour and up to 2 hours”, etc.) were recoded, taking the median of each time interval, and assuming 7.5 hours if the answer was “more than 7 hours” and converted to a continuous variable.
Source: OECD (2023[15]), PISA 2022 Database, https://www.oecd.org/en/data/datasets/pisa-2022-database.html.
When students are asked how much time they spend on the individual digital leisure activities listed above, the combined number of hours significantly exceeds the overall time they estimate spending on digital leisure activities, even though the question does not cover all types of digital leisure activities that students may engage in. While the overall digital leisure time derived from these individual responses should therefore be treated with caution, we may expect the question to provide us with a good sense of the relative amount of time students spend on different digital leisure activities. Students’ responses are captured in two separate PISA indices of the frequency of ICT activity on weekdays and during the weekend (OECD, 2024[16]).15 Students in Sweden score 0.04 in the index for leisure activities during weekdays, only marginally above the OECD average, above Denmark (0.00) and in line with Finland (0.04). The pattern is similar for leisure activities during weekends, where Swedish students score 0.04, only marginally above the OECD average and Finland (0.01) and in line with Denmark (0.05).
In Sweden, boys reported spending significantly more time on digital leisure activities during weekdays than girls (0.07 vs. 0.00 in the index). Disadvantaged students also spend significantly more time on digital leisure activities during weekdays than advantaged students (0.09 vs. -0.05 in the index). Immigrant students spend significantly more time on digital leisure activities during weekdays than non-immigrant students (0.12 vs. 0.01 in the index). These broad differences between student groups are in line with those observed on average across OECD countries, but they are not observed for all activities. For example, boys in Sweden (and across OECD countries) are more likely to spend substantial time playing video games, whereas girls are more likely to spend more than one hour per day browsing social networks.
Association with students’ performance
In contrast the overall time that students estimate spending on digital leisure activities, the time students spend using digital resources for specific leisure activities during weekdays is significantly negatively associated with their performance. In Sweden, a one-unit increase in the index is associated with a 23-point decrease in mathematics performance. This association is similar to the OECD average (-20 points) and the one observed in Finland (-20 points) and slightly weaker than the one observed in Denmark (-29 points). This negative association is observed across all subgroups, apart from students in rural areas or villages for whom the association is not significant (Figure 3.24). The negative association between ICT-based leisure activities on weekdays and performance is more pronounced for advantaged than for disadvantaged students (-30 points vs. -13 points), for non-immigrant than for immigrant students (-24 points vs. -16 points) and in cities vs. rural areas (-24 points vs. -9 points). The time that students reported spending on digital leisure activities during weekends shows a negative association of a similar magnitude (-20 points in Sweden, compared to -17 points on average across OECD countries [not shown here]).
Difference in mathematics performance associated with a one-unit increase in the index (student-reported), by student and school characteristics in Sweden
Note: Statistically significant differences are shown in a blue tone. Each bar represents the coefficient from a bivariate regression of mathematics performance on the index, estimated separately for each subgroup. No control variables are included in these models. Results reflect associations, not causal effects.
Source: OECD (2023[15]), PISA 2022 Database, https://www.oecd.org/en/data/datasets/pisa-2022-database.html.
A multivariate analysis shows that the strong negative association between the time that students spend using ICT for leisure activities and their mathematics performance holds when controlling for socio-economic context. After accounting for students’ and schools’ socio-economic profile, a one-unit increase in the index of the frequency of ICT activity on weekdays remains associated with a 21-point decline in mathematics performance (Annex Table 3.A.10a) and a one-unit increase in the index of the frequency of ICT activity on weekends remains associated with an 18-point decline in mathematics performance (Annex Table 3.A.10b). The associations are only slightly smaller in magnitude than the bivariate estimates presented above (-23 points and -20 points respectively). The indices also explain a significant amount of the performance variation in mathematics (5.7% and 4.4% respectively – slightly more than the OECD averages of 5.1% and 3.9%).
Conclusion
Like many students across OECD countries, Swedish students spend a significant amount of their leisure time engaged with digital resources. The association between students’ engagement in digital leisure activities and their mathematics performance varies across different measures, which makes it hard to interpret. While students’ estimates of the overall time spent on digital leisure activities is associated with slight increases in performance, a more granular measure comprised of students’ estimates of the time spent on individual activities is associated with significantly diminished performance, even after controlling for socio-economic context. Boys, disadvantaged students and immigrant students spend more time using digital resources for leisure activities, which may be a source of inequities, given its negative association with performance. At the same time, digital leisure activities appear to be associated with a greater reduction in performance among student groups that tend to spend less time on them.
Schools’ digital education policies and practices and students’ performance in PISA 2022 (School Questionnaire)
Data from the PISA 2022 School Questionnaire can further enrich our understanding of the use of digital resources in Swedish schools by complementing students’ views with those of their principals. School leaders’ responses provide additional information on the availability and quality of digital resources in their schools, their teachers’ preparedness for digital education, as well as the policies and practices they adopt to promote an effective the use of digital resources (OECD, 2023[14]). As is the case for the Student Questionnaire, data derived from the PISA School Questionnaire need to be interpreted with caution, due to their self-reported nature.
Access to and quality of ICT resources at school (principals’ reports)
As discussed above, access to high-quality ICT resources in schools – while not a guarantee for impactful digital teaching and learning – is a necessary condition for teachers to make effective pedagogical use of digital resources. Swedish students express a high level of satisfaction with the availability and quality of digital resources in their schools and those who reported having better access score higher in mathematics. To triangulate these results, the following sections complement the analysis with principals’ views on the adequacy of their schools’ ICT infrastructure, derived from the PISA School Questionnaire.
International comparison and differences across schools
Like their students, Swedish principals reported very high levels of satisfaction with their schools’ access to digital resources and their quality. 80.2% of Swedish students attend schools whose principals reported that a lack of digital resources does not hinder their schools’ capacity to provide instruction “at all”. This proportion is well above the OECD average (47.2%) and exceeds that of the Nordic peers Denmark (76.2%), Finland (48.1%) and Norway (67.4%). Likewise, 72.8% of Swedish students attend schools whose principals reported that their schools’ capacity to provide instruction is “not at all” hindered by inadequate or poor-quality digital resources. Although slightly lower and in line with that of Denmark (73.7%), this proportion still significantly exceeds the OECD average (45.5%) and the shares of Finland (51.3%) and Norway (56.4%) (OECD, 2023[14]).
Very few Swedish students attend schools whose principals reported that their capacity to provide instruction is hindered “to some extent” or “a lot” by a lack of digital resources (3.5%, compared to the OECD average of 23.9%) or by inadequate or poor-quality digital resources (6.7%, compared to the OECD average of 24.6%), even in comparison to the other Nordic countries (OECD, 2023[14]). In addition, there are no notable inequities in the schools who are affected by a lack of digital resources in Sweden, as perceived by their principals. While principals of disadvantaged schools were more likely to report having inadequate or a lack of digital resources on average across OECD countries, this is not the case in Sweden. Likewise, while inadequate and lacking digital resources were more often reported in rural schools on average across OECD countries, in Sweden these regional differences are not statistically significant. Neither are public schools more likely to be affected by inadequate or lacking digital resources in Sweden, as is the case on average across OECD countries (in this respect, Sweden differs from both Denmark and Finland). However, as discussed above, students’ responses present a somewhat different picture, suggesting that disadvantaged students do suffer from lower quality digital resources.
Association with students’ performance
Students attending schools with insufficient digital resources (based on their principals’ perceptions) do not appear to perform worse in mathematics in Sweden. On average across OECD countries, students in schools whose principals reported that a lack of digital resources hinders their schools’ capacity to provide instruction to some extent or a lot score 10 points lower in mathematics and those in schools whose principals reported having inadequate of poor-quality digital resources score 9 points lower. In both cases, however, the association disappears once students’ and schools’ socio-economic profile is accounted for. In Sweden, the association is not statistically significant either before or after controlling for socio-economic context (Figure 3.25) and no statistically significant differences in the association are observed between subgroups of students (i.e. by gender, socio-economic status, immigrant background or school location). This reflects the equitable distribution of digital resources in Sweden, discussed above. However, it is also important to bear in mind that a very small proportion of principals reported facing issues with the quality of digital resources in Sweden, rendering the statistical analyses less reliable.
Difference in mathematics performance associated with principals reporting that their schools’ capacity to provide instruction is hindered "to some extent or a lot" by the following
Note: Statistically significant differences are shown in a blue tone. Each bar represents the coefficient from a bivariate regression of mathematics performance on the variable. No control variables are included in these models. All values after accounting for students' and schools' socio-economic profile are statistically insignificant. Results reflect associations, not causal effects.
Source: OECD (2023[15]), PISA 2022 Database, https://www.oecd.org/en/data/datasets/pisa-2022-database.html.
Conclusion
Very few principals in Sweden reported that their schools’ capacity to provide instruction is hindered by a lack of or inadequate digital resources. Furthermore, digital resources are distributed very equitably in Sweden and the other Nordic countries, which makes them stand out in comparison with most OECD countries. (The few schools who suffer from a lack of or inadequate digital resources are not more disadvantaged on average.) Students’ performance also does not appear to be correlated with the quality of digital resources, as reported by principals, in Sweden and other Nordic countries, once socio-economic context is taken into account. However, the very small proportion of Swedish principals who reported facing serious problems due to insufficient digital resources may render this result less reliable and it should be noted that problems with the quality and accessibility of digital resources as reported by students (and as captured by the PISA index of the quality of access to ICT) do have a negative association with performance.
Teachers’ preparedness for digital education
Teachers are widely recognised as playing a pivotal role in ensuring that digital resources are used effectively to promote students’ performance. To do so, teachers need to be well prepared and supported to use digital resources to enhance their work (OECD, 2025[8]). As discussed above, the great majority of Swedish students (79%) agree that their teachers have the necessary skills to use digital devices during instruction (see Figure 3.3) and those who agree with this statement tend to perform 8 points higher in mathematics. To triangulate these results, the following sections complement the analysis with principals’ views on their teachers’ preparedness for digital learning as captured by the PISA index of schools’ preparedness for digital learning (OECD, 2024[16]).16 The index covers principals’ views on their teachers’ technical and pedagogical skills, teachers’ professional resources, incentives and time to engage in effective teaching with digital resources, as well as the availability of sufficient technical assistant staff.
International comparison and differences across schools
Principals in Sweden reported that their teachers are exceptionally well prepared to engage in digital education. Sweden’s score in the index (0.60) significantly exceeds the OECD average (-0.02) and that of Finland (0.05) (OECD, 2023[14]). 94.6% of Swedish students attend schools whose principals agree or strongly agree that their teachers have the necessary technical and pedagogical skills to integrate digital devices in instruction. This significantly exceeds the OECD average (87.6%) and constitutes a significant improvement since 2018 (up from 72.0%). Likewise, the share of students whose principals reported that teachers have sufficient time to prepare lessons integrating digital devices increased significantly from 82.3% in 2018 to 90.8% in 2022. This is significantly above the OECD average of 58.8% (OECD, 2023[14]). Nevertheless, not all schools in Sweden are equally well prepared for digital education. As is the case on average across OECD countries, principals of schools in cities reported having a higher level of preparedness than those in rural areas or villages (0.82 vs. 0.35 in the index). The same is true for socio-economically advantaged vs. disadvantaged schools, although the difference is not statistically significant.
Association with students’ performance
Although schools’ preparedness is considered an important condition for the effective use of digital resources, Swedish principals’ perception of their teachers’ preparedness for digital learning is not associated with students’ performance. On average across OECD countries, a one-unit increase in the index of preparedness for digital learning is associated with a small 4-point increase in students’ mathematics scores in PISA. In Sweden, the association is not statistically significant. After accounting for students’ and schools’ socio-economic context in a multivariate analysis, the association between the index of preparedness for digital learning and students’ performance also turns insignificant across OECD countries and remains insignificant in Sweden and Finland (Annex Table 3.A.11).
Insights from international practice
Only effective forms of professional learning on the use of digital resources will lead to meaningful improvements in the classroom. Research evidence has supported a move away from passive, standardised and one-off training formats towards more active, collaborative and sustained forms of professional learning. This includes school-based programmes that allow teachers to improve their practice by co-operating with their peers (Opfer, 2016[42]), individualised instructional coaching carried out by trained teacher coaches (Blazar and Kraft, 2015[43]; Kraft and Blazar, 2017[44]), or matching effective teachers with less effective ones (Papay et al., 2016[45]). Some OECD countries are actively supporting teachers’ engagement in such collaborative forms of professional learning on the use of digital resources. For example, the OECD’s Policy Survey on School Education in the Digital Age indicates that 17 of 35 participating jurisdictions provide central support for teachers’ engagement in peer learning on the use of digital resources (e.g. mentoring schemes or communities of practice) (OECD, 2025[8]). Two examples of such initiatives in Finland and Luxembourg are presented in Box 3.4.
Finland: Promoting teachers’ digital skills through a network of tutor teachers
Finland has supported peer-learning activities to strengthen teachers’ use of digital technologies and pedagogical innovation early on. Between 2016 and 2018, Finland invested EUR 23 million to build a network of tutor teachers with the goal to reach all 2 500 basic education schools in the country. Tutor teachers were paired with basic education teachers to provide peer support, one-to-one guidance, as well as team teaching with a particular focus on the use of digital resources for instruction (OPH, 2018[46]). An evaluation of the programme highlighted the broad availability of tutor teacher services across schools in the country and a positive self-reported impact on teachers’ pedagogical digital skills (Pennanen et al., 2021[47]).
Luxembourg: Training specialised teachers to strengthen their peers’ digital skills
In 2020/21, Luxembourg started recruiting a pool of teachers specialised in digital skills (enseignant spécialisé en compétences numériques, I-CN) to support their peers in the use of digital resources and in teaching ICT skills and media literacy. The specialised I-CN teachers must hold a master's degree in media didactics and ICT and work with teachers across multiple schools. The I-CNs’ work is co-ordinated by the regional directorates for basic education and they are provided with training and a network by the Service for the Co-ordination of Educational and Technological Research and Innovation (Service de co-ordination de la recherche et de l’innovation pédagogiques et technologiques, SCRIPT) (Luxembourg Ministry of Education, 2020[48]).
Source: OECD (2025[8]), “Preparing teachers for digital education: Continuing professional learning on digital skills and pedagogies”, OECD Education Policy Perspectives, No.122, https://doi.org/10.1787/af442d7a-en, p. 12.
Conclusion
Swedish principals’ reports confirm students’ assessment that their teachers are very well prepared to integrate digital resources into their instruction. Principals also confirmed that teachers are provided with a range of supports to facilitate their effective use of digital resources (such as professional resources to learn how to use digital devices, incentives to integrate digital tools in their instruction, and sufficient qualified technical assistant staff). However, the provision of these supports and teachers’ preparedness is not significantly associated with higher student performance. This could suggest that there is scope to further improve the effectiveness of teachers’ preparation for the use of digital resources. At the same time, the results should not be interpreted as evidence that the preparation of teachers makes no difference for their use of digital resources. In practice, it is likely that teachers who are not well prepared for the use of digital tools make less use of them, relying instead on pedagogical strategies that incorporate fewer digital resources, which may be highly effective as well.
School policies and practices concerning the use of digital resources
Education institutions and their leadership teams play a critical role in the successful digital transformation of education systems (OECD, 2023[1]). Particularly in systems with a high level of decentralisation and school autonomy, pedagogical leadership is critical to advance and sustain school improvement processes. School leaders represent an important interface between public authorities and educators, and they are vital in promoting technological innovations and creating the conditions to strengthen teachers’ capacity for digital education (e.g. through systematic collaboration and communities of practice) (Paniagua and Istance, 2018[33]). School leaders’ efforts to lead the effective use of digital resources in their schools is therefore an important factor that may influence students’ learning outcomes. Previous surveys commissioned by the Swedish National Agency for Education (in 2021) have underlined the importance of school leadership and, for example, pointed to the need for principals to work systematically on the digital transformation of their institutions, including through strategic planning and the development of their staff (Skolverket, 2022[35]). The following sub-sections draw on the PISA school questionnaire to analyse some of these school leadership practices and their association with students’ performance in more depth.
International comparison and differences across schools
In PISA 2022, principals were asked to indicate whether their schools implemented a range of policies regarding the use of digital devices, including policies to support teachers in the use of digital devices or to provide guidance on their use in mathematics lessons. In Sweden, the majority of students (61.3%) attend schools whose principals reported that their mathematics teachers are provided with professional development on the integration of digital resources in mathematics instruction. This is slightly below the OECD average (67.7%) and significantly below Denmark (78.2%) and Norway (85.8%), but above Finland (50.2%). On average across OECD countries, the provision of professional development support is more widespread in advantaged schools, schools in cities and private schools. Although similar gaps are observed in Sweden, the differences between schools with different characteristics are not statistically significant.
Comparatively few schools in Sweden have a policy on how to use ICT in mathematics instruction (e.g. specifying the amount of computer use or the use of specific computer programmes). Only 21.9% of students attend schools whose principals reported that such policies were in place, which may reflect the system’s emphasis on teachers’ pedagogical autonomy. By contrast, on average across OECD countries, 39.8% of students attend schools with policies on the use of digital devices in mathematics lessons. The share is also higher in Denmark (37.6%), Finland (52,5%) and Norway (44.6%). As for the provision of professional development, policies on the use of digital devices are – on average across OECD countries – more widespread among advantaged schools, schools in cities and private schools. Again, no such statistically significant differences between schools are observed in Sweden.
Association with students’ performance
On average across OECD countries, schools that have specific policies in place to govern the use of digital resources (based on the PISA DIGDVPOL index) tend to perform marginally better than those who do not (3 points in mathematics, per one-unit increase in the index).17 Consistent with this general pattern, students attending schools that provide mathematics teachers with professional development on the integration of digital resources in mathematics instruction perform 8 points higher in mathematics, on average across OECD countries. Although this association is diminished to 3 points after accounting for students' and schools' socio-economic profile, it remains positive and statistically significant. By contrast, no significant association between professional training on digital resources and students’ performance is observed neither in Sweden or any of the other Nordic countries (Figure 3.26).
A similar pattern can be seen for schools whose principal reported having a policy on how to use digital devices in mathematics instruction. Although students who attend such schools perform 3 points higher in mathematics than their peers on average across OECD countries, the association disappears after accounting for students' and schools' socio-economic profile. In Sweden, Denmark, Finland and Norway, no statistically significant associations between such policies on the use of digital devices and students’ performance can be observed either before or after accounting for socio-economic context (Figure 3.26).
Difference in mathematics performance associated with the principals reporting that their school has the following policies in place
Note: Statistically significant differences are shown in a darker tone. Each bar represents the coefficient from a bivariate regression of mathematics performance on the variable. No control variables are included in these models. The diamonds represent coefficients after accounting for students’ and schools’ socio-economic profile. All results reflect associations, not causal effects.
Source: OECD (2023[15]), PISA 2022 Database, https://www.oecd.org/en/data/datasets/pisa-2022-database.html.
Insights from international practice
Many OECD countries have recognised the pivotal role that school leaders play in ensuring the successful digital transformation of their schools. School leaders’ ability to take a strategic approach to digitalisation and employ their schools’ resources effectively can make a critical difference in unlocking the potential of digital resources to enhance teaching and learning. Not all school leaders are prepared to embrace and successfully meet the challenges of digitalisation, to take strategic short- and medium-term decisions amid an imperfect and rapidly evolving evidence base, and to develop new skills among the teaching workforce and the school management team alike. To avert the risk of widening inequities between schools, some countries have sought to strengthen school leaders’ capacity to plan, implement and resource a digital vision for their schools. An example of such a strategy in Wales (United Kingdom) is provided in Box 3.5.
In 2022, Wales launched a Digital Professional Learning Journey (DPLJ) which aims to support schools in developing their own digital vision and provides a framework for increasing teachers’ and school leaders’ skills and knowledge around digital education. The platform refers to a wide range of resources and learning opportunities designed to help schools plan their own Professional Learning Journey (PLJ). Building on the PLJ, the DPLJ supports school leaders in planning, developing and resourcing a digital vision for their schools. The DPLJ leadership training places a particular emphasis on budgeting for the digital transformation, as well as strategic planning, procurement, hardware and software. The platform is supported by a “360 digi Cymru” self-evaluation toolkit, which enables schools to evaluate the role of digital resources in their workplace and which can be linked with the national EdTech Funding offer for schools (Welsh Government, 2022[49]).
Source: Welsh Government (2022[49]), Digital Professional Learning Journey (DPLJ), https://hwb.gov.wales/professional-learning/career-long-professional-learning/digital-professional-learning-journey/ (accessed on 21 October 2024).
Conclusion
In comparison with other OECD countries, relatively few principals in Sweden reported offering their teachers professional development on the integration of digital resources in mathematics instruction or having a policy on the use of digital devices in mathematics instruction. This may be surprising, given the strong reliance on the use of digital resources in Swedish schools. However, neither of the two practices are significantly associated with changes in students’ performance in Sweden. By contrast, on average across OECD countries, schools that offer professional development on the use of digital devices tend to perform slightly better. As for teachers’ digital preparedness, a direct link between school-level support for digital education and students’ performance is therefore not apparent in Sweden. Nevertheless, this should not be taken as causal evidence of these practices’ lack of effectiveness. For example, it may be that they have a positive impact on other student outcomes or enhance student performance conditional on the intensive use of digital devices in schools.
PISA 2022 data highlight that students in Sweden make intensive use of digital technologies for learning, both at school and at home. Both students and school leaders express a high level of satisfaction with the quality and availability of digital resources in their schools and teachers’ preparedness to integrate digital devices in instruction. Compared with other OECD countries, Sweden also exhibits low levels of socio-economic inequity across a range of relevant indicators. Nevertheless, PISA 2022 data reveal several challenges concerning the effective us of digital education resources i. While the use of digital resources is associated with better student performance in Sweden than in other OECD countries, this is not the case for its use in mathematics lessons. Likewise, the highly digitalised learning environment is associated with a greater prevalence of digital distractions. PISA data also point to some deficits and inequities in students’ self-rated digital literacy as well as the quality of digital devices available in schools. Strategies that could be considered to address these challenges are presented below and further developed in Chapter 5 of this report.
Policy consideration 1: Strengthening teachers’ capacity to make effective pedagogical use of digital resources and managing digital distractions in the classroom
Students in Sweden make extensive use of digital resources for learning at school and PISA 2022 results indicate that the use of digital resources in Sweden is not necessarily detrimental to students’ performance. Compared with the OECD average, the time that students spend learning with digital devices is associated with larger increases in mathematics performance and appears to become detrimental at significantly higher levels of use than in other countries. Most students and principals in Sweden also reported being satisfied with their teachers’ preparation for digital learning. Nevertheless, there is some evidence to suggest room for improvement concerning the effective use of digital resources. The intensive use of digital technology in Sweden comes with a greater prevalence of digital distractions, which are associated with significant reductions in students’ performance. Furthermore, the frequent use of digital devices in mathematics lessons, for example, is associated with lower mathematics performance (in contrast to Denmark). 39% of Swedish students attend schools whose mathematics teachers are not provided with professional development on the integration of digital resources in mathematics instruction. To help teachers to succeed in the digital age, to maintain a good classroom climate and to keep up with rapid technological advances, Swedish schools will need to continue strengthening their systems of professional learning. Several OECD countries have sought to promote effective forms of professional learning on the use of digital resources, two examples of which (in Finland and Luxembourg) are presented in Box 3.4.
Policy consideration 2: Developing strategies to strengthen students’ digital literacy and promoting a healthier engagement with digital devices
Despite Swedish students’ intensive use of digital resources for learning, they rated their digital competency below the OECD average and below students in Finland and Denmark. This is consistent with results from ICILS 2023, which show that students’ use of digital devices does not go hand in hand with a strong level of digital literacy. Advantaged students in Sweden also reported significantly higher levels of digital confidence than disadvantaged students, mirroring trends across OECD countries and Nordic countries. Consistent with the pattern observed across OECD countries, greater self-efficacy in digital skills is associated with better mathematics performance in Sweden, particularly for disadvantaged students. This suggests that strengthening students’ digital literacy should remain a priority for Sweden and could be a means to raise performance, particularly among disadvantaged students. An example of central efforts to strengthen students’ digital literacy through local initiatives can be seen in Australia (see Box 3.2).
As is the case across OECD countries, a small but significant share of Swedish students also reported engaging in problematic habits when using digital devices, such as not turning off notifications in class or at night. Likewise, many students reported feeling pressure to be online and feeling nervous or anxious at least occasionally when they do not have their digital device near them. Both of these behaviours and feelings are associated with significantly diminished mathematics performance, particularly for disadvantaged students, which risks exacerbating inequities in learning outcomes. Like Sweden, many OECD countries have sought to address this challenge by issuing guidance directed at schools and families to promote a healthier engagement with digital devices (Folkhälsomyndigheten, 2025[38]). International examples of such guidance in Australia, Germany, Canada and the United States are presented in Box 3.3. Sweden should consider additional strategies to support students and their parents in developing self-regulation skills and healthy habits when engaging with digital devices while also addressing risks to students’ well-being and equipping them with the skills they needed to navigate the digital environment safely.
Policy consideration 3: Investing in the digital capacity of school leadership and considering the dissemination of evidence and guidance on the effective use of digital resources
To ensure an effective use of digital education technologies, it is essential to build digital capacity at all levels of the education system. Surprisingly, given the extensive use of digital resources in Sweden, relatively few principals reported offering their mathematics teachers professional development on the integration of digital resources and even fewer reported having a policy on the use of digital devices in mathematics instruction. Where these practices are implemented, they are not associated with improvements in students’ performance. Although tentative, this suggests that there is scope to strengthen the capacity of leadership teams to guide the digital transformation of Swedish schools. International approaches offer promising examples of strategies to strengthen school leaders’ capacity to plan, implement and resource a digital vision for their schools, for example in Wales (United Kingdom) (see Box 3.5). Public authorities should also support education institutions in selecting the right digital tools to meet their needs, facilitate their interactions with developers of digital education technologies, disseminate evidence and guidance on effective practices and empower leadership teams to build a culture of continuous pedagogical improvement, peer learning and innovation in their schools. Particularly in decentralised systems like Sweden, strengthening capacity among local authorities will also be critical to enable them to successfully support and hold their schools to account concerning their use of digital resources.
Policy consideration 4: Monitoring and addressing potential inequities in students’ access to high-quality digital resources in schools
Both students and principals in Sweden reported high levels of satisfaction with their schools’ digital infrastructure. Few principals felt that their schools’ capacity to provide instruction is hindered by a lack of or inadequate digital resources and principals’ perceptions do not appear to vary systematically between advantaged and disadvantaged schools. However, students’ responses in PISA provide some evidence of inequities in the access to high-quality digital resources, which should be carefully monitored. Advantaged students in Sweden reported better access to high-quality ICT resources than disadvantaged students and this socio-economic gap is larger than in most OECD and Nordic countries. Students with better access to high-quality digital resources also score significantly higher in mathematics. If confirmed, inequities in the quality of digital resources could therefore contribute to widening socio-economic gaps in students’ performance. Building on its strong tradition of education monitoring and research, Sweden’s education authorities could direct further efforts to monitor inequities in students’ access to high-quality resources and, if confirmed, consider means to ensure a more equitable distribution of digital resources. Two examples of international approaches to providing additional funding for digital resources to schools in rural areas and those serving disadvantaged student populations are described in Box 3.1.
References
[41] American Psychological Association (2024), APA Recommendations for Healthy Teen Video Viewing, https://www.apa.org/topics/social-media-internet/healthy-teen-video-viewing (accessed on 27 August 2025).
[28] Amez, S. and S. Baert (2020), “Smartphone use and academic performance: A literature review”, International Journal of Educational Research, Vol. 103, p. 101618, https://doi.org/10.1016/J.IJER.2020.101618.
[34] Australian Department of Education (2017), Digital Literacy School Grants Round 1 - Successful Projects - Department of Education, Australian Government, https://www.education.gov.au/australian-curriculum/resources/digital-literacy-school-grants-round-1-successful-projects (accessed on 27 August 2025).
[29] Beland, L. and R. Murphy (2016), “Ill communication: Technology, distraction & student performance”, Labour Economics, Vol. 41, pp. 61-76, https://doi.org/10.1016/J.LABECO.2016.04.004.
[43] Blazar, D. and M. Kraft (2015), “Exploring mechanisms of effective teacher coaching: A tale of two cohorts from a randomized experiment”, Educational Evaluation and Policy Analysis, Vol. 37/4, pp. 542-566, https://doi.org/10.3102/0162373715579487.
[26] Boeskens, L. and A. Echazarra (2025), “Using digital resources for learning: Policy insights from PISA 2022”, OECD Education Working Papers, No. 340, OECD Publishing, Paris, https://doi.org/10.1787/eb9453f3-en.
[3] Bulman, G. and R. Fairlie (2016), “Technology and Education: Computers, Software, and the Internet”, Handbook of the Economics of Education, Vol. 5, pp. 239-280, https://doi.org/10.1016/B978-0-444-63459-7.00005-1.
[32] Burns, T. and F. Gottschalk (eds.) (2020), Education in the Digital Age: Healthy and Happy Children, Educational Research and Innovation, OECD Publishing, Paris, https://doi.org/10.1787/1209166a-en.
[39] Canadian Pediatric Society (2022), Screen Time and Preschool Children: Promoting Health and Development in a Digital World, https://cps.ca/en/documents/position/screen-time-and-preschool-children (accessed on 27 August 2025).
[20] Department of Education (2024), Grant Scheme for ICT Infrastructure (Circular 0039/2024), https://www.gov.ie/pdf/?file=https://assets.gov.ie/289703/4faef53c-e6fa-4d9b-b5f0-629298d20507.pdf (accessed on 19 October 2024).
[22] Echazarra, A. and T. Radinger (2019), “Learning in rural schools: Insights from PISA, TALIS and the literature”, OECD Education Working Papers, No. 196, OECD Publishing, Paris, https://doi.org/10.1787/8b1a5cb9-en.
[38] Folkhälsomyndigheten (2025), Hjälp ditt barn på nätet – stöd för föräldrar om barns och ungas skärmanvändning [Help your child online – support for parents about children’s and young people’s screen use], https://www.folkhalsomyndigheten.se/livsvillkor-levnadsvanor/digitala-medier-och-halsa/hjalp-ditt-barn-pa-natet/ (accessed on 23 November 2025).
[6] Forsström, S. et al. (2025), “The impact of digital technologies on students’ learning: Results from a literature review”, OECD Education Working Papers, No. 335, OECD Publishing, Paris, https://doi.org/10.1787/9997e7b3-en.
[7] Forsström, S. et al. (2025), “Key findings and integration strategies on the impact of digital technologies on students’ learning: Results from a literature review”, OECD Education Working Papers, No. 336, OECD Publishing, Paris, https://doi.org/10.1787/ab309c32-en.
[30] Fraillon, J. (2024), An International Perspective on Digital Literacy: Results from ICILS 2023, International Association for the Evaluation of Educational Achievement (IEA), Amsterdam, https://www.iea.nl/sites/default/files/2024-11/ICILS_2023_International_Report_0.pdf (accessed on 21 November 2024).
[2] Ganimian, A., E. Vegas and F. Hess (2020), Realizing the Promise: How Can Education Technology Improve Learning for All?, The Brookings Institution, Washington, https://www.brookings.edu/wp-content/uploads/2020/08/edtech_playbook_full_v2.pdf (accessed on 6 April 2022).
[44] Kraft, M. and D. Blazar (2017), “Individualized coaching to improve teacher practice across grades and subjects: New experimental evidence”, Educational Policy, Vol. 31/7, pp. 1033-1068, https://doi.org/10.1177/0895904816631099.
[48] Luxembourg Ministry of Education (2020), Les compétences numériques (E-skills) [Digital Skills (E-skills)], https://men.public.lu/en/grands-dossiers/enseignement-secondaire/05-competences-numeriques.html (accessed on 27 August 2025).
[21] Ministère de l’Éducation Nationale et de la Jeunesse (2018), L’éducation nationale engagée en faveur d’une école rurale de qualité [National education engaged for high-quality rural schooling], https://www.education.gouv.fr/rentree-2018-l-education-nationale-engagee-en-faveur-d-une-ecole-rurale-de-qualite-8375 (accessed on 26 July 2022).
[40] National Academies of Sciences (2023), Social Media and Adolescent Health: Consensus Study Report, The National Academies Press, Washington, DC, https://doi.org/10.17226/27396.
[37] OECD (2025), How’s Life for Children in the Digital Age?, OECD Publishing, Paris, https://doi.org/10.1787/0854b900-en.
[8] OECD (2025), “Preparing teachers for digital education: Continuing professional learning on digital skills and pedagogies”, OECD Education Policy Perspectives, No. 122, OECD Publishing, Paris, https://doi.org/10.1787/af442d7a-en.
[18] OECD (2024), PISA 2022 Results (Volume V): Learning Strategies and Attitudes for Life, PISA, OECD Publishing, Paris, https://doi.org/10.1787/c2e44201-en.
[16] OECD (2024), PISA 2022 Technical Report, PISA, OECD Publishing, Paris, https://doi.org/10.1787/01820d6d-en.
[15] OECD (2023), PISA 2022 Database, https://www.oecd.org/en/data/datasets/pisa-2022-database.html (accessed on 29 January 2024).
[9] OECD (2023), “PISA 2022 ICT Framework”, in PISA 2022 Assessment and Analytical Framework, OECD Publishing, Paris, https://doi.org/10.1787/9bd299c1-en.
[17] OECD (2023), PISA 2022 Results (Volume I): The State of Learning and Equity in Education, PISA, OECD Publishing, Paris, https://doi.org/10.1787/53f23881-en.
[14] OECD (2023), PISA 2022 Results (Volume II): Learning During – and From – Disruption, PISA, OECD Publishing, Paris, https://doi.org/10.1787/a97db61c-en.
[1] OECD (2023), Shaping Digital Education: Enabling Factors for Quality, Equity and Efficiency, OECD Publishing, Paris, https://doi.org/10.1787/bac4dc9f-en.
[24] OECD (2022), Value for Money in School Education: Smart Investments, Quality Outcomes, Equal Opportunities, OECD Publishing, Paris, https://doi.org/10.1787/f6de8710-en.
[4] OECD (2021), 21st-Century Readers: Developing Literacy Skills in a Digital World, PISA, OECD Publishing, Paris, https://doi.org/10.1787/a83d84cb-en.
[10] OECD (2021), Computer-Based Student Questionnaire for PISA 2022: Main Survey Version, https://www.oecd.org/content/dam/oecd/en/data/datasets/pisa/pisa-2022-datasets/questionnaires/COMPUTER-BASED%20STUDENT%20questionnaire%20PISA%202022.pdf (accessed on 21 February 2025).
[11] OECD (2021), ICT Questionnaire for PISA 2022 (International Option): Main Survey Version, https://www.oecd.org/content/dam/oecd/en/data/datasets/pisa/pisa-2022-datasets/questionnaires/ICT%20QUESTIONNAIRE%20PISA%202022.pdf (accessed on 21 February 2025).
[12] OECD (2021), School Questionnaire for PISA 2022: Main Survey Version, https://www.oecd.org/content/dam/oecd/en/data/datasets/pisa/pisa-2022-datasets/questionnaires/SCHOOL%20QUESTIONNAIRE%20computer%20PISA%202022.pdf (accessed on 21 February 2025).
[19] OECD (2020), PISA 2018 Results (Volume V): Effective Policies, Successful Schools, PISA, OECD Publishing, Paris, https://doi.org/10.1787/ca768d40-en.
[5] OECD (2019), OECD Skills Outlook 2019: Thriving in a Digital World, OECD Publishing, Paris, https://doi.org/10.1787/df80bc12-en.
[23] OECD (2017), PISA 2015 Results (Volume III): Students’ Well-Being, PISA, OECD Publishing, Paris, https://doi.org/10.1787/9789264273856-en.
[42] Opfer, D. (2016), “Conditions and Practices Associated with Teacher Professional Development and Its Impact on Instruction in TALIS 2013”, OECD Education Working Papers, No. 138, OECD Publishing, Paris, https://doi.org/10.1787/5jlss4r0lrg5-en.
[46] OPH (2018), Tutor Teacher Activities in Basic Education in Finland (Facts Express, No. 3C/2018), Finnish National Agency for Education, Helsinki, https://www.oph.fi/sites/default/files/documents/195451_oph_faktaa_express_3c_2018_englanti_sivut.pdf (accessed on 27 August 2025).
[33] Paniagua, A. and D. Istance (2018), Teachers as Designers of Learning Environments: The Importance of Innovative Pedagogies, Educational Research and Innovation, OECD Publishing, Paris, https://doi.org/10.1787/9789264085374-en.
[45] Papay, J. et al. (2016), “Learning job skills from colleagues at work: Evidence from a field experiment using teacher performance data”, NBER Working Papers, No. 21986, https://doi.org/10.3386/w21986.
[27] Pedaste, M. et al. (2015), “Phases of inquiry-based learning: Definitions and the inquiry cycle”, Educational Research Review, Vol. 14, pp. 47-61, https://doi.org/10.1016/J.EDUREV.2015.02.003.
[47] Pennanen, M. et al. (2021), Tutkimus perusopetuksen tutoropettajatoiminnasta ja sen vaikutuksista [Research on tutor teacher activities in basic education and their effects], Finnish National Board of Education, Helsinki, https://www.oph.fi/fi/tilastot-ja-julkaisut/julkaisut/tutkimus-perusopetuksen-tutoropettajatoiminnasta-ja-sen (accessed on 27 August 2025).
[36] Skolverket (2025), Attityder till skolan 2024 [Attitudes Towards School 2024], https://www.skolverket.se/getFile?file=13208 (accessed on 11 August 2025).
[13] Skolverket (2024), 15-åringars användning av digitala verktyg: Nordiska jämförelser utifrån PISA 2022 [15-Year-Olds’ Use of Digital Tools: Nordic Comparisons Based on PISA 2022], Swedish National Agency for Education, Stockholm, http://www.skolverket.se/publikationer (accessed on 26 May 2025).
[31] Skolverket (2024), ICILS 2023: En internationell studie om digital kompetens hos elever i årskurs 8 [ICILS 2023: An International Study of the Digital Competency of Students in Grade 8], https://www.skolverket.se/getFile?file=13168 (accessed on 11 August 2025).
[35] Skolverket (2022), Skolverkets uppföljning av digitaliseringsstrategin 2021 [The Swedish National Agency for Education’s Follow-up of the Digitalisation Strategy 2021], https://www.skolverket.se/getFile?file=9385 (accessed on 11 August 2025).
[25] Skolverket (2022), Tabellverk - Frågor till lärare - Skolverkets uppföljning av digitaliseringsstrategin 2021 [Tables - Questions for teachers - The Swedish National Agency for Education’s Follow-up of the Digitalisation Strategy 2021], https://www.skolverket.se/download/18.47ddd14d17f0282d1f45cf/1645805985409/Tabeller_L%C3%A4rare_2021.pdf (accessed on 23 November 2025).
[49] Welsh Government (2022), Digital Professional Learning Journey (DPLJ), https://hwb.gov.wales/professional-learning/career-long-professional-learning/digital-professional-learning-journey/ (accessed on 21 October 2024).
← 1. Caution is required when interpreting estimates for Denmark and other countries marked with an asterisk in figures and tables because one or more PISA sampling standards were not met. See Reader’s Guide, Annexes A2 and A4 in OECD (2023[17]).
← 2. Values on some indices are reported as missing for Norway and two indices were not included in the analysis since they were missing for Sweden (“PROBSCRI” on problems with schools’ capacity to provide remote instruction and “DIGDVPOL” on digital device policies at school). Results for these indices were suppressed since they did not meet PISA’s technical standards for internal consistency. Indices needed to have a Cronbach’s alpha coefficient of at least 0.60 to be reported for any given country.
← 3. The index of ICT availability at school (ICTSCH) was constructed based on students’ frequency ratings of how often they use the following seven digital resources at school (based on question IC170 of the ICT Familiarity Questionnaire): Desktop or laptop computer; Smartphone (i.e. mobile phone with Internet access); Tablet device (e.g. [iPad®], [Galaxy Book®],) or e-book reader (e.g. [Amazon Kindle™], [Kobo®]); Internet access (except on smartphones); School portal (to consult timetable, absences, etc.); Educational software, games or apps, other learning tools (e.g. [CK-12™] or [Mathalicious®] online support); A learning management system or school learning platform (e.g. [Blackboard®], [Edmodo®], [Moodle®], [Google® Classroom™]). Each of the seven items had six response options (“Never or almost never”, “About once or twice a month”, “About once or twice a week”, “Every day or almost every day”, “Several times a day”, “This resource is not available to me at school”). “This resource is not available to me at school” was recoded as 0, while the five other response options were recoded as 1 prior to scaling. For more details, see Table 19.A.86 in OECD (2024[16]), https://stat.link/v6uq1n.
← 4. The index of ICT availability outside school (ICTHOME) was constructed based on students’ frequency ratings of how often they use the following six digital resources at outside of school (based on question IC171 of the ICT Familiarity Questionnaire): Desktop or laptop computer; Smartphone (i.e. mobile phone with Internet access); Tablet device (e.g. iPad®, Galaxy Book®,) or e-book reader (e.g. Amazon® Kindle™, Kobo®); Internet access (except on smartphones); Educational software, games or Apps, other learning tools (e.g. CK-12™ or Mathalicious® online support); Video or online games (e.g. used with game consoles such as a Play Station 4® or Nintendo Wii®, online gaming platforms such as Steam® or gaming Apps such as Angry Birds®). Each of the six items included in this scale had six response options (“Never or almost never”, “About once or twice a month”, “About once or twice a week”, “Every day or almost every day”, “Several times a day”, “This resource is not available to me outside of school”). “This resource is not available to me outside of school” was recoded as 0, while the five other response options were recoded as 1 prior to scaling. For more details, see Table 19.A.87 in OECD (2024[16]), https://stat.link/v6uq1n.
← 5. The index of the quality of access to ICT (ICTQUAL) was constructed based on students’ reported agreement with the following nine statements about ICT resources at their school (based on question IC172 of the ICT Familiarity Questionnaire): ”There are enough digital resources for every student at my school”; “There are enough digital devices with access to the Internet at my school”; “The school's Internet speed is sufficient”; “Digital resources function properly at my school”; “Digital resources are easily accessible within the classroom”; “Digital learning resources available at my school make learning interesting”; “The school provides sufficient technical support to help students in their use of digital resources”; “Teachers at my school have the necessary skills to use digital devices during instruction”; “Teachers at my school are willing to use digital resources for teaching.” Each of the nine items included in this scale had four response options (“Strongly disagree”, “Disagree”, “Agree”, “Strongly agree”). For more details, see Table 19.A.88 in OECD (2024[16]), https://stat.link/v6uq1n.
← 6. Since there is no separate subject for computer science, information technology or informatics for 15-year-old students in Sweden, results for this subject were not covered in this report, in line with previous analyses by the Swedish National Agency for Education (Skolverket, 2024[13]).
← 7. The index of subject-related ICT use during lessons (ICTSUBJ) was constructed based on students’ frequency ratings of how often digital resources are used in lessons of the following four subjects (based on question IC173 of the ICT Familiarity Questionnaire): Lessons in the test language; Lessons in mathematics; Lessons in science; Lessons in computer science, information technology, informatics or similar lessons. Each of the four items included in this scale had five substantive response options (“Never or almost never, “In less than half of the lessons”, “In about half of the lessons”, “In more than half of the lessons”, “In every or almost every lesson”) and an additional response option “I do not have this subject”, which was recoded as missing prior to scaling. For more details, see Table 19.A.89 in OECD (2024[16]), https://stat.link/v6uq1n.
← 8. The index of the use of ICT in enquiry-based learning activities (ICTENQ) was constructed based on students’ frequency ratings of how often they use digital resources for the following 10 school-related activities (based on question IC174 of the ICT Familiarity Questionnaire): Create a multimedia presentation with pictures, sound or video; Write or edit text for a school assignment (e.g. using [Google® Docs™], [Microsoft® Word]); Find information online about real-world problems or phenomena (e.g. climate change, oil spills, measuring the height of a building); Collect and record data (e.g. using data loggers, [Microsoft® Access™], [Google® Forms], spreadsheets); Analyse data that you have collected yourself (e.g. using [Microsoft® Excel™]); Report or share your results from your own experiments or investigations; Plan and manage work or projects (e.g. dividing tasks, managing deadlines); Track the progress of your own work projects; Collaborate with other students to create digital content (e.g. presentation, etc.); Play digital learning games (e.g. [Atlantis Remixed], [Duolingo®]). Each of the 10 items included in this scale had five response options (“Never or almost never”, “About once or twice a year”, “About once or twice a month”, “About once or twice a week”, “Every day or almost every day”). For more details, see Table 19.A.90 in OECD (2024[16]), https://stat.link/v6uq1n.
← 9. The index of support or feedback via ICT (ICTFEED) was constructed based on students’ frequency ratings of how often they use digital resources in the following four activities related to support or feedback (based on question IC175 of the ICT Familiarity Questionnaire): Read or listen to feedback sent by my teachers regarding my work and academic results; Read or listen to feedback sent by other students on my work; Read or listen to feedback automatically selected for me by an educational software or app; Work on drill and practice exercises using an educational software or app. Each of the four items included in this scale had five response options (“Never or almost never”, “About once or twice a year”, “About once or twice a month”, “About once or twice a week”, “Every day or almost every day”). For more details, see Table 19.A.91 in OECD (2024[16]), https://stat.link/v6uq1n.
← 10. The index of the use of ICT for school activities outside of the classroom (ICTOUT) was constructed based on students’ reports of how often they use digital resources for the following eight school-related activities outside of the classroom (based on question IC176 of the ICT Familiarity Questionnaire): See my grades or results from specific assignments (e.g. homework or tests); Browse the Internet for schoolwork (e.g. for preparing an essay or presentation); Browse the Internet to follow up lessons (e.g. for finding explanations); Receive or download assignments or instructions (via [digital resources]) from my teacher; Upload my work (e.g. documents or presentations) to a school portal or another platform to share materials with my teacher and/or students; Communicate with my teacher; Communicate with other students about schoolwork; Search for information on school-related activities or assignments. Each of the eight items included in this scale had five response options (“Never or almost never”, “About once or twice a year”, “About once or twice a month”, “About once or twice a week”, “Every day or almost every day”). For more details, see Table 19.A.92 in OECD (2024[16]), https://stat.link/v6uq1n.
← 11. The index of self-efficacy in digital competencies (ICTEFFIC) was constructed based on students’ ratings of how well they can do the following 14 tasks using digital resources (based on question IC183 of the ICT Familiarity Questionnaire): Search for and find relevant information online; Assess the quality of information you found online; Share practical information with a group of students; Collaborate with other students on a group assignment; Explain to other students how to share digital content online or on a school platform; Write or edit text for a school assignment; Collect and record data (e.g. using data loggers, [Microsoft® Access™], [Google® Forms], spreadsheets); Create a multimedia presentation (with sound, pictures, or video); Create, update and maintain a webpage or a blog; Change the settings of a device or app in order to protect my data and privacy; Select the most efficient programme or app that allows me to carry out a specific task; Create a computer programme (e.g. in [Scratch®], [Python®], [Java®]); Identify the source of an error in a software after considering a list of potential causes; Break down a problem and represent a solution as a series of logical steps, such as an algorithm. Each of the 14 items included in this scale had four substantive response options (“I cannot do this”, “I struggle to do this on my own”, “I can do with a bit of effort”, “I can easily do this”) and an additional response option “I don’t know what this is” which was recoded as missing prior to scaling. For more details, see Table 19.A.97 in OECD (2024[16]), https://stat.link/v6uq1n.
← 12. The index of students’ practices regarding online information (ICTINFO) was constructed based on students’ reported agreement with the following six statements about their practices regarding online information (based on question IC180 of the ICT Familiarity Questionnaire): “When searching for information online I compare different sources”; “I check the accuracy of online information before sharing it on social networks”; “I discuss the accuracy of online information with my teachers or in class”; “I discuss the accuracy of online information with friends or other students”; “I discuss the accuracy of online information with my parents (or family)”; “I try to flag wrong information when I encounter it online.” Each of the six items included in this scale had four response options (“Strongly disagree”, “Disagree”, “Agree”, “Strongly agree”). For more details, see Table 19.A.96 in OECD (2024[16]), https://stat.link/v6uq1n.
← 13. This association is not accounted for by gender differences in mathematics performance and students’ engagement in digital leisure activities.
← 14. Authors' analysis based on OECD (2023[15]), PISA 2022 Database, https://www.oecd.org/en/data/datasets/pisa-2022-database.html; Question IC178 of the ICT Familiarity Questionnaire.
← 15. The index of frequency of ICT activity – Weekday (ICTWKDY) was constructed based on students’ reports of how much time they spent engaged in the following seven leisure activities using ICT during a typical weekday (based on question IC177 of the ICT Familiarity Questionnaire): Play video-games (using my smartphone, a gaming console or an online platform or apps); Browse social networks (e.g. [Instagram®], [Facebook®]); Browse the Internet (excluding social networks) for fun (e.g. reading NEWs, listening to podcasts and music or watching videos); Look for practical information online (e.g. find a place, book a train ticket, buy a product); Communicate and share digital content on social networks or any communication platform (e.g. [Facebook®], [Instagram®], [Twitter®], emails, chat); Read, listen to or view informational materials to learn how to do something (e.g. tutorial, podcast); Create or edit my own digital content (pictures, videos, music, videos, computer programmes). Each of the seven items included in this scale had six response options (“No time at all, “Less than 1 hour a day”, “Between 1 and 3 hours a day”, “More than 3 hours and up to 5 hours a day”, “More than 5 hours and up to 7 hours a day”, “More than 7 hours a day”). For more details, see Table 19.A.93 in OECD (2024[16]), https://stat.link/v6uq1n.
← 16. The index of preparedness for digital learning (DIGPREP) was constructed based on school administrators’ scaled ratings of how much they agreed with the following six statements about their school’s capacity to use digital devices to enhance learning and teaching (based on question SC155 of the School Questionnaire): “Teachers have the necessary technical and pedagogical skills to integrate digital devices in instruction”; “Teachers have sufficient time to prepare lessons integrating digital devices”; “Effective professional resources for teachers to learn how to use digital devices are available; An effective online learning support platform is available”; “Teachers are provided with incentives to integrate digital devices in their teaching”; “The school has sufficient qualified technical assistant staff.” The four response categories for the six items in the scale were “Strongly disagree”, “Disagree”, “Agree”, and “Strongly agree”. For more details, see Table 19.A157 in OECD (2024[16]), https://stat.link/v6uq1n.
← 17. The index is not available for Sweden, Denmark, Finland or Norway in PISA 2022.