Using digital resources for learning: Policy insights from PISA 2022

Abstract

Advances in digital technology have the potential to profoundly transform teaching and learning. Prepared as part of the OECD project "Resourcing School Education: Policies for the Digital Transformation of Education and Future-Readiness of Teachers", this working paper draws on data from PISA 2022 to analyse how students and teachers are using digital resources for learning activities today. Following a description of schools' digital infrastructure, the paper analyses how students use digital resources for learning and which factors are associated with a more intensive use of digital technology in schools. Drawing on previously unexamined data from the PISA Teacher Questionnaire, it then analyses the digital tools that teachers use and how they employ them. Using cluster analysis, the paper identifies distinct patterns in students', teachers' and schools' approaches to digital learning, revealing significant heterogeneity across and within countries. The paper concludes with insights that can guide the design and implementation of more effective and equitable digital education policies.

Full text

Translate

Turn on search term navigation

Headnote

Abstract

Acknowledgements

This working paper was prepared by Luka Boeskens and Alfonso Echazarra as part of the OECD project "Resourcing School Education: Policies for the Digital Transformation of Education and Future-Readiness of Teachers", which is included in the 2025-26 Programme of Work of the OECD's Education Policy Committee (EDPC). The authors would like to thank delegates of the OECD Group of National Experts on School Resources (GNE-SR), which oversees the project, as well as members of the EDPC and the PISA Governing Board (PGB) for their insightful feedback on drafts of the paper. The working paper also benefited from the input and advice of José-Luis Alvarez-Galvan and Miyako Ikeda. The authors are grateful to Paulo Santiago (Head of the Policy Advice and Implementation Division of the OECD Directorate for Education and Skills) and Andreas Schleicher (OECD Director for Education and Skills) for their overall guidance and feedback. The authors also thank Christina Mitrakos and Sophie Limoges for their support during the formatting and publication process.

1 Introduction

Background

Digital education technologies have become a key resource in OECD education and training systems and significant investments over the past decade have enabled their increasing integration into schools and classrooms. When used effectively, digital tools can enhance teaching, foster student engagement and collaboration, and support self-regulated learning (OECD, 202371). However, their positive impact depends on a careful design of tasks, their alignment with learning objectives and the appropriate selection of tools (Forsstrôm et al., 2025p). Teachers increasingly face the challenge of integrating digital resources into their practice to meet individual students' needs while maintaining engagement and fostering meaningful learning. Many education systems have responded by investing in guidance and professional learning to strengthen teachers' capacity for digital education (OECD, 2025137).

Data from the 2022 Programme for International Student Assessment (PISA) show substantial international variation in the intensity of digital resource use in schools (OECD, 20231). On average across OECD countries, 15-year-old students reported spending two hours per day using digital resources for learning activities at school, ranging from under 1.5 hours in Czechia, France, Germany, Greece, Ireland and Slovenia to 3 hours or more in Denmark, Iceland, Norway and Sweden (OECD, 2023, pp. 410, Table 11.B1.5.62). Whether and how these digital education resources are used also varies significantly across schools, with important consequences for students' learning experience and outcomes. Yet, comparatively little is known about the types of digital tools that are used, how they are used by teachers and students, and what explains the variation observed across countries, schools and teachers.

In most education systems, schools and teachers have some autonomy in selecting digital resources and integrating them in lessons. This is underlined by the significant variation in the digital resource use observed both between and within schools (see Figure 20). At the same time, public authorities can play an important role in shaping the digital learning environment, by addressing infrastructure gaps, removing technical barriers, and strengthening the professional capacity of teachers and school leaders. They can also provide guidance on the effective pedagogical use of digital resources or adapt assessment and evaluation frameworks to reflect their use, among a diverse range of other policy levers. The successful design and implementation of such interventions requires policy makers to have a clear understanding of the way students and teachers are using digital technology today.

Objectives and structure

This working paper uses PISA 2022 data to analyse the use of digital resources for learning in schools and to derive policy-relevant insights concerning variations in digital teaching and learning practices across individuals, schools and countries. It was prepared as part of the OECD project "Resourcing School Education: Effective Digitalisation and Future-Ready Teachers" (OECD, 202515). Launched in 2023, the project supports the development of system-level policies to enable the effective use of digital technologies and to prepare teachers for the digital age. The project provides targeted policy advice through in-depth country reviews, offers opportunities for international peer learning and strengthens the international evidence base through comparative data collections and research.'

This working paper aims to strengthen the projects evidence base and to inform policy makers and practitioners seeking to enhance equity and students' learning outcomes through digital resources. It seeks to enrich our understanding of how students and teachers use digital resources for learning, and to identify systematic variation within countries as well as the inequalities that may arise from it. The paper also identifies individual-, school-, and system-level factors related to the use of digital resources for learning, which provide insights into the policies and practices that might enable or hinder the adoption of digital technologies in schools.

Following this introduction, Section 2 provides an overview of the state of schools' digital infrastructure and students' access to high-quality digital resources, which are important preconditions for digital learning. Section 3 analyses when, how often and for what purposes students use digital resources for learning, exploring variations based on student and school characteristics. Cluster analysis is used to identify distinct profiles of digital learners across countries. Section 4 shifts to the perspective of teachers, examining the tools and practices they employ in digital instruction across a smaller number of countries that administered the PISA 2022 Teacher Questionnaire. It also investigates the factors shaping teachers' engagement with digital tools, including their demographic characteristics, training and school environment. Section 5 concludes with a discussion of seven policy insights emerging from the analysis. An annex includes supplementary tables with additional country-specific results. All figures and underlying data are available on the working paper's support materials page: https://doi.org/10.1787/eb9453f3-en.

Data and coverage of countries and economies

The analysis presented in this working paper is based on data from PISA 2022, which provided the first comprehensive and representative international data collection covering the use of digital technologies for teaching and learning since the COVID-19 pandemic. Created by the Organisation for Economic Cooperation and Development (OECD), the Programme for International Student Assessment (PISA) tests the skills and knowledge of 15-year-old students in mathematics, reading and science. Some 690 000 students, representing about 29 million 15-year-olds in 81 countries and economies, participated in the 2022 assessment, which focussed on mathematics.

The analysis draws on data from four PISA background questionnaires: The Student Questionnaire (STQ) (OECD, 2021), administered to students in all participating countries and economies; the School Questionnaire (SCQ) (OECD, 20217), administered to school principals in all participating countries and economies; the ICT Familiarity Questionnaire (ICQ) (OECD, 2021781), administered to students in 52 of 81 countries and economies; and the Teacher Questionnaire (TQ) (OECD, 2021197), administered to teachers in 18 of 81 countries and economies (OECD, 2024, pp. 21, Annex Table 1.A.1j10). The Teacher Questionnaire was administered to up to 25 respondents per school, with the aim to reach an equal number of teachers of the 2022 cycle's major domain (mathematics) and teachers of other subjects. Since the Teacher Questionnaire aims to reflect the learning environment experienced by 15-year-old students sitting the PISA test, it was distributed to teachers who were currently teaching the modal grade attended by 15year-olds.

Given the target population of PISA, all analyses and findings presented in this paper apply to 15-year-old students and their teachers and may not extend to older and younger students. It is also important to bear in mind the self-reported nature of the background questionnaires, which may introduce systematic measurement error related, for example, to social-desirability bias in the hours that students report using digital resources for leisure or school leaders' assessments of their teachers' digital competency. Finally, all findings presented in this working paper reflect the situation at the time of the PISA testing period (for most countries, between March and October 2022) and - given the rapid developments in the use of digital technology - may have evolved in the meantime.

The analysis covers all OECD Member countries that participated in PISA 2022, the eight accession candidate countries at the time of writing (Argentina, Brazil, Bulgaria, Croatia, Indonesia, Peru, Romania and Thailand) as well as Singapore, which participated in the Policy Survey on School Education in the Digital Age.? These are collectively referred to as the "covered countries and economies". Analyses based on items of the PISA 2022 Teacher Questionnaire cover all 18 countries and economies that administered the optional module. Several jurisdictions participating in PISA 2022 did not meet one or more of the test's sampling standards and it is not possible to exclude the possibility of more than minimal bias based on the information available at the time of data adjudication. They are marked with an asterisk (·) in figures and tables and caution is advised when interpreting their data. For more information, see Annexes A2 and A4 in the Reader's Guide of PISA 2022 Results (Volume 1): The State of Learning and Equity in Education (OECD, 20237111).

Previous comparative work on the use of digital resources for learning

Teachers' and students' use of digital resources for learning has attracted significant policy interest and is the subject of a rapidly expanding body of research. Much of this work focusses on describing how digital resources are used in the classroom or identifying their causal impact on students' learning and well-being - a topic explored in a separate literature review of the Resourcing School Education project (Forsstrôm et al., 2025121; Forsstróm et al., 2025/12). This working paper uses PISA 2022 data to provide a comparative international perspective, identifying patterns across education systems and highlighting differences that can inform policy or provide opportunities for international peer learning. The working paper focusses on the use of digital resources for learning and complements previous analyses that focussed on their use for leisure activities.

Previous international surveys and assessments have provided insights into the use of digital technology for learning, including the 2018 Teaching and Learning International Survey (TALIS), the 2021 Progress in International Reading Literacy Study (PIRLS), the 2023 Trends in International Mathematics and Science Study (TIMSS), and the 2023 International Computer and Information Literacy Study (ICILS), Which focusses on students' digital literacy. PISA 2022 data make a distinct contribution to this body of international comparative work, by virtue of its broad coverage of OECD Member countries and accession candidate countries. PISA 2022 also allows for students', teachers' and principals' reported use of digital technology to be linked with students' performance in mathematics, reading and science.

Previous OECD analyses of PISA 2022 data on digital education have focussed on: 1. The relationship between students' use of digital devices (for leisure and learning), digital distractions, students' performance (incl. in creative thinking) and students' well-being (OECD, 2024113); OECD, 2023, pp. 194, Figure I1.5.14(4)); and 2. The quality and availability of digital resources in schools (incl. inequalities within countries), their relationship with performance across countries and schools' preparedness for digital learning (OECD, 202371). PISA 2022 Volume V included some information on students' use of digital resources at school and how it relates to their self-efficacy in digital competencies (OECD, 2024, p. 203 ff.r41). However, less is known about the individual-, school-, and system-level factors that relate to the use of digital devices for learning (by students and teachers) and how patterns of use vary across schools. This working paper addresses this gap, including by analysing items from the PISA 2022 Student and Teacher Questionnaires that have not been subject to comparative analyses before.

2 Schools' digital infrastructure and access to high-quality digital resources

Students' and teachers' access to high-quality digital resources is a prerequisite for their effective use in the learning process. Drawing on PISA 2022 data, this section provides the context for the subsequent analyses by offering a brief overview of the state of schools' digital infrastructure. The availability of digital infrastructure in schools has increased dramatically over the past decade and is now widespread across OECD countries. Nevertheless, students' and school leaders' reports suggest that limited access digital devices in schools or their insufficient quality remain a barrier and potential source of inequalities in some schools (OECD, 2023[4]).

General access to digital infrastructure in schools has further expanded and is now close to universal across OECD countries

Access to digital resources in schools has expanded significantly over the past decade. In PISA 2022, on average across OECD countries, principals reported that there are about 0.8 computers (laptops or desktop computers) and 0.4 tablet devices or e-book readers available for educational purposes for every 15-year-old student at their school (OECD, 2023, pp. 411, Tables II.B1.5.24 and II.B1.5.27[4]). Between 2012 and 2022, the computer-to-student ratio for 15-year-olds has increased significantly in 21 of 37 OECD countries (by 0.13 on average). While most of this expansion took place in the earlier part of the decade and has since tapered off, a number of OECD countries (including Finland, France, Japan, Poland and Portugal), saw their computer-to-student ratio increase significantly between 2018 and 2022. The OECD average did not change during this time period (OECD, 2023, pp. 411, Table II.B1.5.25[4]).

On average across OECD countries, there is no difference in the number of computers that are available to students in socio-economically advantaged and disadvantaged schools and where gaps exist, they tend to favour disadvantaged schools (OECD, 2023, pp. 411, Table II.B1.5.24[4]). This may reflect targeted funding and other efforts to compensate for students' lack of access to digital devices at home. It could also reflect that disadvantaged students are more likely to attend small rural schools (Echazarra and Radinger, 2019[15]), where the number of devices per student tends to be higher (OECD, 2023, pp. 409, Table II.B1.5.24[4]).

The widespread availability of digital technology reported by school leaders is echoed in students' responses to the PISA 2022 ICT Familiarity Questionnaire. On average, 96.1% of students across OECD countries with available data reported that a desktop or laptop computer is available to them at their school and at least 92% of students reported having access in all of the 35 countries with available data covered in this working paper. Similarly high proportions of students reported having access to tablets or e-book readers, a school portal, learning management system or school learning platform, or to educational software, games, apps or other learning tools (Figure 1).

Students' reports suggest that the quality and accessibility of schools" digital infrastructure remain a barrier in some OECD countries

Despite the increasing presence of digital devices in schools, students' reports in PISA 2022 suggest that their quality and accessibility remain limited in some countries. Across OECD countries, about a third of students (29%) do not agree that there are enough digital resources for every student at their school (ranging from fewer than 15% of students in Denmark, Sweden and the United States to more than 50% in Germany and Turkiye). This is slightly higher than the proportion of students (24%) whose principal reported that the school's capacity to provide instruction is hindered to some extent or a lot by a lack of digital resources (OECD, 2023, pp. 409, Table 1.B1.5.17y)). Likewise, across OECD countries, only 67% of students agree that digital resources are easily accessible within the classroom, underlining the difficulty of effectively integrating digital resources in the teaching process (Figure 2). (For example, digital devices may be kept in a dedicated room, making it difficult for subject teachers to use them on a regular basis or for multiple classes to use them simultaneously).

At least 17% of 15-year-old students in every OECD country (and 29% on average) also reported that the digital resources at their school do not function properly. This is close to the proportion of students (25%) Whose principal reported that the school's capacity to provide instruction is hindered to some extent or a lot by inadequate or poor-quality digital resources (OECD, 2023, pp. 409, Table ||.B1.5.17j). This points to the challenge of ensuring that devices are of high quality and continuously maintained once they have been purchased. In many countries, internet speed remains another limiting factor in students' views. On average across OECD countries, 46% of students consider it insufficient and no more than 70% of students reported being satisfied with their schools' internet speed in any OECD country (Figure 2).

In contrast to the number of computers available in schools, the quality and accessibility of digital resources also constitute a potential source of inequity. Across OECD countries, disadvantaged students were significantly more likely to report that ICT resources at their school were not functioning properly, were not easily accessible within the classroom, or that their internet speed was not sufficient (OECD, 20237167). On average, disadvantaged students scored significantly lower than their advantaged peers (-0.06 vs. 0.07) on PISA's "index of quality of access to ICT", which measures these factors.? This significant gap was observed in 18 of 28 OECD countries with available data and likely accounts for some of the socioeconomic differences in students' digital learning experience, which are explored in the next section (OECD, 20231161).

In most countries, few students frequently use laptops at school, but smartphones are both often used and increasingly regulated

Although the use of digital resources in the learning process has undoubtedly increased in recent years, their ubiquity in OECD countries' classrooms should not be overstated. In PISA 2022, 37% of 15-year-old students reported making daily or almost daily use of computers at school and only 21% reported making use of computers several times a day (Figure 3). Although more than 90% of students reported that computers are in principle available to them at school (see Figure 1), this suggests that most students use them sporadically. Notable exceptions are Australia, Denmark, Sweden and the United States, where more than half of all students reported using computers several times a day at school.

More so than computers, the use of smartphones has become part of many students' daily experience at school. Across OECD countries, 61% of students reported using smartphones every day or almost every day and 37% of them reported using them several times a day (Figure 4). Students did not specify whether they use either computers or smartphones for leisure or learning purpose at school. However, at least some of this smartphone use is likely to be unrelated to learning activities, given that students reported spending 1.1 hours a day using digital resources for leisure at school and about 30% of students reported that students in their mathematics lessons get distracted using digital devices in most or every mathematics lesson (OECD, 2023, pp. 410, Tables 11.B1.5.40 and I1.B1.5.67/116]).

In recent years, concerns about cyberbullying, classroom distractions and a desire to foster students social interactions have led many OECD education systems to limit mobile phone use in schools. On average in 2022, 34% of 15-year-old students attend schools whose principals reported that a mobile phone ban was in place, though only Greece had a near-universal prohibition (Figure 5). Other countries have enacted laws encouraging schools to regulate device use or restrict it to teacher-led instructional activities.

3 Students' use of digital resources for learning

This section focusses on students' use of digital resources for learning. In PISA 2022, 15-year-old students were asked how many hours a day they usually use digital resources for learning and for leisure activities at school, before and after school, and on weekends.· Based on this data, we first examine when and how often students spend time using digital resources for learning and how this varies across different groups of students. Second, we examine systematic differences across schools in the time students spend using digital resources. Third, it looks in greater detail at the types of digital resources students use for learning and the activities that they use them for, using cluster analysis to identify distinct profiles of digital learners across countries.

How intensively do students use digital resources for learning?

The time that students spend using digital resources for learning at school varies widely across countries

The amount of time that students spend using digital resources (for both leisure and learning activities) varies widely across education systems (OECD, 2023, pp. 410, Table ||.B1.5.62;4). Figure 6 shows the time that students spend using digital resources at school for learning and leisure activities, compared with the time they spend in regular lessons per day. On average across OECD countries, 15-year-old students reported having 4.7 hours of regular school lessons per day and spending 2 hours per day using digital resources for learning activities at school. (It is important to keep in mind that students may use digital resources at school but outside of regular lessons). Students in Czechia, France, Germany, Greece, Ireland and Slovenia spend fewer than 1.5 hours per day learning using digital resources at school, while students in Denmark, Norway, Iceland and Sweden spend 3.0 hours per day or more (OECD, 2023, pp. 195, Figure 1.5.15).

Although the use of digital resources for leisure and for learning tend to go together, there are notable differences in emphasis across countries

Across countries, the time that students reported spending using digital resources for leisure and for learning (across settings, over the course of the week) is positively correlated, meaning that countries where students spend a lot of time engaged in digital learning activities tend to be the ones where they spend more time using digital resources for leisure too. Nevertheless, there are notable differences in the emphasis placed on the pedagogical vs. leisure use of digital resources across countries. In the majority of education systems, students spend more time using digital resources for leisure than for learning (26.5 vs. 20.5 hours per week on average), especially in Czechia, Estonia, Hungary, Israel and Poland. However, in a few countries, including Denmark, Iceland and Japan, students reported spending more time using digital resources for learning than for leisure (Figure 7).

Students use digital resources to learn in different settings and at different times

15-year-old students not only use digital devices to learn at school, but also at home. Overall, students reported spending 20.5 hours per week learning using digital resources on average across OECD countries. This includes 9.9 hours per week at school, but also a substantial amount of time (7.5 hours per week) before and after school on weekdays and 3.1 hours on the weekend (Figure 8). Although there is more cross-country variation in the time students spend learning with digital resources at school than outside of school, in countries where students spend more of their school hours learning with digital resources, they also tend to spend more time learning with digital resources before and after school. This may reflect that countries' general attitudes towards the use of digital resources or access to digital infrastructure affects learning both inside and outside of school. It could also indicate that teaching and learning styles modelled at school influence how students approach learning outside of school or whether they are expected to use digital resources to complete their homework.

The moderate use of digital resources for learning is positively associated with mathematics performance and several indicators of effective teaching practices

Although PISA 2022 data do not permit the identification of causal relationships, they suggest that the moderate use of digital resources for learning at school is associated with higher mathematics performance (OECD, 2023, р. 193 f.m). On average across OECD countries, students who reported not spending any time using digital resources for learning at school (14% of all students) score 456 points in mathematics. Students who spend up to one hour per day using digital resources for learning activities at school (31% of students) score 14 points higher, after controlling for students' and schools' socio-economic profile. This positive relationship is observed in over half of all systems with available data. The association starts to noticeably reverse once the use of digital resources for learning at school exceeds 3-5 hours. On average across OECD countries, students who spend between 5 and 7 hours per day using digital resources for learning activities at school (8% of students) score 10 points lower in mathematics than students who spend between 3 and 5 hours per day, after controlling for students' and schools' socio-economic profile. Those who spend over 7 hours per day score even lower (OECD, 2023, pp. 410, Table 11.B1.5.664)).

A similar pattern can be observed in the amount of time that students use digital resources for learning activities outside of school. Taken together, the overall time that students spend learning using digital resources across settings tends to be positively associated with mathematics performance when controlling for students' and schools' socio-economic profile - up to a point. Across countries, the positive association between performance and overall digital learning time tends to diminish above 11 or 15 hours a week before diminishing more rapidly and turning negative for usage above 31 or 41 hours a week (compared to students who reported using digital resources for learning fewer than 5 hours per week) (Figure 9).

The PISA 2022 Student Questionnaire asked students a range of questions concerning their mathematics teachers' pedagogical practices and the perceived quality of their mathematics lessons. For example, students were asked how often their mathematics teacher showed behaviours that encouraged mathematical thinking (e.g. if the teacher "encouraged us to 'think mathematically" or "asked us how different topics are connected to a bigger mathematical idea"). Students' responses to these items were scaled to create the index of "Cognitive activation in mathematics: Encourage mathematical thinking" (OECD, 2024, р. 403110). To explore whether the use of digital resources for instruction can help to promote such desirable teaching practices and raise the perceived quality of mathematics lessons, we tested Whether they are associated with students reporting that they spend more time using digital resources in mathematics lessons than in test language lessons.

Across OECD countries with available data, more time spent using digital resources in mathematics than in test language lessons is associated with statistically significant increases in a range of positive teacher behaviours, as perceived by students. This includes higher cognitive activation and self-efficacy, increased teacher support, reduced mathematics anxiety, as well as higher perceived quality of instruction (Figure 10). Students who spend more time using digital resources in mathematics than in test language lessons are also more likely to name mathematics as one of their favourite subjects (not shown in this figure). By contrast, a relatively more frequent use of digital resources in mathematics lessons is also associated with a worse disciplinary climate (Figure 10) as well as a greater likelihood that students get distracted by their own or their peers' digital devices in mathematics lessons (not shown in this figure).

Socio-economically advantaged students and girls tend to spend more time learning with digital resources

Teenagers living in different socio-economic contexts tend to use their time differently, which can contribute to the reproduction of social inequalities. Research on time-use patterns in Germany, for example, indicates that teenagers in high-income families are more likely to spend substantial parts of their weekday afternoons engaging in cultural activities, sports or homework while children in low-income families are more likely to spend long sequences of their weekday afternoons watching TV (Bettháuser, Siglreitmaier and Fasang, 2024, р. 5417). PISA 2022 data suggest that socio-economically advantaged students tend to spend more time using digital resources than disadvantaged students in most contexts, with the exception of their use for leisure at school (Figure 11). ·

Advantaged students spend more time using digital resources for learning (at school, before and after school, and on weekends), as well as for leisure before and after school, and on weekends. The only context in which disadvantaged students use digital devices more is for leisure at school. This is also the usage with the strongest negative association with mathematics performance (controlling for socioeconomic status), which makes it a potential source of learning inequalities (OECD, 2023, pp. 410, Table 11.81.5.77(4)). This pattern can be observed across most OECD countries, although there are some exceptions (see Table A A.1). That disadvantaged students spend less time on digital leisure activities on weekends and after school than their advantaged peers could be explained by differences in their access to digital devices or more trade-offs between digital leisure and other obligations, such as part-time work or care (OECD, 2025718).

There are also clear gender differences in the intensity with which students reported using digital resources in PISA 2022. While boys spend slightly more time using digital resources for leisure at school, girls spend significantly more time using digital resources for learning before and after school, and on weekends (Figure 12). This pattern can be observed across most OECD countries, although there are some exceptions (see Table A A.6). This could indicate that girls have a preference for learning using digital devices. It could also be explained by the fact that girls tend to spend more time studying and doing homework than boys in general (OECD, 2023, pp. 410, Table 11.B1.5.58p4)). However, the pattern remains when controlling for gender differences in students' overall learning time and their level of academic engagement (analyses not included in this working paper).

In which schools do students spend more time learning with digital resources?

School characteristics shape how teachers and students use digital education resources. An inadequate digital infrastructure or an insufficient supply of digital devices and software, for example, can place a constraint on teachers' ability to draw on digital resources in the classroom (Gil-Flores, Rodriguez-Santero and Torres-Gordillo, 20177191). Likewise, schools with a socio-economically advantaged student body may be able to mobilise additional resources or rely on students' private digital devices to support digital learning activities at school or at home (see Figure 11). School leadership practices can also play an important role in shaping schools' approach to digital learning (e.g. whether school leaders encourage and enable teachers' use of digital devices, invest in their schools' digital materials and issue guidance on the use of digital devices) (Tondeur et al., 20127201). Depending on schools' level of pedagogical autonomy, their distinct educational profiles will also influence how much they emphasise the use of technology compared to traditional teaching methods. Finally, teachers' digital preparedness, their confidence and motivation to use digital resources is likely to affect both their own use of digital tools and that of their students.

Table 1 shows the results of a multivariate analysis, which confirms that the time students spend using digital resources for learning at school is associated with a number of school characteristics. Across OECD countries, PISA 2022 data suggest that technical barriers, digital capacity (of both students and teachers), as well as certain school policies play an important role in shaping the use of digital resources. Two factors are particularly strongly and consistently (across countries) associated with a more intensive use of digital resources for learning at school: students' efficacy to engage in self-directed learning and (perhaps surprisingly) the number of computers/tablets in students' homes. As expected, principals' views on their schools' access to digital resources and teachers' preparedness for digital learning are also important factors, though less consistently across countries. The following sections explore each of these factors in turn, highlighting notable differences across OECD countries.

Access to digital infrastructure (particularly in students' homes) is associated with greater use of digital resources for learning at school

Despite the fact that nearly all 15-year-old students in OECD countries now have access to a computer or tablet at their school, their satisfaction with the quality and accessibility of digital devices remains low in some countries (see Section 2). This is also borne out in the association between different measures of schools' digital infrastructure and the time that students spend on learning with digital resources in their schools. Across OECD countries, students attending schools whose principal reported that the capacity to provide instruction is hindered to "some extent" or "a lot" by a lack of digital resources spend 0.6 hours fewer per week learning with digital resources, controlling for a range of other school characteristics. Likewise, having fewer computers per student is associated with reduced digital learning time, on average across OECD countries (Table 1).

This suggests that limited digital equipment can be a barrier to digital learning, even if students have access to a computer at their school in principle. It might be, for example, that there are not enough computers for multiple students or classes to use them simultaneously. However, it should be noted that a lower number of computers per student and principals' reported lack of resources are not consistently associated with reduced digital learning time. While significant across OECD countries, the associations are only significant in 5 and 9 of 31 OECD countries respectively (Table 1) and point in the opposite direction in some countries. It is also conceivable that part of the association between the availability of digital resources and digital learning time is the result of reverse causation, i.e. that schools invest more in digital devices if they place a greater emphasis on digital learning.

By contrast, the number of computers and tablets that students reported having access to at home is consistently associated with more time spent learning with digital devices at school across nearly all OECD countries (Table 1). While this association may be surprising, it could reflect that a more intensive use of digital resources in the classroom often relies on students bringing their personal devices to school. Schools that cannot or choose not to implement a Bring-Your-Own-Device (BYOD) policy may be more restricted in the way teachers can use digital resources for their lessons (OECD, 2023[1]). Schools may also refrain from making intensive use of digital resources for learning if they know that not all students have access to digital devices to use for their homework.

Both students' and teachers' digital preparedness are associated with greater use of digital resources for learning

Teachers' digital capacity is widely recognised to be critical for the successful digital transformation of schooling (OECD, 2023[1]). Results of the OECD's Policy Survey on School Education in the Digital Age show that 29 of 32 participating jurisdictions have addressed teachers' capacity to integrate digital resources in their practice in their central-level digital education strategies (OECD, 2025[3]). PISA captures teachers' digital capacity in the "index of preparedness for digital learning".7 The index is based on principals' perceptions of different factors related to their teachers' preparedness for digital instruction, e.g. whether teachers have the necessary technical and pedagogical skills to integrate digital devices in instruction, whether they have sufficient time to do so and whether they have access to relevant professional learning resources. On average across OECD countries, a one-standard-deviation increase in the index is associated with students spending an additional 0.4 hours per week learning with digital resources at school. However, this relationship is positive and statistically significant in only 9 of 37 countries with available data (Figure 13).

It should be noted that not all principals may have an accurate assessment of their teachers' digital skills. Indeed, students' views in PISA 2022 suggest that their teachers' learning needs with respect to the use of digital resources may be greater than school leaders believe. On average across OECD countries with available data, 30.4% of 15-year-olds did not agree that their teachers had the necessary skills to use digital devices during instruction. By contrast, only 11.6% of school leaders in the same countries reported that their teachers lacked the necessary technical and pedagogical skills to integrate digital devices in instruction (OECD, 2023Зрв)). This gap was observed relatively consistently across countries and might be explained by students having higher expectations for their teachers' use of digital resources. It could also be that students' regular, direct experience of their teachers' classroom practice allows them to appraise their digital skills more accurately (OECD, 2025, р. 8p).

PISA 2022 data also point to the important role that students' own capacity plays in enabling the integration of digital resources in the classroom. The PISA index of self-directed learning self-efficacy captures students' confidence in their ability to independently engage in a number of tasks should their school building close again in the future.8 This includes whether students would be confident in using a learning management system, using a video communication programme or finding learning resources online on their own. On average across OECD countries, a one-standard-deviation increase in the index is associated with students spending an additional 0.5 hours per week learning with digital resources at school. This relationship - in contrast to teachers' preparedness - is significant across nearly all OECD countries (Table 1 and Figure 13).

The association between students' self-efficacy and their digital learning time at school could indicate that learning with digital devices makes students more confident in their use. It could also be that teachers moderate their emphasis on the use of digital devices based on their students' ability to successfully engage in digital learning (similar to the role played by students' access to digital resources at home). Students' self-directed learning self-efficacy is even more strongly associated with the time they spend using digital resources before and after school (analysis not presented in this working paper). This is in line with expectations, given that digital learning outside of school depends to a greater extent on students' voluntary and independent engagement.

Different aspects of teachers' digital preparedness appear to matter for their use of digital resources in different countries

Multiple dimensions of teachers' digital preparedness are associated with an increased use of digital education resources, though the extent to which they matter varies across school systems. On average across OECD countries, the time that students spend using digital resources for learning at school is significantly and positively associated with each aspect captured by the OECD index of schools' digital preparedness (Figure 14). For example, students attending schools whose principal agreed (or strongly agreed) that there are effective professional resources for teachers to learn how to use digital devices spend an additional 0.7 hours per week using digital resources for learning.

Similar increases are associated with principals' view that teachers have the necessary technical and pedagogical skills to integrate digital devices in instruction (0.7 hours); that teachers have sufficient time to prepare lessons integrating digital devices (0.6 hours); that the school has sufficient qualified technical assistant staff (0.6 hours); that teachers are provided with incentives to integrate digital devices in their teaching (0.5 hours); and that an effective online learning support platform is available (0.5 hours) (Figure 14). While each of these dimensions of teachers' digital preparedness is associated with an increased use of digital resources for learning across OECD countries, none of the associations is statistically significant in more than 11 of the 38 education systems. It thus appears as though the dimensions of teachers' preparation matter to different degrees across school systems and that no single one is consistently more strongly associated with students' digital learning time than the others (see Table A A.7).

School policies shape the time that students spend learning with digital resources, but most policies aim at the quality, rather than quantity of digital learning

While central education authorities play an important role in promoting the effective use of digital technology, schools in many OECD countries have considerable autonomy in shaping their approach to digital education. School-level policies and practices are therefore expected to influence how often students use digital resources for learning. PISA 2022 data indicate that schools with written statements or guidelines on the use of digital devices tend to report higher levels of digital learning time (Figure 15), although the strength of the associations varies across countries (see Table A A.8). This could reflect that central guidance encourages teachers' use of digital resources or increases their confidence. It could also be that the need for rules or guidelines arises particularly in schools where teachers and principals already embrace the integration of digital resources.

By contrast, prohibiting the use of mobile phones on school premises is associated with students reporting significantly less digital learning time. Restricting the use of phones may limit opportunities for students to engage in digital learning activities between lessons or for teachers to integrate personal devices into learning activities. It could also be that schools with mobile phone bans take a more cautious approach to the use of digital devices in teaching and learning in general (Figure 15). These findings are significant since the rationale for restricting the use of mobile phones at school tends to focus on their use for leisure activities and digital distractions, rather than the use of digital resources for teaching and learning.

As expected, school-level policies that are explicitly designed to promote the use of digital technology are also associated with students spending more time using digital resources for learning. On average across OECD countries, students in schools that schedule time for teachers to meet to share, evaluate or develop instructional materials and approaches that employ digital devices spend an additional 0.4 hours per week using digital resources for learning. Schools with a specific programme to promote collaboration on the use of digital devices among teachers made more use of digital resources for learning as well (Figure 15).

Both of these strategies are relatively common, but - as is the case for most policies included in Figure 15 - they are far from universal in most OECD countries. This suggests that they are implemented at the discretion of schools (or with their involvement), rather than mandated centrally. On average, 58% of 15year-old students attend schools that schedule dedicated time for teachers to work on digital education approaches and materials, ranging from 30% or less in Argentina, Belgium, Croatia and Hungary, to more than 90% in Iceland and Singapore. Although the association of the policy with digital learning time is positive in the majority of countries, it is only statistically significant in seven countries, controlling for students' and schools' socio-economic profile (Figure 16).

We see similar results for schools that implement specific programmes to promote collaboration on the use of digital devices among teachers. On average, 55% of 15-year-old students attend schools with such programmes in place, ranging from less than 30% in Croatia and Poland, to 80% or more in Portugal and Singapore. Although the association of the policy with digital learning time is positive in most countries, it is only statistically significant in seven countries, controlling for students and schools socio-economic profile (Figure 17).

Schools' socio-economic profile matters for the intensity of digital learning in some countries and students in private schools tend to spend more time learning with technology

As discussed above, socio-economically advantaged students tend to spend more time using digital resources for learning both at school and at home (Figure 11). Once school-level factors related to the availability of digital devices, teachers' preparedness for digital learning and schools' digital policies are controlled for (see Table 1 for a full list of controls), this association largely disappears, both across OECD countries and in most individual countries (Figure 18). (The association between socio-economic status and the use of digital resources before and after school or on weekends persists, indicating the reduced influence of the school environment on students' learning at home).

Yet, students' learning experience and use of digital devices at school may not only be affected by their own socio-economic conditions, but also by the socio-economic composition of their school. To the extent that such an association is observed at the school-level, it could be indicative of peer effects or advantaged schools' better access to resources. As for students' individual socio-economic status, there is little evidence of a systematic relationship between schools' socio-economic profile and the amount of time that their students spend using digital resources for learning at school across OECD countries, once technological and skills-related barriers are controlled for (Table 1). However, significant associations in both directions do exist in some countries. Students in socio-economically advantaged schools are spending significantly more time using digital resources for learning in countries including Australia, Canada, New Zealand, Switzerland and the United Kingdom. By contrast, advantaged schools make less use of digital resources for learning in countries including Belgium, Finland, Hungary, Poland and the Slovak Republic (Figure 18).

On average across OECD countries, attending a private school is associated with students spending an additional 1.1 hours per week using digital resources to learn at school, after controlling for students' and schools' socio-economic profile. This association is significant and positive in 10 of the 35 countries with available data and exceeds 3 hours in Australia, Italy and Portugal. Colombia and Slovenia are the only countries where the use of digital resources for learning is significantly more intensive in public than in private schools (Figure 19). The greater use of digital resources in private schools can be triangulated with teachers' reports from the PISA 2022 Teacher Questionnaire, which show a similar pattern across the 16 countries and economies with available data.

Several factors could explain these differences between school sectors. For example, principals of private schools tend to report that their schools are better prepared for digital learning than public school principals (OECD, 2023, pp. 410, Table II.B1.5.30[4]) and are less likely to report a lack of digital resources (OECD, 2023, pp. 410, Table II.B1.5.19[4]). Greater autonomy may permit private schools to embrace new technological developments more quickly than public schools, which may be required to use central curricula or assessments. It could also be that the intensive use of digital resources is a way for private schools to attract families' attention and distinguish themselves with a unique pedagogical profile in education systems where schools compete for students (OECD, 2017[21]).

In some countries, schools' approaches to digital learning appear to vary more than in others

Asking students how much time they spend using digital resources for learning at school will inevitably produces a range of responses, depending on the school they attend, the classes they take, the teachers they are assigned to, their individual approaches to learning and even their ability to accurately estimate the time. Exploring how much of this variation occurs between schools provides an insight into how similar or different schools in each country are in their approaches to digital learning. If the proportion of the overall variation observed between schools is low, schools play a less important role in determining how much time students spend learning with digital resources; instead, it is the characteristics of students, their teachers, or different classes within schools that account for most of the variation and vice versa (OECD, 2023, р. 6411).

On average across OECD countries, differences between schools account for 9% of the overall variation in students' reported digital learning time. In comparison, as much as 32% of the observed variation in students' mathematics performance occurs between schools (OECD, 2023, pp. 252, Table |.B1.2.12j1n). However, the between-school variation in students' digital learning time differs widely across countries. In four countries (Australia, Ireland, Spain and Switzerland), between-school differences account for more than 15% of the total variation. By contrast, in five countries (Croatia, Estonia, Italy, Romania and Slovenia), differences between schools account for less than 2% of the total variation in digital learning time at school (Figure 20).10

More between-school variation in students' learning experience could be indicative of a higher level of school autonomy. However, in the case of digital learning time, the amount of variation does not appear to be strongly related to the level of school autonomy or schools' responsibility for their curricula (as measured in PISA 2022). Among the countries with the highest levels of between-school variation are both those with high levels of school autonomy (such as the United Kingdom) and low levels of school autonomy (such as Spain). The same is true for countries with low levels of between-school variation, which include both lowautonomy systems, such as Greece, and high-autonomy systems, such as Estonia and Denmark (Figure 20). One possible explanation is that teachers' use of digital resources for learning remains relatively independent of central prescriptions and regulations, even in school systems with low levels of pedagogical autonomy. It could also be that other factors, such as variation in schools' technical infrastructure or staff competency, exert a greater influence on the heterogeneity observed across schools.

The variance decomposition analysis presented above can be triangulated with data from the PISA 2022 Teacher Questionnaire and the index of teacher use of ICT, which is based on teachers' reports of the activities for which they use digital resources (rather than students' reports of the time they spend engaged in digital learning). The average proportion of the total variation in teachers' ICT use observed between schools is similar to that of students' digital learning time (8%, compared to 9%) across the 18 countries and economies that administered the PISA 2022 Teacher Questionnaire. However, among the countries with available data, the between-school heterogeneity in digital learning differs significantly depending on the indicator used. For example, Australia has a high level and Brazil a low level of between-school heterogeneity if measured by students' digital learning time. However, the opposite is the case when using the measure based on teachers' reported ICT use (analysis not shown in this working paper). These discrepancies underline the need for further work to better understand the diversity of approaches to digital education between and within schools as well as the factors that explain it.

Patterns in schools' level of digitalisation (cluster analysis)

The analyses above highlight a range of factors that are associated with a more intensive use of digital resources for learning in schools, as well as cross-country differences, for example in the extent to which schools' use of digital resources is affected by their socio-economic profile or teachers' digital preparedness. The results also point to significant variation within countries, suggesting that some schools more readily embrace the digital transformation whereas others emphasise more traditional pedagogical approaches or use digital tools primarily outside the classroom. The following sections further explore these distinct patterns in schools' use of digital resources, investigate their prevalence across countries and identify what distinguishes schools with different levels of digitalisation.

We use cluster analysis to group schools based on four variables that represent relevant dimensions of their use of digital resources for learning:

1. Teachers' capacity to use digital resources for instruction (as measured by the PISA index of preparedness for digital learning, based on principals' reports);

2. The quality of the school's digital infrastructure (as measured by principals' reports whether the school's capacity to provide instruction hindered by lack of digital resources);

3. The school's preparedness for remote instruction (as measured by principals' assessment of whether their school would be prepared for providing remote instruction if the school building were to be closed to students for an extended period in the future); and

4. The intensity with which digital resources are used for learning (as measured by the hours that students reported using digital resources for learning activities at school).

The clusters were created using k-means clustering and Euclidean distances, a technique that minimises within-cluster variance. The four selected variables are weakly correlated (no correlation is above 0.25 at the school level), which suggests that each adds unique information to the clusters. The number of clusters was chosen based on the interpretability of the resulting groups, the model's parsimony and its statistical fit (using the Caliñski-Harabasz pseudo-F statistic). ·· All countries and economies with available data were used in the estimation of the clusters.

The four identified clusters represent distinct patterns in schools' approaches to the use of digital resources (see Table 2 and Figure 22). It should be noted that the analysis is primarily intended to describe patterns in schools' use of digital resources, rather than to evaluate their pedagogical efficacy. The clusters' "level of digitalisation" should therefore not be equated with the quality of instruction they offer, or even the effectiveness of their use of digital resources. In each of the clusters, a majority of principals reported feeling confident in their schools' ability to engage in remote instruction and most reported that they do not suffer from a lack of digital resources. However, the groups vary significantly in teachers' preparedness to use digital resources for instruction, as reported by their principals, and in the amount of time that students spend learning with digital resources:

* Cluster 1 (High level of digitalisation): Students in these schools make intensive use of digital resources for learning. Most schools have a reliable digital infrastructure and nearly all are confident in their ability to engage in remote instruction. Their teachers are very well prepared to use digital resources.

* Cluster 2 (Medium-high level of digitalisation): Students in these schools make moderate-to-intensive use of digital resources for learning. Most schools have a reliable digital infrastructure and are confident in their ability to engage in remote instruction. Their teachers are relatively well prepared to use digital resources.

* Cluster 3 (Medium-low level of digitalisation): Students in these schools make limited-to-moderate use of digital resources for learning. Many schools have a reliable digital infrastructure and are confident in their ability to engage in remote instruction. Their teachers are less prepared to use digital resources

* Cluster 4 (Low level of digitalisation): Students in these schools make limited use of digital resources for learning. Many schools have a reliable digital infrastructure and are confident in their ability to engage in remote instruction. Their teachers are not well prepared to use digital resources.

Schools with "medium-low" or "low" levels of digitalisation dominate, while highly digitalised schools are a small minority in most countries

Based on the cluster analysis, most schools across OECD countries have a "medium-low" (37%) or "low" (33%) level of digitalisation. Schools with a "medium-low" level are the most prevalent group in 20 of the 40 systems with available data, whereas schools with a "low" level of digitalisation constitute a plurality in 11 systems. Schools with a "high" level of digitalisation are a small minority in most countries, comprising less than 10% of schools in 33 of the 40 systems with available data and 8% on average across OECD countries. Notable exceptions are Australia, where highly digitalised schools constitute the largest group, and Iceland and Sweden, where they are the second largest group, following those in the "medium-high" category. By contrast, schools with a "low" level of digitalisation make up over 30% of all schools in 18 systems, and over 50% in 7 of 40 systems (Figure 22).

Highly digitalised schools are more likely to have policies intended to foster teachers' effective use of digital devices

PISA 2022 asked principals about their schools' policies concerning the use of digital devices. This includes schools' approach to regulating their use (e.g. based on written statements, rules or formal guidelines) and specific practices aimed at supporting teachers' effective use of digital devices in the classroom (e.g. by promoting collaboration on the subject or by scheduling time for teachers to share, evaluate or develop instructional materials and approaches that employ digital devices). In some countries and economies, schools' policies concerning digital devices may be set at the central level (for example in the case of mobile phone bans). Nevertheless, for each of the policies covered by the questionnaire, PISA data show substantial variation across schools within countries.

As discussed above, some of these policies are associated with increased digital learning time (Figure 15). However, on the whole, they are not strongly related to schools' level of digitalisation, as identified in the cluster analysis. For instance, whether or not schools regulate the use of digital devices is not associated with their degree of digitalisation (as identified in the cluster analysis above) even though students tend to spend less time using digital resources for learning in countries that ban the use of mobile phones (Table 1). Having a written statement about the use of digital devices on school premises or guidelines for their use in class also bears little relationship with schools' level of digitalisation. By contrast, highly digitalised schools are more likely to have policies aimed at fostering teachers' effective use of digital devices (e.g. scheduling time for teachers to meet to work on the use of digital devices or having a specific programme to promote teachers' collaboration on the use of digital devices) (Figure 23).

How do students use digital resources for learning?

What matters most for students' learning outcomes and well-being is not the time they spend learning with digital resources, but how they are integrated in instruction (Forsström et al., 2025[2]). In PISA 2022, students were not only asked how much time they spend using digital resources for learning, but also how frequently they use digital resources to engage in different school-related activities (both during lessons and outside of school, e.g. when completing homework) (OECD, 2021[6]). The following sections analyse how students use digital resources for learning, identify distinct patterns in students' use and discuss how they are aligned with the evidence on effective practices.

Students' use of digital resources for different learning activities remains relatively limited and infrequent

The PISA 2022 Student Questionnaire covers ten types of learning-related activities that students can engage in using digital resources. These activities can be grouped into broad categories based on their role in the learning process and the competencies they require or promote, for example:

* Accessing and managing information (finding information online; collecting and recording data);

* Creating and communicating digital content (writing or editing text for a school assignment; creating a multimedia presentation; collaborating with other students to create digital content);

* Analysing and interpreting data (analysing data that you have collected yourself; reporting or sharing results from your own experiments or investigations);

* Planning and managing work (planning and managing work or projects; tracking the progress of your own work or projects);

* Engaging in interactive digital learning (playing digital learning games).

At the country level, the frequency with which students engage in different digital learning activities is highly correlated with the overall time they spend learning with digital resources (Figure 8). In countries where students reported spending more time learning with digital resources (at school and at home) (e.g. Denmark, Sweden and Iceland), they were also more likely to report regularly engaging in the different learning activities described above. The opposite is true in many countries where students reported spending little time using digital resources for learning (e.g. Czechia, Germany and Japan). However, there are notable exceptions. Although students in Finland and Korea reported spending long periods of time learning with digital resources, comparatively few students reported regularly engaging in digital learning activities like creating multimedia presentations, analysing data or playing digital learning games. The opposite is true in countries like Ireland and Costa Rica (see Table A A.9). № is therefore important to distinguish between the time that students spend learning with digital resources and the range of specific digital learning activities that they engage in.

Across OECD countries, students tend to engage in these digital learning activities relatively infrequently. For seven of the ten activities, only between 21% and 27% of students reported engaging in them at least once a week. The regular creation of multimedia presentations is even less widespread (18%), while slightly more students reported regularly writing or editing text for a school assignment (37%) or "finding information online about real-world problems or phenomena" (35%) (Figure 24). The activities that students most frequently engage in are of relatively low complexity and make limited use of the distinct potential of digital technologies. However, more engaging and cognitively demanding activities, such as the engagement in digital learning games or data analysis, are not much less frequent (OECD, 2023, р. 262/22).

Patterns in students' approaches to digital resources (cluster analysis)

Beyond international differences, there is considerable heterogeneity within countries and schools in how students use digital resources. Even in systems with limited overall use, some students and schools make intensive and diverse uses of digital resources and vice versa. Identifying these outliers and their characteristics can be important for policy makers trying to identify successful practices or schools in need of targeted support. Analysing the use of digital resources at the student-level also helps to reveal patterns of digital learning behaviours that country-level averages may obscure. We use cluster analysis to capture these patterns and group students based on the digital learning activities that they reported engaging in at least once a week. The analysis produced four clusters of students, each representing distinct digital learning patterns (visualised in Figure 25):

* Cluster 1 ("highly engaged digital learners", 13.7%): Students in this group make frequent and intensive use of digital resources for learning. They engage in almost all types of activities on a weekly basis.

* Cluster 2 ("moderate exploring digital learners", 15.4%): Students in this group make moderate use of digital resources for learning. They use digital resources mostly for the collection and presentation of information, and less for collaboration, analysis or project management.

* Cluster 3 ("moderate planning digital learners", 14.0%): Students in this group make moderate use of digital resources for learning. They use digital resources mostly to plan and track projects and to collaborate with others, and less for independent creative or analytical tasks.

* Cluster 4 (low-intensity digital learners, 57.0%): Students in this group make hardly any use of digital resources for learning. Few of them engage in any type of activity on a weekly basis.

"Low-intensity" digital learners are a plurality in nearly every country, but "highly engaged" digital learners constitute a sizeable minority in some

Students exhibiting each of the digital learning patterns identified above can be found in every OECD country, though their prevalence varies significantly. The plurality of students in every country belongs to the "low-intensity digital learners" group, with the exception of Denmark, where the "moderate exploring" and "highly engaged" learners are the largest groups. On average across OECD countries, the "lowintensity" digital learners account for 57% of all students, ranging from only 16% in Denmark and 32% in the United States to 70% or more in Czechia, Germany, Japan, Korea and Turkiye. On average across OECD countries, 14% of students are classified as "highly engaged" users, ranging from less than 7% in Czechia, Germany and Japan to 25% or more in Denmark and the United States. Both the "moderate planning" learners and the "moderate exploring" learners account for 14% and 15% of students on average respectively (Figure 26).

In some countries, digital resources play an important role in supporting the learning process indirectly outside of classroom hours

In addition to their direct use for teaching and learning, digital resources can also play a more indirect role, supporting teachers and students with tasks such as communicating with teachers and peers or sharing learning materials and instructions. Across OECD countries, 35% of 15-year-old students reported using digital resources outside classroom hours at least once a week to communicate with their teachers and 54% reported using them to communicate with other students about schoolwork. About half use digital resources on a weekly basis to browse the internet to follow up on lessons (46%), to browse the internet for schoolwork (53%), or to search for information on school-related activities or assignments (49%). A similar proportion reported regularly uploading their work to a school portal or another platform to share materials with their teachers and/or students (43%), to receive or download assignments or instructions (45%), or to see their grades or results from specific assignments (53%)

4 Teachers' use of digital resources for instruction

As discussed above, the success of digital education critically depends on how teachers integrate technologies into the learning process (Bulman and Fairlie, 20167231; OECD, 2023, р. 261). Results from PISA underline the significant progress that teachers have made in using digital resources since the COVID-19 pandemic. Across OECD countries, the share of students whose principals reported that teachers have the necessary technical and pedagogical skills to integrate digital devices in instruction increased from 64.5% in 2018 to 87.6% in 2022 (OECD, 2023, pp. 409, Table |1.B1.5.3214). The previous sections have considered the implications of this digitalisation by analysing teachers' use of digital resources from the perspective of students. This section complements the picture with the perspective of teachers, drawing on data from the PISA 2022 Teacher Questionnaire (OECD, 2021191). The analysis focusses on the digital tools that teachers report using in their lessons as well as the activities that teachers report using them for. For both, we use cluster analysis to identify distinct patterns in teachers' use of digital resources and to describe their prevalence across countries. Finally, we examine factors that might shape teachers' engagement with digital resources, including their demographic characteristics, training and school context.

To the extent that students' use of digital resources at school reflects their engagement in teacher-led activities and teachers' pedagogical decisions, we expect students' and teachers' experience of digital learning to be aligned. Nevertheless, teachers' reports provide additional information that can enrich our understanding of the way digital resources are used for teaching and learning. Teachers can provide detailed insights into their uses of digital education technologies that may not be apparent to students (e.g. in preparing learning materials or analysing student data). They can also illuminate how digital teaching practices are shaped by teachers' training and other characteristics that are relevant to policy makers seeking to support educators in using digital resources to enhance their practice. The PISA 2022 Teacher Questionnaire was administered in 18 countries and economies (6 OECD Member countries, 2 accession candidate countries and 10 non-Member economies). ? Despite the limited geographical coverage of the Teacher Questionnaire, its analysis can generate generalisable insights that are relevant to other OECD education systems.

Which digital tools do teachers use in lessons?

Teachers frequently use lower-complexity digital tools for a range of purposes

The PISA 2022 Teacher Questionnaire asked educators about their use of 14 different types of digital tools and whether they had used them "never", "in some lessons", "in most lessons" or "in every or almost every lesson" this school year. The tools vary in their complexity and the technological competency required to operate them, including, for example, word processors as well as simulation software. While some of the resources can be assumed to be primarily operated by the teacher (e.g. data logging and monitoring tools), others place students in an active role (e.g. digital learning games). The full list of tools is presented below, grouped based on their primary purpose, although many of them can be used for a range of activities, depending on how teachers employ them. (Teachers responding to the questionnaire were not presented without this classification or further contextual information):

* Instruction and assessment tools (tutorial software or practice programmes; digital learning games; interactive digital learning resources, e.g. quizzes and instructional content);

* Content creation tools (word processors or presentation software; multimedia production tools; graphing or drawing software);

* Monitoring tools (data logging and monitoring tools; e-portfolios);

* Data analysis tools (spreadsheets; simulations and modelling software);

* Knowledge organisation tools (concept mapping software; computer-based information resources, e.g. websites and wikis);

* Communication tools (social media; communication software, e.g. email and blogs).

Across the 18 countries and economies with available data, the digital tools most frequently used by teachers (i.e. that more than 30% of teachers reported using in most lessons) cover a range of purposes. This includes communication tools (communication software [30%]), instruction and assessment tools (tutorial software or practice programmes [34%] and interactive digital learning resources [40%]), knowledge organisation tools (computer-based information resources [42%]) and content creation tools (word processors or presentation software [53%]) (Figure 28). Tools for data analysis (spreadsheets [20%] or simulations/modelling software [11%]) and tools for the monitoring of student progress (data logging/monitoring tools [20%] or e-portfolios [15%]) are used less frequently.

Notably, for any given purpose, the lower-complexity tools tend to be more frequently used than highercomplexity tools. For example, among the content creation tools, a smaller share of teachers frequently uses graphing or drawing software (15%) and multimedia production tools (22%) than word processors. Likewise, among the data analysis tools, spreadsheets are more frequently used than modelling software. This could be because higher-complexity tools have a narrower range of applications or because their use requires a higher level of ICT competency on the part of both teachers and students (Figure 28).

Patterns in the digital tools that teachers use (cluster analysis)

Although the aggregate analysis presented above suggests that teachers employ digital tools for a range of purposes (with a more frequent use of lower-complexity tools), prior research indicates significant heterogeneity in teachers' approaches to the use of digital resources (Graves and Bowers, 2018[24]; Lee, 2024[25]). A large body of research has tried to systematise this variation by developing typologies to describe teachers' use of ICT, considering both teachers' attitudes and beliefs concerning the use of ICT as well as their integration of ICT in the classroom. Efforts to classify technology users based on their adoption of new ideas and technologies go back at least to Rogers (1962[26]), who distinguished between five groups of users (innovators, early adopters, early majority, late majority and laggards) based on their role in the diffusion of technologies. Since then, several typologies have focussed specifically on the teaching profession, seeking to describe the process by which they adopt innovative practices and use digital resources (Graves and Bowers, 2018[24]).

While earlier studies have mostly employed small-n qualitative research designs (Donnelly, McGarr and O'Reilly, 2011[27]; Mama and Hennessy, 2013[28]), more recent studies have studied large datasets to identify patterns in teachers' approach to digital learning using latent class analysis (LCA) (Graves and Bowers, 2018[24]; Lee, 2024[25]). This working paper is, to our knowledge, the first to use PISA 2022 data to analyse teachers' use of digital resources in this way. As above, we use cluster analysis to identify distinct groups of teachers based on their reported use of digital tools for teaching. The analysis yielded three clusters of teachers, each characterised by a distinct set of digital tools that they reported using in every, almost every, or most lessons (Figure 29):

* Cluster 1 ("high-intensity users" of digital tools, 12.5%): Teachers in this group make frequent use of a wide range of digital tools for teaching and account for most of the profession's use of digital technology. The tools they use mirror the overall pattern, with a more moderate use of some narrower or higher-complexity tools (see Figure 28).

* Cluster 2 ("selective users" of digital tools, 33.6%): Teachers in this group make moderate to frequent use of some of the more commonly used digital tools (e.g. word processors, presentation software, computer-based information resources or communication software). They make more infrequent use of most other tools, indicating a focus on a limited set of essential or more established digital tools.

* Cluster 3 ("minimal users" of digital tools, 53.9%): Teachers in this group make little use of digital tools for their teaching, which suggests that they mostly focus on "analogue" teaching methods. For all tools covered in the analysis (with the exception of word processors, presentation software and tutorial software), fewer than 10% of the teachers in this group reported making frequent use of them.

Previous cluster analyses drawing on PISA 2018 data in Korea, Germany and the USA distinguished between "minimal", "moderate" and "versatile users", based on whether or not teachers reported ever using different types of digital tools (including "in some lessons") (Lee, 2024[25]). By focussing the analysis on the share of teachers who use digital tools habitually (i.e. not just in some, but at least "in most lessons"), we distinguish more clearly the group of teachers with a distinctly technology-driven approach to teaching (the "high-intensity digital users") and the large group of teachers (the "selective digital users") for whom a limited range of digital tools have become an integral part of their regular teaching practice.

The majority of teachers are "minimal" or "selective users" of digital tools, but most countries also have a small share of "high-intensity users"

The three clusters of teachers identified above are represented in each of the 18 countries and economies with available data, although their prevalence varies significantly. The "high-intensity digital users" account for the smallest share of teachers across countries (12%). With the exception of the United Arab Emirates and Peru, they make up the smallest group of teachers in each jurisdiction, constituting as little as 1% in Germany, 4% in Australia and 5% in Hong Kong (China). The "minimal digital users", by contrast, account for the largest share of teachers across countries and economies (54%) and make up the plurality of teachers in 15 of the 18 jurisdictions with available data. Their share is above 70% in Germany and Morocco. The "selective digital users" are the second largest group of teachers on average (34%) and they are the predominant group in Australia, Portugal and the United Arab Emirates (Figure 30).

For which activities do teachers use digital resources?

Most teachers use digital resources to design tasks, but a sizeable share also involve students (providing feedback and enabling collaboration)

The PISA 2022 Teacher Questionnaire asked teachers how often they had used digital resources to engage in different teaching-related activities this school year ("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")?"· Prior research has demonstrated that teachers' use of technology is best understood as a multidimensional concept, not as a single action or activity. Teachers use technology to different degrees and for different purposes, including for class preparation and direct instruction, to engage in professional communication or research, to accommodate special education needs, to lead students' use of technology, as well as for feedback and assessment (Russell et al, 200329); Graves and Bowers, 201824); McKnight etal, 201669). The PISA 2022 Teacher Questionnaire captures this diversity by providing data on teachers' engagement in nine activities, which could be grouped into broad domains:

* Professional exchange with peers (using digital resources to share ideas or resources with colleagues; taking part in professional communities of practice online);

* Communication with parents and students (using digital resources to communicate with parents or guardians; using digital resources to provide access to instructional material for students who cannot physically attend class; using digital resources to provide feedback to students);

* Instruction and assessment (using digital resources to design tasks; using digital resources to explore new teaching methods; using digital resources to enable student collaboration; using online tools or computer-based testing to assess students' learning).

Between a third and half of the surveyed teachers reported regularly engaging in most of these activities (ie. at least once a week), across the 18 countries and economies with available data. Although teachers' responses do not specify how exactly they use digital resources to carry out, enhance or transform the activities in question, a sizeable share of teachers use digital resources for tasks that are directly related to instruction. Around half of the teachers, for example, regularly use digital resources to design tasks (50.8%). Many also reported using ICT to explore new teaching methods (42.7%) and to enable student collaboration (41.2%). Furthermore, more than a third of teachers (37.6%) reported using ICT to provide feedback to students (this could happen either inside or outside the classroom), while only a quarter (25.9%) use online tools or computer-based testing to assess students' learning (Figure 31).

Teachers also use ICT in their work outside the classroom and for activities that are more indirectly related to instruction. For example, more than a third of teachers reported using digital resources at least once a week to communicate with parents or guardians (35.5%) and to provide access to instructional material for students who cannot physically attend class (44.9%). Teachers also use digital resources to support interactions and professional exchange with colleagues. On average, 40.1% reported using them to share ideas or resources with colleagues and 23.6% engage in online professional communities of practice at least once a week (Figure 31).

Is teachers' use of digital resources aligned with effective teaching practices?

As discussed above, PISA 2022 data lend support to the hypothesis that the moderate use of digital resources for learning (in mathematics lessons) can be associated with higher levels of student performance as well as a range of behaviours associated with effective teaching (see Figure 9 and Figure 10). However, given the cross-sectional nature of the PISA data and the risk of confounding variables or reverse causality, it is difficulty to establish a causal relationship between the overall time spent learning with digital resources and students' performance. It is therefore important to investigate how students and teachers employ digital resources for learning (see Figure 24 and Figure 31) and whether these uses are aligned with practices that were identified as effective in causal studies.

Discussions of digital teaching sometimes distinguish between uses of technology that merely substitute existing instruction activities with their digital equivalents (perhaps rendering them more efficient in the process), and more transformative uses of ICT that enable new, more engaging or more personalised learning activities (OECD, 2023, р. 2631221). For example, reviews of earlier causal studies underlined that computer-assisted learning based on (sophisticated) tutoring systems or educational software can improve students' cognitive achievement because it allows teachers to carry out multiple practices simultaneously (e.g. proposing learning activities tailored to fit students' needs and providing immediate feedback). Another characteristic of these effective tutoring systems is their ability to help teachers integrate new forms of digital content, such as videos and simulation tools (Bulman and Fairlie, 20163); Escueta et al., 2017131).

While a detailed discussion of more recent evidence on effective digital learning practices is beyond the scope of this working paper, an in-depth OECD literature review suggests that the impact of digital tools on student outcomes depends on teachers' ability to implement them strategically (Forsstróm et al., 2025121; Forsstrôm et al., 2025112). Effective digital learning requires teachers to act as pedagogical designers who strategically select tools, scaffold learning to manage cognitive load, guide students in developing digital literacy, and thoughtfully balance digital and analogue methods. Managing the increased risk of digital distractions requires teachers to design structured tasks to keep students focussed and implement classroom management strategies that build on relational aspects, such as trust, mutual respect and positive teacher-student relationships. Evidence also suggests that the use of digital resources needs to be carefully aligned with curriculum standards and linked to students' learning goals (Forsstróm etal., 2025121).

It is difficult to establish an equivalence between the characteristics of effective digital teaching practices identified in the literature and the data derived from the PISA 2022 Teacher Questionnaire, given the broad nature of the survey items and their reliance on teachers' self-reporting. The "transformative" practices identified in the literature are most closely related to the activities related to "instruction and assessment" identified above (i.e. using digital resources to design tasks, using digital resources to explore new teaching methods, using digital resources to enable student collaboration, and using online tools or computer-based testing to assess students' learning). However, for each of these activities it is possible to imagine digital technologies being used in either transformative or merely "substituting" ways. Likewise, teachers' use of digital technologies for more peripheral tasks (such as professional exchange with peers or communication with parents and students) could - while not directly enhancing student learning - have an indirect effect on instruction, e.g. by saving time for teaching-related activities or by facilitating professional learning.

Patterns in teachers' use of digital resources for different activities (cluster analysis)

The PISA data clearly show that teachers, in the aggregate, engage in some teaching-related activities using digital resources more frequently than in others. Whether teachers engage in any given activity is likely to be shaped by their individual characteristics and dispositions as well as structural opportunities and constraints. For example, teachers with a high level of digital self-efficacy serving in schools with abundant digital resources are likely to use digital tools for a broader range of activities, whereas teachers who lack the requisite infrastructure or prefer traditional teaching methods may be less inclined to use digital resources for any of the activities described above. We expect these factors to give rise to systematic variation and patterns in teachers' digital behaviour. Following the same method as above, we identify four distinct clusters of teachers based on the activities they reported engaging in at least once a week using digital resources (visualised in Figure 32):

* Cluster 1 ("Highly engaged", 21.9%): Teachers in this group make frequent of digital tools to engage in the full spectrum of professional activities in and outside of the classroom. At least two thirds of teachers in this group report regularly using digital resources for each of the activities covered by the questionnaire. In contrast to Clusters 2 and 3, the "highly engaged" teachers also regularly use digital resources to assess and provide feedback to students and to engage in professional communities of practice.

* Cluster 2 ("Moderately engaged [instructional]", 21.2%): Teachers in this group make regular use of digital resources with a focus on pedagogical activities in and outside the classroom, such as designing tasks, exploring new teaching methods and enabling student collaboration. They make little use of digital technologies to engage in professional collaboration or external communication.

* Cluster 3 ("Moderately engaged [communicating]", 15.2%): Teachers in this group make regular use of digital resources to communicate with parents, guardians or peers and to share information with students. In contrast to Cluster 2, they make little pedagogical use of digital resources to prepare their lessons or directly in the classroom.

* Cluster 4 ("Minimally engaged", 41.7%): Teachers in this group rarely use digital tools for professional activities, with no more than 13% of them reporting that they use digital tools regularly for any of the covered tasks. The few digital activities that some teachers in this group regularly engage in tend to take place outside the classroom (communicating with parents and colleagues or designing tasks).

Across countries, teachers in the "highly engaged" group are marginally younger than those in the "minimally" or "moderately engaged" groups and are more often female. On average across the 18 jurisdictions, the "highly engaged" teachers are 42.3 years old, compared to 42.4 years for the "moderately engaged (instructional)", 43.2 years for the "moderately engaged (communicating)" and 44.4 years for the "minimally engaged". 63-64% of the "highly engaged" and "moderately engaged" teachers are female, compared to 56% of the "minimally engaged" teachers (analysis not included in this working paper). In general, the groups of teachers identified in the cluster analysis differ more with respect to their professional learning behaviour than their demographic characteristics. This contrasts with the clear age and gender gaps that are evident in teachers' use of digital resources as measured by the PİSA index of teacher use of ICT (discussed further below).

The results of the cluster analysis complement previous work on teachers' engagement with digital technologies. Graven and Bowers (20181241), for example, used latent class analysis to develop a typology of technology-using teachers in the United States, based on their use of ICT for instruction and productivity, their professional preparation, and their disposition towards professional learning. The analysis identified four distinct clusters of teachers: "Dexterous" (using technology for a variety of functions) (24.4%), "Evaders" (using technology little and not directing students to use technology either) (22.2%), "Assessors" (using technology mostly to practice basic skills or with drill and practice software) (28.4%), and "Presenters" (using technology mostly for their own and their students' classroom presentations) (24.8%). Whereas the "dexterous" teachers and "evaders" share similarities with the highly and minimally engaged teachers identified in the PISA 2022 data respectively, the "moderately engaged (communicating)" and "moderately engaged (instructional)" teachers identified in this analysis have no obvious equivalent in previously reported patterns.

While "minimally engaged" teachers are the largest group in most countries, the behaviour of digitally active teachers varies considerably across education systems

The four patterns of teachers' digital technology use identified above can be found, to different extent, in each of the 18 countries and economies with available data. The "minimally engaged" account for 41.7% of teachers on average and are the largest share of the teacher population in 15 of 18 education systems. In most countries and economies, the minimally engaged account for 40-60% of the teacher population (and more than 60% in Korea and Morocco). The "highly engaged" teachers are the second largest group, making up 21.9% of the teacher population on average. They are the largest group in the Dominican Republic and the United Arab Emirates, while accounting for less than 10% of teachers in Germany and Morrocco (Figure 33).

The "moderately engaged (instructional)" and "moderately engaged (communicating)" technology users typically account for 10-30% of teachers, respectively. While teachers in some countries (such as Australia, Peru and Portugal) are relatively evenly divided across the four identified groups, teachers in countries like Brazil or Hong Kong (China) are more polarised between highly and minimally engaged users of technology. The data also show that some countries have a large share of teachers making moderate use of technology for instructional purposes (e.g. 30.7% in Germany). In other countries, few teachers are minimally engaged but a large share use technology primarily for communication, rather than pedagogical purposes (e.g. 32.2% in Australia) (Figure 33).

Teachers' use of digital resources and their emphasis on ICT competencies do not always go hand in hand

In recent years, many OECD countries have reformed their curricula to place greater emphasis on students' digital skills, recognising the importance of preparing students to thrive in an increasingly digital world (OECD, 202032). The OECD's Policy Survey on School Education in the Digital Age shows that curricula in two thirds of the responding jurisdictions require the teaching of digital skills while the curricula in the remaining jurisdictions recommend it (Boeskens and Meyer, 2025, pp. 50, Annex Table 2.1133). The PISA 2022 Teacher Questionnaire provides an insight into how this growing emphasis on students' ICT skills in curricula is reflected in the classroom. The index of teachers' emphasis on ICT competencies captures how much emphasis teachers reported placing on different types of digital skills in their instruction. '·

When compared with the index of teacher use of ICT (which captures how frequently teachers reported using digital resources to engage in the nine activities described in Figure 31), it is evident that teachers' use of ICT does not always go hand in hand with a greater emphasis on ICT competencies (Figure 34). Teachers in some countries, such as the Dominican Republic and the United Arab Emirates, score highly in both indexes and teachers in others, such as Korea and Morocco, have low scores in both. However, teachers in countries like Colombia and Costa Rica reported placing strong emphasis on students' ICT skills while making relatively little use of digital resources, whereas the opposite is the case in Australia (Figure 34). Although digital resources can undoubtedly support the promotion of students' ICT skills, these findings serve as a reminder that a country's intensive use of ICT should not be confused with a strong emphasis on students' digital literacy (and vice versa). The same is true at the classroom level, where pedagogical approaches like "computational thinking unplugged" have been used to promote digital literacy without the use of digital devices - an approach particularly adapted to younger students (Bers, Strawhacker and Sullivan, 20221341).

What factors are associated with teachers" use of digital resources?

Whether or not teachers use digital resources for instruction and the way they integrate ICT in the classroom is shaped by factors at the system, school and individual levels. The emphasis on digital learning in curricula, regulations concerning the use of digital devices in schools, and the role of digital methods in initial teacher education can all be expected to shape how educators use digital resources in the classroom. Likewise, characteristics of schools, like the quality of their ICT infrastructure, their student composition, professional learning opportunities and leadership practices can encourage or constrain the use of digital tools for learning (Lee, 202425; Gil-Flores, Rodriguez-Santero and Torres-Gordillo, 2017719). Finally, teachers' individual characteristics, such as their attitudes, subject area and self-efficacy, can affect how they integrate technology into their practice (OECD, 20231; Starkey, 20205). The following sections explore how a selection of these factors is associated with teachers' use of ICT, drawing primarily on the PISA index of teacher use of ICT, which captures the frequency with which teachers reported using digital resources to engage in a selection of nine teaching tasks. '·

Teachers of technology and practical or vocational skills make the most use of digital resources

Teachers of some subjects are more likely to make frequent use of digital technologies in the classroom. Teachers of technology, followed by those teaching practical and vocational skills, stand out as the most frequent users of ICT for teaching (in 15 of the 18 countries and economies with available data). Teachers of physical education are least likely to use ICT for instruction (in 13 of the 18 countries and economies). Among the remaining subjects, mathematics teachers stand out as making significantly less frequent use of ICT than their peers (ranking last in 4 and second to last in 5 of the 18 countries and economies) (Figure 35). These patterns remain when controlling for teacher characteristics, such as gender and age.16

These results, based on teachers' reports, are broadly consistent with students' experiences in PISA 2022. When asked how frequently they use digital resources in different subjects, only 26% of students reported using them in more than half of their mathematics lessons, on average across OECD countries, compared to 29% for lessons in the test language and 33% for science lessons. Likewise, 39% of students reported using digital resources never or almost never in their mathematics lessons, compared to 34% for test language lessons and 28% for science lessons. While broadly consistent across countries, there are some notable exceptions to this pattern, for example Austria and Túrkiye, where students reported using digital resources more frequently in mathematics lessons and less in test language lessons. In contrast to most countries, students in Denmark reported using digital resources less in science lessons than in test language or mathematics lessons (albeit at much higher levels overall) (Figure 36).

Younger teachers are more likely to use digital resources and there are gender gaps in some countries

Younger teachers are significantly more likely to use digital resources for teaching than their older peers (as measured by the index of teacher use of ICT). The gap between teachers aged 56 years or older and 35 years or younger is statistically significant in 13 of the 18 countries and economies with available data from the PISA 2022 Teacher Questionnaire, with a few notable exceptions, including Portugal (Figure 37). In most of the 18 countries, the use of ICT decreases gradually across age groups, i.e. 35 years or younger, 36-45 years, 46-55 years and 56 years or older.17 Similar age gaps, although slightly less pronounced, can be observed in the emphasis that teachers reported placing on students' ICT competencies (Figure 37). These results stand in contrast to the cluster analysis presented above, which found relatively minor age differences between teachers classified as "highly engaged" and "minimally engaged" users of digital technology.

Many countries also show pronounced gender gaps in teachers' use of digital technologies. Female teachers reported using ICT significantly more frequently than their male peers in 8 of the 18 countries and economies with available data. Among OECD countries, the gap is particularly large in Korea, Australia and Portugal, while no gender gap was observed in Colombia. In 9 of the 18 countries and economies, female teachers also reported placing significantly more emphasis on students' ICT skills (Figure 38).

Demand for ICT training remains high and teachers' professional learning plays an important role in shaping their use of digital resources

The importance of teachers' training in the use of digital resources is widely acknowledged among OECD countries. 29 of the 32 jurisdictions with available data in the Policy Survey on School Education in the Digital Age mentioned strengthening teachers' digital capacity in their central strategies for digital education policy (Boeskens and Meyer, 20251331). In TALIS 2018, ICT skills for teaching were among the top three areas in which teachers reported high training needs (OECD, 2019, pp. 165, Figure 1.5.613 ;) and - across the 18 countries and economies that administered the PISA 2022 Teacher Questionnaire - more than half of all teachers reported a moderate or high level of need for further training on the topic. !8

The importance of teachers' initial education and their continuing professional learning is also reflected in the association between teachers' training experience and their classroom practice. Across the 18 countries and economies with available data, teachers for whom ICT skills for teaching were included in their initial teacher education (ITE) or their recent professional development (PD) activities score higher in the index of ICT use than their peers who had not received either. Teachers who were exposed to training on ICT skills during both their ITE and recent PD reported making even more intensive use of digital resources. This pattern is remarkably consistent across the participating countries and economies (Figure 39).

These findings are in line with earlier research on Germany, Korea and the United States using PISA 2018 data, which indicated that teachers' ICT training (both in their ITE and their in-service professional development) is the most important factor associated with teachers' use of digital tools (Lee, 20247251). The association between continuing professional learning and teachers' use of ICT is also reflected in the results of the cluster analysis above (see Figure 32). Teachers who are classified as "highly engaged" users of digital resources are significantly more likely to have taken part in professional development activities over the previous 12 months than those classified as "minimally engaged". This difference is particularly pronounced (25 percentage points or more) for professional learning activities that are directly related to the use of digital resources (i.e. "listening to or watching recorded seminars or online courses about the use of digital resources for teaching" and participating in "in-service training courses about the use of digital resources for teaching"). '9

Teachers who dedicate a greater share of their working time to instruction report engaging less with digital resources

Besides their potential to transform teaching and learning, digital resources are frequently discussed as a means to help teachers engage in some aspects of their work more efficiently (OECD, 2023, р. 19811). Tasks such as lesson preparation, administrative record keeping or student feedback have been argued to hold significant potential for automation, which could help teachers to spend more of their time on core instructional activities (Bryant et al., 20207). At the same time, learning about digital resources and integrating them effectively into activities inside and outside the classroom takes time. PISA 2022 data indicate that teachers who spend a greater share of their working time on teaching engage less with digital resources, as measured by the index of teacher use of ICT (Figure 40). That means they less frequently use digital resources for tasks like lesson preparation, exploring new teaching methods, providing feedback to students, sharing ideas with colleagues or assessing students' learning. This underlines the trade-offs that teachers face when allocating their limited time between competing priorities (Boeskens and Nusche, 20217381). Conversely, across the 18 countries and economies with available data, there is no evidence to suggest that teachers who make frequent use of ICT for teaching-related activities spend less time engaging in general administrative work, which one might expect the use of digital resources could render more efficient (Figure 40).

Teachers in disadvantaged schools make less use of digital resources and there is insufficient data on their use to accommodate immigrant students or students with SEN

Many OECD countries continue to face challenges trying to bridge digital divides between advantaged and disadvantaged students or schools. Across OECD countries, principals of socio-economically disadvantaged schools reported lower levels of preparedness for digital learning and a greater lack of digital resources (OECD, 2023, pp. 410, Tables 11.B1.5.30 and I1.B1.5. 19/47). However, less is known about classroom practices and teachers' use of digital tools in disadvantaged schools. Responses to the PISA 2022 Teacher Questionnaire suggest that teachers in disadvantaged schools (as measured by two different indicators) may indeed make less use of digital resources (Figure 41). Among OECD countries with available data, large, statistically significant gaps could be observed in Costa Rica and Colombia, as well as (to a lesser extent) in Australia and Germany. In Korea, teachers' use of ICT is not significantly correlated with schools' socio-economic profile. In Portugal, the opposite is the case, i.e. teachers in disadvantaged schools make more frequent use of digital resources (Figure 41).

Digital resources can enable teachers to engage in more differentiated forms of instruction and to support students who might otherwise have difficulties following regular lessons. The use of assistive technologies can support students with special education needs (SEN) or overcome language barriers faced by students who are not native in the language of instruction (OECD, 2023, р. 16211). Although data from the PISA 2022 Teacher Questionnaire do not measure the use of assistive technologies directly, they do not indicate that teachers in schools with a high share of immigrant students or students with SEN make more use of digital resources (across the 16 countries and economies with available data) (Figure 41). More granular data on education technologies would be needed to investigate the use of assistive technologies in the future.

Geographical differences in teachers' use of digital resources are small on average but there are rural-urban gaps in some education systems

Digital technologies can be used to improve access to high-quality education for students in rural and remote areas through different forms of blended or distance learning (OECD, 2023, рр. 28, 14171; European Commission et al., 20217391). Chile, France and some US school districts, for example, have therefore invested significantly in rural schools' digital infrastructure (Echazarra and Radinger, 20197151). At the same time, digital divides persist and - on average across OECD countries - rural schools score slightly lower in the index of preparedness for digital learning than those in both towns and cities (OECD, 2023, pp. 410, Table 11.B1.5.30p4).

On average, across the 15 countries and economies with available data from the PISA 2022 Teacher Questionnaire, there are few geographical differences in teachers' use of digital resources. Teachers in rural schools (i.e. in villages or towns with fewer than 15 000 inhabitants) report using ICT only marginally less than their peers in larger agglomerations. However, significant rural-urban gaps in both directions can be observed in some countries. In Australia and Colombia (as well as Panama, Malaysia and Morocco), for example, rural teachers reported using ICT somewhat less frequently than their peers in towns and cities, while ICT is most frequently used in the large cities. This pattern is consistent with digital resources, infrastructure and capacity being more concentrated in urban centres. By contrast, in Germany and Portugal (as well as the United Arab Emirates and the Dominican Republic), rural teachers reported making more frequent use than their peers in (large) cities. This may indicate that teachers in rural or remote schools rely more heavily on digital resources to engage in distance education or professional exchanges with their peers (Figure 42).

5 Policy insights and areas for further investigation

The analysis of PISA 2022 data presented in this working paper provides a range of policy-relevant insights into the use of digital resources for learning activities. Its findings highlight the profound changes that the digital transformation has meant for teachers and students. They also point to the persistent challenges that school systems face in ensuring that digital resources are used effectively to enhance teaching and learning for all while also managing risks related to distractions, inequities and students' well-being. A holistic approach to digital education policy that links curriculum, assessment, infrastructure, teacher development and monitoring is essential to achieve sustainable progress. PISA 2022 data also reveal significant heterogeneity in the way teachers and students approach digital learning, which is critical for policy makers to bear in mind when designing and implementing digital education policies. At the same time, the analysis points to areas that merit further investigation in order to strengthen the evidence base that informs digital education policy.

This section presents seven policy insights emerging from the analysis and outlines their implications for the future of digital education. Taken together, these insights suggest that successful digital transformation requires a holistic and system-wide approach. In that sense, progress depends not only on providing an accessible, high-quality infrastructure, but also on aligning curricula, pedagogical approaches, student assessment, professional learning, leadership and system evaluation. By addressing these dimensions together, education systems can ensure that digital technologies contribute meaningfully to equity, student learning and teacher well-being. Like the analyses presented throughout this working paper, these insights concern PISA's target population of 15-year-old students and may not be applicable to other age groups. Given the diversity of the countries and economies covered by the analysis (with respect to the structures of their school systems, their educational cultures and traditions, their economic and social conditions etc.), these policy insights also need to be considered in context and complemented by other forms of evidence.

Policy insight 1: Monitoring and ensuring not just the general provision, but also the quality and accessibility of digital infrastructure in schools

Schools' digital infrastructure has expanded significantly over the past decade and access to a basic level of digital devices is now close to universal across OECD countries. Well above 90% of 15-year-old students in PISA 2022 reported having access to a computer, the internet, a learning management system and educational software at their school. However, students' reports suggest that the quality and accessibility of their digital infrastructure remain a barrier in some OECD countries. A sizeable share of students reported that their schools do not have enough digital resources for every student or that these resources are not easily accessible within the classroom. Problems with the functionality of digital devices and the speed of their internet connections also appear to be widespread. In contrast to the number of computers available in schools, their quality and accessibility of digital resources also constitute potential sources of socio-economic inequities. In the majority of OECD countries, disadvantaged students are significantly more likely to report problems with the quality of their schools' digital resources, which may account for some of the socio-economic gaps we observe in students' digital learning experience. Now that OECD countries have mostly succeeded in ensuring basic connectivity and access to digital devices in schools, it will therefore be important to ensure that schools' digital infrastructure keeps pace with technological progress and the educational needs of teachers and students. Monitoring should include not only the provision but also the quality, accessibility and functionality of digital infrastructure. It should be supported by clear benchmarks and evaluation frameworks that can be used to adapt resourcing mechanisms to assist schools in achieving their goals. Further analyses should explore which aspects concerning the quality and accessibility of digital resources are most critical to their effective use for instruction.

Policy insight 2: Promoting the evidence-based and purposeful use of digital resources to enhance teaching and learning

PISA 2022 data suggest that students' moderate use of digital resources for learning is associated with higher mathematics performance as well as some indicators of effective teaching practices. On average across OECD countries, students who spend between one and five hours per day using digital tools for learning at school perform better than their peers who use them less or more, after controlling for socioeconomic economic differences. These findings challenge binary narratives advocating for either the rejection or uncritical promotion of digital technologies. At the same time, PISA results underline the importance of managing risks, such as classroom distractions and negative impacts on students' well-being. These findings are also consistent with causal evidence suggesting that digital tools can promote students' learning outcomes if they are used strategically and pedagogically adapted to support specific learning objectives. Yet, data from the PISA 2022 Teacher Questionnaire also show that many teachers rely primarily on low-complexity digital tools and rarely employ them in transformative ways, e.g. to provide interactive feedback or foster higher-order learning activities. This suggests that the full pedagogical potential of digital technologies is yet to be realised. To promote the adoption of more effective digital education practices, policy makers could strengthen the evidence base on effective digital pedagogies, promote school-based research and disseminate practical examples of what works. Professional learning, curriculum guidance as well as monitoring and evaluation frameworks could also be aligned to encourage teachers' evidence-based, purposeful use of digital resources and to generate stronger evidence on what works in different contexts. While PISA 2022 provides important initial evidence concerning the association between the use of digital resources and learning outcomes, future waves of PISA will provide a clearer picture of the progress that teachers are making in using technology to support their students.

Policy insight 3: Considering students' access to digital resources at home in the design of digital education policies and strategies

Students in many OECD countries spend a considerable amount of time learning before and after school or on weekends at home. As schools increasingly integrate digital resources into the teaching and learning process, students rely on access to high-quality digital infrastructure at home to complete digital assignments, engage in self-directed learning or participate in blended learning environments. Despite the expansion of broadband access and the proliferation of digital devices, socio-economic gaps remain. On average across OECD countries, almost 15% of students in the bottom socio-economic quartile reported not having access to a computer that they can use to learn at home, compared to 2% in the top quartile. Similar socio-economic gaps are observed in students' access to educational software and apps. Advantaged students also spend significantly more time learning with digital resources at home. PISA 2022 data confirms that students' access to digital infrastructure at home is strongly associated with their use of digital resources for learning both inside and outside of school. To enable all schools and students to harness the potential of digital tools, policy makers and school leaders therefore need to consider students home environment as part of their broader digital education strategies. This could include providing subsidised access to broadband services and computers to low-income families, central device-lending schemes, or requiring schools to ensure access to digital resources if needed. Digital inclusion strategies should also consider the role of assistive technologies and accessible design to support students with special education needs and diverse learning profiles. More granular data on the use of digital resources will be needed to assess how they are and can be used to support inclusive learning environments.

Policy insight 4: Identifying and addressing barriers that prevent some schools from using digital education resources effectively

Across OECD countries, socio-economically disadvantaged students tend to spend less time using digital resources for learning at school. This pattern is largely explained by school-level factors, such as the availability of digital devices, teachers' preparedness for digital learning and schools' digital policies, all of which can limit schools' ability to make effective use of digital education resources. If leftunidentified or unaddressed, these barriers risk exacerbating existing educational inequalities. Policy makers should therefore seek to identify these challenges where they arise and target support to schools that face structural disadvantages. Depending on the context, this may include such policy levers as earmarked funding to upgrade digital infrastructure, professional development for teachers in disadvantaged schools, targeted guidance or the deployment of support personnel. Although rural-urban gaps are modest across countries, some school systems also display pronounced regional divides and students in private schools tend to use digital resources more intensively for learning than those in public schools. In some cases, these differences are likely to reflect deliberate pedagogical choices rather than resource constraints or other barriers. Yet, it is important to remain attuned to widening gaps in schools' ability to benefit from the digital transformation and to address inequalities where they emerge. These gaps are not only material but also concern pedagogical practices, highlighting the need for targeted capacity building and professional learning alongside infrastructure investments.

Policy insight 5: Continuing to support teachers' professional learning on the effective use of digital resources

Teachers are central to the successful digital transformation of education and they have made remarkable progress in acquiring new technical and pedagogical skills to promote digital learning since the COVID-19 pandemic. Nevertheless, data from the PISA 2022 Teacher Questionnaire suggest that the use of digital resources remains a priority area for further training. It also shows that professional learning can make a difference for teachers' practice. Across the 18 countries and economies with available data, teachers who reported having recently engaged in professional development on the use of ICT are more likely to employ it in the classroom. Likewise, students make greater use of ICT at school if their principals are more confident in their teachers' preparedness for digital learning. These findings underline the need for policy makers to ensure teachers' access to high-quality training on the effective pedagogical use of digital resources. This could include actively incentivising and supporting relevant learning opportunities and recognising new formats of professional development (such as MOOCs or micro-credentials) (OECD, 2025[3]). It could also involve including digital skills in teachers' professional competency frameworks and recognising them for their formative evaluations or career advancement (OECD, 2025[40]). Aligning teachers' professional learning with curriculum and assessment frameworks can help to ensure its coherence and sustained impact. Further analyses should also explore how some countries have managed to prevent gender and age gaps to emerge in teachers' use of digital resources.

Policy insight 6: Promoting peer learning and knowledge sharing among teachers to take advantage of the diversity of digital practices within the profession

Data from the PISA 2022 Teacher Questionnaire reveal significant variation in the way teachers integrate digital resources in their practice, not only across but also within countries. Most teachers employ a relatively narrow range of digital tools on a regular basis and engage in a limited set of activities. However, a cluster analysis reveals that there are "highly engaged" teachers drawing on a wide range of digital tools in every country and economy. The analysis also shows that younger teachers and those teaching technology and vocational subjects are more likely to be intensive users of digital resources. This diversity of digital expertise within the profession provides opportunities for systems to scale up effective practices and build collective capacity for digital innovation through peer learning. Professional learning formats that promote knowledge sharing among teachers (including those of different subjects and levels of expertise in ICT) could help to disseminate effective digital practices, foster a sense of professional community and enhance teachers' confidence in experimenting with new digital tools. Policy makers should therefore consider strategies to promote such peer learning on digital education, for example though mentoring programmes, professional learning communities or school-based innovation hubs. Successful international examples of peer learning and collaboration could also be highlighted to demonstrate how diverse practices can be scaled (OECD, 2025131). Education authorities could support such initiatives, for example by recognising mentoring activities in career progression frameworks or by providing time and resources for collaboration.

Policy insight 7: Leveraging digital resources to alleviate teachers' workload and to help them make more effective use of their time

Besides their potential to enhance teaching and learning, digital technologies offer opportunities for teachers to make more effective use of their time. By supporting tasks like lesson preparation, marking or administrative duties, digital tools could alleviate teachers' workload and allow them to spend more of their time directly engaging with their students and providing them with tailored support. However, so far, PISA 2022 data provide little evidence that the use of digital tools has helped to reduce the time that teachers spend on administration or communication. Conversely, the data indicate that teachers who spend much of their working time inside the classroom may have fewer opportunities to explore using digital resources for teaching-related activities, underlining the difficult trade-offs that teachers face, given their limited time and competing priorities. Policy makers and school leaders should therefore continue to explore opportunities to increase schools' operational efficiency, streamline routine tasks and reduce teachers' administrative burden. This requires the careful co-design of tools with teachers to ensure that they address real needs and deliver genuine efficiency gains. This could also involve the integration of administrative platforms, enhancing the interoperability of digital tools or ensuring the secure exchange of data to facilitate processes like curriculum planning and performance tracking. Public authorities can also play an important role in facilitating collaboration between digital developers and the teaching profession to create tools that are user-friendly, aligned with teachers' needs and have a direct impact on their daily work (OECD, 202540). Improving the efficiency of teachers' non-instructional work through technology can create more time for teaching and student support, ultimately enhancing both educational quality and teachers' professional well-being.

Sidebar

OECD EDUCATION WORKING PAPERS SERIES

This work was approved and declassified by the Education Policy Committee and Group of National Experts on School Resources on 05/11/2025.

OECD Working Papers should not be reported as representing the official views of the OECD or of its member countries. The opinions expressed and arguments employed herein are those of the author(s).

Comments on Working Papers are welcome and may be sent to [email protected] or the Directorate for Education and Skills, OECD, 2 rue André-Pascal, 75775 Paris Cedex 16, France.

This document, as well as any data and map included herein, are without prejudice to the status of or sovereignty over any territory, to the delimitation of international frontiers and boundaries and to the name of any territory, city or area.

The statistical data for Israel are supplied by and under the responsibility of the relevant Israeli authorities. The use of such data by the OECD is without prejudice to the status of the Golan Heights, East Jerusalem and Israeli settlements in the West Bank under the terms of international law.

Attribution 4.0 International (CC BY 4.0)

This work is made available under the Creative Commons Attribution 4.0 International licence. By using this work, you accept to be bound by the terms of this licence (https://creativecommons.org/licenses/by/4.0/).

Attribution - you must cite the work.

Translations - you must cite the original work, identify changes to the original and add the following text: In the event of any discrepancy between the original work and the translation, only the text of original work should be considered valid.

Adaptations - you must cite the original work and add the following text: This is an adaptation of an original work by the OECD. The opinions expressed and arguments employed in this adaptation should not be reported as representing the official views of the OECD or of its Member countries.

Third-party material - the licence does not apply to third-party material in the work. If using such material, you are responsible for obtaining permission from the third party and for any claims of infringement.

You must not use the OECD logo, visual identity or cover image without express permission or suggest the OECD endorses your use of the work.

Any dispute arising under this licence shall be settled by arbitration in accordance with the Permanent Court of Arbitration (PCA) Arbitration Rules 2012. The seat of arbitration shall be Paris (France). The number of arbitrators shall be one.

References

References

Bers, M., A. Strawhacker and A. Sullivan (2022), "The state of the field of computational thinking in early childhood education", OECD Education Working Papers, No. 274, OECD Publishing, Paris, https://doi.org/10.1787/3354387a-en. [34]

Bettháuser, B., М. Siglreitmaier and A. Fasang (2024), "Change and inequality in German teenagers' typical time-use sequences between 1991 and 2013", Journal of Time Use Research 1, pp. 42-62, https://doi.org/10.32797/jtur-2024-3. [17]

Boeskens, L. and K. Meyer (2025), "Policies for the digital transformation of school education: Evidence from the Policy Survey on School Education in the Digital Age", OECD Education Working Papers, No. 328, OECD Publishing, Paris, https://doi.org/10.1787/464dab4d-en. [83]

Boeskens, L. and D. Nusche (2021), "Not enough hours in the day: Policies that shape teachers' use of time", OECD Education Working Papers, No. 245, OECD Publishing, Paris, https://doi.org/10.1787/15990b42-en. [38]

Bryant, J. et al. (2020), How artificial intelligence in education will impact K-12 teachers, McKinsey & Company, https://www.mckinsey.com/industries/public-sector/our-insights/howartificial-intelligence-will-impact-k-12-teachers#/ (accessed on 10 March 2025). [37]

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. [23]

Caliñski, T. and J. Harabasz (1974), "A dendrite method for cluster analysis", Communications in Statistics, Vol. 3/1, pp. 1-27, https://doi.org/10.1080/03610927408827 101. [41]

Donnelly, D., O. McGarr and J. O'Reilly (2011), "A framework for teachers' integration of ICT into their classroom practice", Computers & Education, Vol. 57/2, pp. 1469-1483, https://doi.org/10.1016/J.COMPEDU.2011.02.014. [27]

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. [15]

Escueta, M. et al. (2017), "Education Technology: An Evidence-Based Review", NBER Working Papers, No. 23744, National Bureau of Economic Research, Cambridge, MA, http://www. nber.org/papers/w23744 (accessed on 6 April 2022). [31]

European Commission, M. et al. (2021), Enhancing learning through digital tools and practices : how digital technology in compulsory education can help promote inclusion, Publications Office of the European Union, Luxembourg, https://doi.org/10.2766/365846. [39]

Forsstróm, 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. [2]

Forsstróm, 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. [12]

Gil-Flores, J., J. Rodríguez-Santero and J. Torres-Gordillo (2017), "Factors that explain the use of ICT in secondary-education classrooms: The role of teacher characteristics and school infrastructure", Computers in Human Behavior, Vol. 68, pp. 441-449, https://doi.org/10.1016/j.chb.2016.11.057. [19]

Graves, K. and A. Bowers (2018), "Toward a typology of technology-using teachers in the "New Digital Divide": A latent class analysis of the NCES Fast Response Survey System Teachers' Use of Educational Technology in U.S. Public Schools, 2009 (FRSS 95)", Teachers College Record, Vol. 120/8, https://doi.org/10.1177/016146811812000808. [24]

Lee, J. (2024), "Latent class profiles of teacher use of digital tools in PISA 2018 data", Education and Information Technologies, Vol. 29/10, pp. 12229-12261, https://doi.org/10.1007/s10639023-12305-3. [25]

Mama, M. and S. Hennessy (2013), "Developing a typology of teacher beliefs and practices concerning classroom use of ICT", Computers & Education, Vol. 68, pp. 380-387, https://doi.org/10.1016/J. COMPEDU.2013.05.022.[28]

McKnight, K. et al. (2016), "Teaching in a digital age: How educators use technology to improve student learning", Journal of Research on Technology in Education, Vol. 48/3, pp. 194-211, https://doi.org/10.1080/15391523.2016.1175856. [30]

OECD (2025), "Finite time to learn and play: Whole student development and students' digital leisure outside of school", OECD Education Policy Perspectives, No. 130, OECD Publishing, Paris. [18]

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. [3]

OECD (2025), Resourcing School Education: Policies for the Digital Transformation of Education and Future-Readiness of Teachers [website], https://www.oecd.org/en/about/projects/resourcing-school-education-for-the-digital-age. html! (accessed on 8 January 2025). [5]

OECD (2025), "Teacher policies to support the use of digital resources in the classroom", OECD Education Policy Perspectives, No. 128, OECD Publishing, Paris, https://doi.org/10.1787/d16e9614-en. [40]

OECD (2024), PISA 2022 Results (Volume lll): Creative Minds, Creative Schools, PISA, OECD Publishing, Paris, https://doi.org/10.1787/765ee8c2-en. [13]

OECD (2024), PISA 2022 Results (Volume V): Learning Strategies and Attitudes for Life, PISA, OECD Publishing, Paris, https://doi.org/10.1787/c2e44201-en. [14]

OECD (2024), PISA 2022 Technical Report, PISA, OECD Publishing, Paris, https://doi.org/10.1787/01820d6d-en. [10]

OECD (2023), PISA 2022 Database, https://www.oecd.org/en/data/datasets/pisa-2022- database.html (accessed on 29 January 2024). [16]

OECD (2023), "PISA 2022 ICT Framework", in PISA 2022 Assessment and Analytical Framework, OECD Publishing, Paris, https://doi.org/10.1787/9bd299c1-en. [22]

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. [11]

OECD (2023), PISA 2022 Results (Volume II): Learning During - and From - Disruption, PISA, OECD Publishing, Paris, https://doi.org/10.1787/a97db61c-en. [4]

OECD (2023), Shaping Digital Education: Enabling Factors for Quality, Equity and Efficiency, OECD Publishing, Paris, https://doi.org/10.1787/bac4dc9f-en. [1]

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/COMPUTERBASED%20STUDENT%20questionnaire%20PISA%202022.pdf (accessed on 21 February 2025). [6]

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). [8]

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.pd f (accessed on 21 February 2025). [7]

OECD (2021), Teacher Questionnaire for PISA 2022 (International Option): Main Survey Version, https://www.oecd.org/content/dam/oecd/en/data/datasets/pisa/pisa-2022- datasets/questionnaires/TEACHER%20QUESTIONNAIRE%20PISA%202022.pdf (accessed on 21 February 2025). [9]

OECD (2020), What Students Learn Matters: Towards a 21st Century Curriculum, OECD Publishing, Paris, https://doi.org/10.1787/d86d4d9a-en. [32]

OECD (2019), TALIS 2018 Results (Volume I): Teachers and School Leaders as Lifelong Learners, TALIS, OECD Publishing, Paris, https://doi.org/10.1787/1d0bc92a-en. [36]

OECD (2017), School choice and school vouchers: An OECD perspective, OECD Publishing, Paris, https://www.oecd.org/edu/School-choice-and-school-vouchers-an-OECD-perspective.pdf (accessed on 10 January 2018). [21]

Rogers, E. (1962), Diffusion of Innovations, Free Press of Glencoe, New York, NY. [26]

Russell, M. et al. (2003), "Examining teacher technology use: Implications for preservice and inservice teacher preparation", Journal of Teacher Education, Vol. 54/4, pp. 297-310, https://doi.org/10.1177/0022487 103255985. [29]

Starkey, L. (2020), "A review of research exploring teacher preparation for the digital age", Cambridge Journal of Education, Vol. 50/1, pp. 37-56, https://doi.org/10.1080/0305764X.2019.1625867. [35]

Tondeur, J. et al. (2012), "Preparing pre-service teachers to integrate technology in education: A synthesis of qualitative evidence", Computers & Education, Vol. 59/1, pp. 134-144, https://doi.org/10.1016/J.COMPEDU.2011. 10.009. [120]

Footnote

Notes

1 To collect international comparative evidence on digital school education policies, the project has administered the Policy Survey on School Education in the Digital Age, which was completed by 37 jurisdictions over the course of 2024. The survey's results were published in Boeskens & Meyer (20251337).

2 Since data for the PISA index of economic, social and cultural status (ESCS) are not available for Costa Rica, the country is not included in some Figures and Tables. Likewise, data on the daily hours that students use digital resources for different activities (Question ST326 of the Student Questionnaire) are missing for the United States.

3 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 (202410), https://stat.link/v6ug1n.

4 Students were asked to think of different kinds of digital resources, such as desktop computers, laptops and tablets as well as educational software and other digital learning tools (OECD, 2021).

5 Socio-economically disadvantaged/advantaged students are defined as those in the bottom/top quarter of their country's distribution of the PISA index of economic, social and cultural status (ESCS) - a composite score derived from three indicators related to students' family background: parents' highest education in years, parents' highest occupational status, and home possessions.

6 The ESCS index, like other PISA scaled indices, has an average of zero and a standard deviation of one across OECD countries. Moving from 0 to 1 in the ESCS index, for instance, means comparing a student With a socio-economic status close to the OECD average with one close to the 84!" percentile, and moving from 1 to 2 in the ESCS index means comparing a student around the 84!" socio-economic percentile with the one around the 98" percentile.

7 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[10]), https://stat.link/v6uq1n.

8 The index of students' self-directed learning self-efficacy (SDLEFF) was constructed based on students' scaled ratings of how confident they felt about having to do the following eight self-directed learning tasks should their school building close again in the future (based on question ST335 of the Student Questionnaire): Using a learning management system or school learning platform (e.g. Blackboard®, Edmodo®, Moodle®, Google® Classroom™); Using a video communication programme (e.g. Zoom™, Skype™, Google® Meet™, Microsoft® Teams); Finding learning resources online on my own; Planning when to do school work on my own; Motivating myself to do school work; Focussing on school work without reminders; Completing school work independently; Assessing my progress with learning. Note that this scale used a within-construct matrix sampling design. Each of the eight items included in this scale had four response options ("Not at all confident", "Not very confident", "Confident", "Very confident"). For more details, see Table 19.A63 in OECD (2024[10]), https://stat.link/v6uq1n.

9 Authors' analysis based on PISA 2022 data, not included in this working paper.

10 It should be noted that the within-school variation in students' responses also captures differences in students' perceptions and individual response styles (even among students attending the same classes) and "noise" arising from unsystematic mistakes. However, the share of between-school variation is significantly lower when students are asked how much time they spend using digital devices for learning before and after school rather than for learning at school. This is in line with the hypothesis that schools exert less influence over students' learning at home and strengthens our confidence that we do measure the influence of schools (analysis not included in this working paper).

11 To test the robustness of the results, clusters were calculated using different seed numbers, keeping the lowest seed number that provided results without extreme values (i.e. 0% or 100%) on any of the items. Results without extreme values were similar, while those with extreme values differed drastically and seemingly at random. The Calinski-Harabasz statistic was calculated to avoid clusters with low internal homogeneity but was not used as the main criterion of selection (Caliński and Harabasz, 1974[41]).

12 The PISA 2022 Teacher Questionnaire was administered in the following 18 countries and economies: Australia, Baku (Azerbaijan), Brazil, Colombia, Costa Rica, the Dominican Republic, Georgia, Germany, Hong Kong (China), Korea, Kosovo, Macao (China), Malaysia, Morocco, Panama, Peru, Portugal and the United Arab Emirates.

13 In the PISA 2022 Teacher Questionnaire, "digital resources" referred to digital devices or hardware (e.g. computers, tablets, smartphones, 3D printers), software (e.g. programmes, apps, communications tools, educational learning tools), and online resources (e.g. websites, web portals) (OECD, 2021[9]).

14 The index of teachers' emphasis on ICT competencies (ICTCOMP) was constructed based on teachers' scaled ratings of how much emphasis they reported placing on the following five ICT competencies (based on question TC221 of the Teacher Questionnaire): Evaluating the credibility of digital information; Sharing digital information with others (e.g. through presentations, documents, images or diagrams); Using [digital resources] to solve real-world problems (e.g. measuring the height of a building, finding directions on а map); Using digital tools to work collaboratively; Understanding the consequences of public dissemination of information online (privacy and safety). Each of the five items included in this scale had four response options ("No emphasis", "Little emphasis", "Some emphasis", "A lot of emphasis"). For more details, see Table 19.A178 in OECD (202410), https://stat.link/v6ug1n.

15 The index of teacher use of ICT (TCDIGRES) was constructed based on teachers' scaled frequency ratings of how often they use digital resources for the following nine teaching tasks (based on question TC220 of the Teacher Questionnaire): to design tasks; to explore new teaching methods; to enable student collaboration; to provide feedback to students; to provide access to instructional material for students who cannot physically attend class; to communicate with parents or guardians; to assess students' learning; to share ideas or resources with colleagues; to take part in professional communities of practice online. Each of the nine 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.A177 in OECD (2024110), https://stat.link/v6ug1n.

16 Authors' analysis based on PISA 2022 data, not included in this working paper.

17 Idem.

18 Idem.

19 Idem.

Word count: 21648

Show less

Using digital resources for learning: Policy insights from PISA 2022

Content area

Abstract

Full text