PROTOCOL: A systematic review of mobile device use in the primary school classroom and its impact on pupil literacy and numeracy attainment

Children today are born into a world surrounded by technology. For this “i‐Generation” (Twenge, 2017), most aspects of daily life can be conducted online, including socialising, shopping, learning and engaging with the world around them, and the lines between the online and offline life are becoming increasingly blurred. OFCOM, (2019) reports that 82% of 5–7‐year‐olds go online regularly, averaging 9.5 h per week, while for 12–15‐year olds, 99% go online regularly, averaging 20.5 h per week. This increasingly technological world has influenced educational practice, with the use of technology in classrooms evolving from desktop computers, to interactive whiteboards, to the more recent use of tablets and other mobile devices to deliver the school curriculum. A recent development (UNCRC, 2021) has seen the Convention on the Rights of the Child adopt General Comment 25 which recognises children's rights in the digital world. This includes children's right to the educational benefits that technology can bring, and places responsibilities on States to ensure schools are equipped with the infrastructure, knowledge and skills to support this. Access to technology in the classroom is increasing, however remains disparate across the globe. OECD (2015) reported that in 2012, 72% of 15‐year‐olds in OECD countries use a computer, laptop or tablet at school (unchanged from 2009), although usage in some countries was as low as 50%. The latest OECD (2020) figures report a global average of almost one computer for every 15‐year‐old pupil, with variation between 1.25 computers per pupil in countries such as UK or United States, and 0.25 computers per pupil in countries such as Brazil or Greece. While these figures do not reflect practice in primary‐aged schools, they give some idea of the scale of usage and the potential impact for children and young people around the world. For this reason, the role and impact of technology in the classroom has been a growing area of interest for researchers, spanning topics including the impact on educational outcomes (such as reading, numeracy attainment and critical thinking) (e.g., Bebell & Pedulla, 2015); classroom interaction and pupil motivation to learn (e.g., Campbell & Jane, 2012; Ciampa, 2014) and teacher skills and attitudes to technology (e.g., Ciampa & Gallagher, 2013; Lincoln & Barney, 2017). As technology has advanced, so too has the range of classroom activities it can be used for. Technology can be used to enhance and support teacher presentation, or can directly engage each child with an individual device or in groups. The integration of technology with pedagogical approaches is therefore of critical importance. Baskerville (2012); Grieffenhagen (2004); Murcia (2014), and others, have considered the ways in which technology can be used to enhance and indeed transform pedagogy, rather than to deliver the same activities with new tools (e.g., a teacher writing a question on a blackboard versus writing it on an interactive whiteboard—the activity has not changed, just the means by which it is presented). The most recent technological advances, specifically mobile devices, provide an unprecedented opportunity for transformative pedagogy. While tablet‐style computers have been available since the early 2000s, the introduction of the user‐friendly and innovative Apple iPad in 2010, closely followed by similar tablets of other brands, made these tools more accessible to the wider population.

Children are particularly skilled in the use of mobile devices from an early age; OFCOM (2019) report that 58% of 3-4-year olds, and 76% of 5-15-year olds use a tablet device regularly, while just under half of 5-11-year olds own one of their own. Additionally, while only 35% of 7-11-year olds own a smartphone, this figure has been steadily rising (from 24% in 2015). Research by the British Educational Suppliers Association (BESA, 2015) found that 71% of UK primary schools surveyed reported using tablets in the classroom.
This was a significant increase from the previous year, with figures predicted to rise significantly by 2020. More recently, and as individual ownership of tablets and indeed smartphones rises, "bring your own device" policies are beginning to be seen across schools to harness the resource now available in pupils' own pockets. The intuitive nature of tablets and smartphones, coupled with affordability and the potential to "bring your own", make them ideally placed to influence traditional teaching methods. The NMC Horizon Report K-12 Edition (Johnson et al., 2014) identified such "intuitive technology" as having the potential to significantly impact educational practice over the next 5 years. Children now learn through gameplaying, use the endless body of information available to research topics of interest, and are adept at recording, editing and presenting videos as part of the ordinary school day.
Literacy and numeracy are a central focus of primary education, and not only provide the tools through which a child can engage in wider curricular subjects, but also have far-reaching application across the life-course. Indeed, research by the National Literacy Trust (Clark & Teravainen-Goff, 2018) shows that children who are more engaged with literacy have better mental wellbeing. Yet the National Literacy Trust website also reports that 16.4% of adults in England,26.7% in Scotland,17.9% in Northern Ireland and 12% in Wales have "very poor literacy skills", while OECD (2016) finds that UK young adults (age 16-24) have lower basic literacy and numeracy skills than young adults in many other countries. The 2018 round of PISA tests (OECD, 2019), designed to assess reading, science and maths skills of 15 year-olds globally, show the UK moving up the rankings in maths (18th, up from 27th in 2016) and in reading (14th, up from 22nd in 2016), yet still falling below many other countries. It is clear that a focus on literacy and numeracy must be a priority for primary age children.
Mobile devices are one way in which teachers in primary schools are enhancing literacy and numeracy education, and there is a general feeling amongst teachers that this is a positive development. The National Literacy Trust (Picton, 2019) surveyed 219 teachers across the UK, and found that just under 60% believe technology can help pupils to overcome learning barriers, and over three quarters feel technology should be available for pupils right across the curriculum to support their literacy development.
However despite this belief, trends in technology access and training do not follow. Just under half of the sample of teachers stated that their pupils had access to technology (either laptops or tablets) in the classroom, around 20% of teachers said they never use technology to support literacy, and a quarter reported never having had training to make use of technology to support literacy.
Similarly, BESA, 2015 reported that 34% of schools felt their technology implementation was poor, citing an ineffective infrastructure and lack of adequate training and support as barriers.
Given the rapid increase in use of mobile devices in the classroom, it is critical that teachers are adequately equipped with the skills, resources and guidance to enable them to effectively and safely embed such mobile technology within pedagogy. Ultimately, this technology must benefit pupils and support learning. The OECD (2015) report also found that those countries reporting heavy investment in technology in schools demonstrated no significant improvement in reading, writing or maths.
While mobile device usage in the classroom continues to grow, its application therefore remains an area of uncertainty, particularly in terms of the impact on pupils' educational experience, and, critically, learning outcomes and attainment. It is important that we can identify what works, for whom, and why. For this reason, this systematic review of existing research on mobile devices in the classroom (specifically tablet devices, smartphones and handheld games consoles), and their impact on attainment, is timely, with an important contribution to make towards ensuring that future developments across educational policy and practice, and in the professional development of educational practitioners, are informed by evidence of good practice.

| Description of the intervention
There are three key elements for consideration in how mobile technology is used in the primary school classroom: 1. Devices 2. Activities

Outcomes
A logic model has been developed to demonstrate how these key components interact (see Figure 1).

| Devices
The most common mobile devices used in schools are tablet computers. On first introduction, Apple was the device of choice in US schools, reporting a 95% education market share (Apple, 2013), and while Apple remains the market leader in tablet sales with 36.5% of sales (International Data Corporation, 2020), advances from other brands using Android or Chrome operating systems, and producing less expensive devices, have seen Apple market share in education drop (Futuresource Consulting, 2019). Smartphones are less commonly used in schools; indeed, there is ongoing discussion in the media as to whether children should be permitted to bring such devices to school. In late 2018, the French government passed legislation to ban children and young people (up to age 15) from using their phone in the school grounds during the school day. This move has prompted much debate; those in agreement with this approach feel it addresses concerns such as cyber-bullying and distraction from studies. Research from the London School of Economics is commonly cited; Beland and Murphy (2015), reviewed exam results following mobile phone bans in schools across England and found not only an improvement in results, but a more significant improvement for those pupils from disadvantaged backgrounds, therefore contributing to a reduction in educational inequalities. Yet the increase in children's ownership of mobile devices undoubtedly presents an opportunity for schools. As ownership of smartphones and tablets increases among children, schools are now considering how they can harness the opportunities presented by children having an Internet-enabled device in their own pockets (e.g., Rae et al., 2017). Such "Bring your Own Device" approaches also bring challenges, with online safety, appropriate behaviour policies, infrastructure capacity and ensuring equality of access only some of the necessary considerations.

Definition of mobile device
For the purposes of this systematic review, the focus will be on mobile devices. These are defined as handheld computing devices, including: • tablet computers of varying sizes (iPad and other brands) • smartphones (defined as those with a touch screen interface which can connect to the Internet) • small handheld games consoles, such as a Nintendo Switch or Nintendo 2DS (again, with a touch screen or integrated buttons and can connect to the Internet or have games loaded).
Within the classroom, the mobile devices will either directly access the Internet, or make use of device applications ("apps") or inbuilt device functions. While overall, the functionality of these different types of mobile devices will be similar, the screen size of a tablet is likely to be larger than on smartphones or games consoles. Tablets are typically anything between 7 and 10 inches, with some of the more professional models measuring up to 12 inches, while smartphones range in size from 3 to 6.5 inches. There is some evidence that screen size may impact effectiveness, depending on usage. Alghamdi et al. (2014) found that it took longer to read the same information on a small screen than a larger one, however there was no difference in information understanding or retention, while Albó et al. (2019) found there was less opportunity for pupils to collaborate on a small smartphone when used in class, as compared to a larger tablet.

| Activities
Mobile devices have applicability right across the curriculum, and can be used in any number of ways, both individually by pupils or in groups. There are also many ways in which the teacher can use F I G U R E 1 Logic model representing implementation of mobile technology in the classroom DORRIS ET AL. | 3 of 30 mobile devices to support his or her teaching, however these will not be included within the scope of this review. From a brief review of relevant publications, a summary of potential usage and types of activities is included below. Puentedura's SAMR model (Substitution, Augmentation,

Modification, Redefinition)
In any discussion on the impact of mobile device usage in the classroom, the activities undertaken, and the context in which they occur, rather than the device itself, are the important factors. Given the breadth of potential activities, researchers and practitioners have sought a framework to help classify and therefore compare activities.
One such model is the SAMR model, developed by Puentedura (2006) (Figure 2). The SAMR model compares activities undertaken using technology with the everyday activities they are replacing (e.g., reading an e-book rather than a paperback), and asks what the use of technology has added to the learning experience.
Puentedura proposes four levels of activity: • Substitution: Technology is used in substitution for the usual classroom tools or activities, for example, reading an online textbook rather than a paper copy.
• Augmentation: Technology substitutes for the usual tools however also improves function slightly, for example, a computer word processor used to write an assignment, therefore allowing for inclusion of pictures or diagrams.
These are both considered to enhance pedagogy; • Modification: Technology allows for an activity to be undertaken significantly differently, for example, accessing the Internet to independently research content for an assignment.
• Redefinition: Technology allows for new, previously unachievable activities to be undertaken, for example, creating a multimedia assignment using video, audio and other creative tools.
These are both considered to transform pedagogy. Puentedura (2013) emphasises that in practical terms, the model should be seen as a spectrum along which classroom activities sit, and proposes that for true transformation of learning, activities using technology should aspire towards redefinition of pedagogy, rather than simply substituting one activity for another. Geer et al. (2017) also note that as mobile devices were not primarily designed as tools for an educational setting, effort will be needed to adapt them to and embed them within existing peda-

| How the intervention might work
The focus of this review is on the specific use of mobile devices in the primary/elementary classroom. Children using the devices will therefore be aged between four and eleven. Given the broad age range, there will be a wide variety in the types of interventions, the complexity and skill needed, and the aims and potential outcomes.
The use of the SAMR model to classify the types of activities that might be undertaken and what they can add to existing pedagogy has already been discussed. There have also been a number of theories proposed which can help us to understand how digital interventions might contribute to positive learning outcomes.
Some of these are discussed briefly below.
• Play-based learning: Play has a central role in early years and primary education curricula, with a widely established body of research showing the effectiveness of play in learning, from free play through to instructive games. Digital interventions can make learning more fun, creating a positive attitude towards a subject, encouraging creativity (e.g., Livingstone, 2012) and generally enhancing enjoyment of learning (Oliemat et al., 2018).
• Supporting agency and self-directed learning: Geer et al.
(2017) report that teachers found tablets to contribute positively to student-centred learning, with pupils more in control of their learning than traditional teaching could have allowed.
This allows students to learn at their own pace, leaving teachers free to provide one to one support where needed and allowing more advanced pupils to move on to more challenging materials.
• Increasing motivation: a common focus of research is the role of mobile devices in increasing motivation to learn, which can give pupils a more positive attitude to school work, help them engage more actively in lessons and in turn contribute positively to academic achievement (Ciampa, 2014;Clarke & Abbott, 2015). Tasks associated with increased motivation should provide autonomy, be challenging without being impossible, stimulate the senses, build curiosity, and provide an element of competition, with others or one self (Malone & Lepper, 1987). The use of mobile devices provides opportunity across all these elements.
• Providing opportunity for formative assessment: The use of digital devices in the classroom provides additional opportunities for teachers to review and assess pupil progress in real time, and offer feedback on the spot which can support learning (Dalby & Swan, 2019). Many digital interventions are designed with an immediate feedback function which allows the child to see their mistakes and learn from them.
While the content of interventions will vary, the activities (as classified through the SAMR framework) and the characteristics highlighted above, are key factors in how interventions might work, and therefore the potential impact they can have on child outcomes.

| Examples of interventions
Some examples of the widely available interventions aimed at primary age children are detailed below.

| Why it is important to do this review
The impact that mobile devices actually have on educational outcomes remains unclear. Regardless, investment in tablets for use in the primary school classroom is increasing year on year (BESA, 2015). Given that funding cuts due to wider austerity are commonplace in education as in other areas of life, it is critical that investment is made in the most effective tools and approaches to best support educational outcomes. Educators must therefore be able to access the most up to date evidence to support decision making. A systematic review of existing literature will provide one such accessible resource. The review team has elected to focus on primary education, rather than the full spectrum of educational experience. Primary and post-primary education, and the use of technology within these, are very different, in terms of the subjects studied, the approach to pedagogy, and activities undertaken. As a team with limited resources, and a desire to produce a review which has the potential to influence practice, we would rather invest in the detail of primary education rather than try to cover both and risk dilution of the discussion.

| Existing literature reviews
Within the systematic review discipline, a number of "Coordinating bodies" or "brokerage agencies" (Sundberg, 2009, in Levinson & Prøitz, 2017 have emerged, of which the Campbell Collaboration is one. The primary role of such bodies is to support the review process and provide quality assurance through peer review, so that practitioners, policy makers and others seeking the best available evidence can trust in the robustness of the review. There is currently no existing registered systematic review on this specific topic, therefore the team feel it will be an important addition to the robust evidence base. There are some meta-analytic studies or literature reviews on a similar theme; these are discussed below. Haßler et al. (2015) review: "Tablet use in schools: a critical review of the evidence for learning outcomes" is the most similar to the proposed review, however a number of key considerations mean that there remains value in the proposed review: • This is not a registered systematic review.

Maths
Motion Math is an instructional app designed to be accessed via a mobile device, although can also be accessed via a desktop computer. The Motion Math model includes hundreds of levels of mathematics content, aimed at children aged approximately 4-11 and covering general arithmetic concepts aligned to the school curriculum, such as fractions, addition and subtraction, and percentages. Pupils can access the app via their own or school device, log-in and work through the games and activities at their own pace. Motion Math uses the "tilt" facility on mobile devices so that children physically manipulate the device to engage with activities, for example directing a falling star to the right place. The game facilitates formative assessment, through tracking performance, providing direct feedback to the child, supplying hints and tips if answers are incorrect, and increasing difficulty when answers are correct. Students can work at their own pace through the activities, and teachers receive feedback on pupil usage and performance. A class password is needed to log in.

Top Marks Maths and literacy
A free website with a range of maths and literacy games for 3-14 year olds. As it is web-based, it can be accessed via a tablet, smartphone or other Internet-enabled device, or the app downloaded. Games vary in difficulty, and are broken down into age groupings. Each game provides immediate feedback so the player can try again if they get a question wrong. The website also includes resources for teachers to use as teaching tools via an interactive whiteboard, and can be accessed at home to continue learning for homework.

Maths and literacy
A free website developed and maintained by the BBC (therefore UK-specific). The site can be accessed at any time via tablet or smartphone, as well as traditional computer, so can be used at home or school. Specific resources are aligned to the curricula across the four nations. A wide range of resources and subject areas are covered for primary, secondary and post-16 pupils, and combine games, videos and instruction. For the younger users, activities are fun and game based, while for older users, activities make use of "real life" examples.

Mathletics Maths
Mathletics is a learning platform designed for use in schools and aligned to the UK primary school curriculum, however can also be used at home. Activities can be accessed via tablet or desktop computer, and include a range of tutorials and interactive games. There is a test option available, and pupil activities are marked automatically with detailed reports provided for the teacher. There is also a facility to assign homework. Activities incorporate challenges to motivate individual pupils, with points awarded for completion. Mathletics also includes scheduling and customisation facilities for teachers to support planning.
growth and evolution of the use of mobile technology in the classroom.
Initial literature searches have found a wealth of research since then; the proposed review would draw on this up to date research.
• Haßler et al.'s review considered both primary and secondary school use, did not include smartphones, and focused on wider learning outcomes. There is no subgroup analysis completed, either across age groups or specific learning outcomes, therefore the team feels our proposed review will expand upon Haßler et al.'s findings to better understand primary use specifically. This is particularly important as interventions may impact one group of pupils differently than others, for example younger pupils versus older pupils, or may be effective in maths interventions but not in science. To enable evidence-informed practice, these details are critical.
There is a current protocol registered with the Campbell Collaboration on a similar theme (title: Free Provision of Information and Communications Technology (ICT) for improving academic achievement and school engagement in students aged 4-18: A systematic review, Liabo et al., 2016). This registered protocol focuses on the impact on academic achievement (including literacy, numeracy and wider knowledge) and on school engagement (as measured by attendance patterns and school enjoyment), of schemes seeking to increase pupils' wider access to ICT, such as discounted laptop schemes or facilitating home Internet access. This ICT is not necessarily for use within the classroom, rather may be used at home or within the community.
A full list of further reviews identified is included in Appendix B along with details of their area of focus and differences in relation to this proposal. This is not a criticism of limitations of existing reviews, rather an attempt to reflect the ways in which the proposed review will add to existing work.
In summary, existing reviews tend to be: • Focused on older or younger age groups of children (pre-school, post primary, higher education) without subgroup analysis on our area of interest, therefore while we can get a sense of overall impact of devices, we cannot fully understand how the intervention specifically impacts primary aged children.
• Focused specifically on pupils requiring additional support or with special needs, rather than general usage in the classroom.
• Inclusive of all technologies (including interactive white boards, desktop computers, etc.) rather than focused specifically on mobile devices, or more narrowly focused (e.g., iPad branded tablets only).
• Outdated, particularly given the topic of interest and the rapid evolution of technology applicability in the classroom.
• Reviews or mapping of the types of research carried out in this field, rather than an analysis of the effects of mobile learning interventions.
• Outside of the standards set out by Campbell in terms of systematic review methodology (for example including peer reviewed journals only, or not including grey or unpublished literature).
It is therefore our opinion that a systematic review of research on the specific impact of mobile devices in the primary classroom on literacy and numeracy achievement will build on existing reviews and is of merit and indeed timely.

| Policy relevance
Any innovation in the classroom has the capacity to impact all children and young people to a greater or lesser extent; it is therefore critical that educators are equipped with the skills and knowledge to use emerging technology appropriately and effectively to best support pupil attainment. The proposed review has important policy and practice implications across a number of areas, including: • Curriculum development and delivery; • Technical provision in schools; • Teacher training and ongoing professional development; • Online safety.

| OBJECTIVES
Specifically, the review will aim to answer the following question: • What is the effect of mobile device integration in the primary school classroom on children's literacy and numeracy attainment outcomes?
This systematic review will focus on the use of tablets (including iPads and other branded devices), smartphones (defined as those with a touchscreen interface and which can connect to the Internet) and handheld games consoles (again with touchscreen interface and Internet connectivity) within the primary school classroom, aimed at improving literacy and/or numeracy for children aged 4-11. The primary objective of this systematic review will be to identify and synthesise high quality research (published and unpublished) to determine the impact of mobile devices in the primary classroom on literacy and numeracy attainment outcomes. Any activities taking place using mobile devices, including apps and Internet sites accessed through them, will be included within the scope of the review This review aims to support policy makers and practitioners, working in the primary education sector, to make informed decisions about the use of mobile devices in the classroom. The review will therefore seek to address a number of issues within this • Are there specific devices which are more effective in supporting literacy and numeracy? (tablets, smartphones or handheld games consoles) • Are there specific activities which are more effective in supporting literacy and numeracy? (aligned to the 4 stages of the SAMR framework-substitution, augmentation, modification, redefinition) • Do any moderator variables impact on the effectiveness of mobile devices in supporting literacy and numeracy? (specifically gender of user, or intensity of intervention) Findings from the review will be used to highlight implications for policymakers and practitioners to support their evidenceinformed decision making in relation to the use of mobile devices in primary education. This review also seeks to identify areas where further research is needed in this regard, to ensure that children are getting the best possible support in developing strong literacy and numeracy skills.

| Stakeholder engagement
This systematic review will seek to draw available evidence from across the globe, and consider implications for policy and practice which might be relevant to all. Chapter 2 of the Cochrane Handbook  highlights the importance of stakeholder engagement throughout the review process, from defining priority topic and review questions through to interpreting review findings in relation to everyday practice. A participatory approach has therefore been incorporated to ensure stakeholder engagement throughout the review process. The involvement of stakeholders in the systematic review process is a less common approach, however the review team feels it is an important inclusion. Cottrell et al. (2015) reviewed studies which considered stakeholder engagement in the systematic review process, to identify benefits and challenges. While the study sample largely reflected medical reviews, the findings are relevant. Benefits and challenges identified include: • Increased credibility • Ability to anticipate controversy • Transparency and accountability • Improved relevance • Enhanced quality • Increased opportunity for dissemination and uptake of findings Challenges identified include: • Time required to engage stakeholders • Training and resources needed • Engaging appropriate people • Balancing multiple inputs • Understanding when and how to engage the stakeholders in the process The lead reviewer has significant experience of engaging stakeholders in various research and policy development processes, and will use this experience to enhance the review process through the involvement of a small Advisory Group. This group will help to ensure that the review itself, including scope, research questions and interpretation of findings, reflects everyday practice as far possible. The group has already been established, and includes four members representing: • Primary school teachers (two members, both ICT Coordinators within their schools) • Education Authority (with significant previous experience at the Council for Curriculum, Examinations and Assessment with responsibility for primary school ICT) • 1 × Parent (also with significant experience in participatory work with children and young people within a voluntary sector organisation) To date, the group has been engaged in refining the methodology to ensure this systematic review reflects the current needs & experiences of primary school practitioners as they embed mobile technology in their classrooms. A number of face to face, telephone and email discussions have taken place to inform the direction of the protocol. Initial proposals for the review focused on general technology use in the classroom, including whiteboards, desktop and laptop computers. However, the Advisory Group members advised that while traditional technology has been a feature of classrooms for many years, mobile devices are rapidly becoming a priority for primary schools and bring the potential for more exciting and transformative practice. In their own schools, they have seen significant investment in tablet computers, yet they also noted that the knowledge of colleagues on how to make best use of such devices was limited. The group also discussed "bring your own device" approaches in schools, and while this is not yet commonplace, they felt this was an emerging area of discussion. While mobile devices are being used right across the primary curriculum, literacy and numeracy are areas of commonality across schools, and a significant part of the primary curriculum.
The stakeholder group therefore felt that a systematic review of the evidence on mobile device usage to support literacy and numeracy would be of benefit to them and a wide range of colleagues.
Throughout the review, the group will be engaged further to: • Support the identification of relevant research studies.
• Highlight any relevant groups, activities or events, research, policy or practice developments which may be of interest.
• Support the interpretation of findings of the review in a userfriendly way which can be understood by a wide range of stakeholder audiences for whom the findings may be relevant • Support the dissemination of findings to appropriate audiences following completion.
As a voluntary group of expert advisors, time commitment and engagement will be kept to a minimum. Engagement will take place via email where possible, with telephone or face to face meeting only where essential. RCTs in educational research and found that their use has increased significantly and their applicability has been demonstrated. We will therefore seek only studies which report effect sizes for the comparison of an intervention and control group or groups through RCTs (or cluster RCTs). Comparison interventions will include either: • Traditional teaching methods which do not incorporate technology (no intervention) • An alternative technology (e.g., desktop computers) The review will not consider qualitative studies.

| Types of participants
The target population will be children within mainstream primary/ elementary education settings. Evidence from all countries will be included as long as it meets the wider search criteria. These children will usually be in the age range four to eleven, however on occasion may include children aged twelve.
There may be cases in which both primary and post-primary aged pupils have been included within a study. In such cases, attempts will be made to isolate the data relating to primary aged children only. If this is not obvious, contact will be made with the author/s to request relevant data; where this is not available, the study will not be excluded, rather will be recorded as having no appropriate data.
Only studies taking place in mainstream schools will be considered. Those which consider use of mobile devices in special schools, educational provision other than at school, or indeed home schooling, will not be included. Additionally, studies which focus on interventions to provide additional support to low-performing students, rather than the class as a whole, will not be included.

| Types of interventions
As already discussed, this systematic review will consider interventions within the classroom in which mobile devices are used to support pupil literacy and numeracy development. Included in the review will be: • Any intervention or activity within the primary school classroom (with children aged 4-11) that makes use of mobile devices (as defined above) to intentionally support learning in either literacy or numeracy attainment.
• Interventions which are a one-off or regular activity (however dosage will be taken into account when comparing studies at analysis stage).
• Interventions which engage the class as a whole.
• Interventions where pupils directly use the device, either individually or in pairs or groups.
• Studies focused on specific applications or websites which are accessed through the mobile device will also be included as long as the conditions above are still met.
Excluded will be: • Interventions which use other technology (such as desktop or laptop computers) rather than mobile devices as specifically defined.
• Interventions which use mobile devices but have no specific focus on literacy or numeracy.
• Interventions where the teacher uses the mobile device to support their own teaching delivery, but pupils have no direct engagement with the device.
• Interventions which do not take place as part of core curriculum delivery (e.g., where pupils take part in activities during free time).
• Interventions which do not take place within mainstream classroom (for example where homework is set via activities on mobile devices, or where the device is used in an after school group).
• Interventions targeted at children with learning difficulties or delays.

Primary outcomes
This review will focus on primary outcomes, that is, interventions which have a direct impact on children's academic achievement in literacy and numeracy. A wide range of measurement methods are employed to assess outcomes across literacy and numeracy, including both standardised, national assessments and bespoke tools, usually quantitative but sometimes observational in nature.
The core academic assessment of the country (such as key stage 1 or 2 in the UK or Children's Progress Academic Assessment [CPAA] in the United States) is also commonly used where scores are allocated to individual pupils. For the purposes of this systematic review, studies which focus on improvement in any element of literacy or numeracy will be considered for inclusion.
However, specific outcome measures will not be used as a criteria for study inclusion or exclusion.
Some common elements of literacy and numeracy have already been discussed. These will be further distilled in developing search strings for information retrieval.

Secondary outcomes
While the literature on technology integration in primary school classrooms considers a wide range of secondary outcomes, including enhanced motivation and engagement with peers (e.g., Ciampa, 2014), this systematic review will focus only on primary outcomes relating to academic attainment.

| Search methods for identification of studies
The following section describes the proposed methodology for identifying studies for inclusion. The methodology is designed to minimise the risk of publication bias, ensuring that all relevant studies are captured.

Search limitations
Year of publication: while the first commercially available touch-screen smartphones were introduced in the mid 2000s, and tablets followed towards the end of the decade, there were a number of Personal Digital Assistants available in the 1990s, for example the Delaware Fingerworks devices (later bought over by Apple), Palm Pilot or the Apple eMate, which may have been used in education. It is important that any research on these early devices is captured in the searches to enable a reflection on advancement in technology and approaches, therefore it is reasonable to limit searches to1990 onwards, given that the technology of interest was not available before then.
Language: Language will not be used as a search limiter, as this would leave the review open to bias. Effort will be made to obtain an English translation of any study thought to have relevance, using available online translation tools or contact with the author/s. Where a translation cannot be obtained, the study will be included as "awaiting classification" and the potential for bias in this regard will be reflected upon in the final report.
Geography: No limitations will be placed on geographical location of the study.
Search terms. Search terms will be grouped as follows: The search strategy has been designed to deliver a more comprehensive, or "sensitive", search. While this increases the risk of irrelevant studies being identified, these will be screened out through abstract review (discussed in the next section). Search strings will be combined as follows: (1 OR 2) AND (3a AND (3bi OR 3bii)) AND 4. An initial list of search terms within each grouping has been drawn up using a combination of keywords (Table 1)

Search sources
The search will incorporate relevant journal and other databases, accessed through the QUB library, with a particular focus on education. As recommended by Campbell Method Guide 1: Searching for studies (Kugley et al., 2016), both field-specific and multi-disciplinary databases will be searched. We aim to retrieve published and unpublished studies, journal and non-journal studies (including NGO and government research), conference papers and reports on proceedings, technical reports, dissertations and theses, white papers and other relevant literature. The following electronic databases will be prioritised: Journal databases (and interface through which they will be accessed-if applicable): • British Education Index (EBSCOhost) • Child Development & Adolescent Studies (EBSCOhost) • Cochrane Central Register of Controlled Trials • Directory of Open Access Journals (DOAJ) • Education Abstracts (EBSCOhost) • ERIC (Education Resources Information Centre) (EBSCOhost) • International Bibliography of the Social Sciences (IBSS) (ProQuest) • Education Journals (ProQuest) • PsychInfo (OVID) • Scientific Electronic Library Online (SciELO) • Scopus • Social Science Citation Index (Web of Science) Review databases:

• Campbell Collaboration
• Cochrane Database of Systematic Reviews
Google activity controls will be used to turn off search history, location services and other personalisation options to ensure this does not impact search results.
• Microsoft Academic Search

• ProQuest Dissertation and Theses
• Government websites (limited to those available in English) • European Documentation Centre Again, search statements will be modified to suit the source; advanced search options will be used where available.

| Searching other resources
Alongside the main electronic searches, a number of other activities will take place to ensure inclusion of all eligible studies.
• Contact will be made with authors prominent in the subject area (first and second authors of included studies, plus any others who have appeared regularly in excluded but relevant studies) to identify any unpublished studies or work in progress, either of their own or known to them. • Alongside any conference proceedings identified through the grey literature searches above, authors have identified the following conference/s as being highly relevant: International Society for Technology in Education; BETT; British Educational Research Conference and the European Conference on Education. These have been selected given their global reach, relevance to primary education and technology, and focus on research and pedagogy rather than marketing opportunities for technological products. The conference proceedings from the past 5 years will be searched by hand to identify those not yet indexed in the commercial databases.
• Reference lists of included studies will be reviewed, relevant studies identified and articles retrieved online (via QUB database).
Bibliographies of other relevant systematic reviews or metaanalyses will also be reviewed and relevant studies identified and retrieved.
• A citation index search will also be carried out through relevant databases to identify any more recent studies which have cited already identified studies.
• Where a potentially relevant study has been identified however is not available online, the author/s will be contacted via email to request a copy.
All searches will be fully documented to the degree that the searches will be fully replicable.

| Selection of studies
All initial searches will be carried out by the lead author (C. D.), following the strategy set out above. Eligible studies will be imported into EPPI-Reviewer and duplicates identified and removed before screening. A number of trial searches have already been run and the number of returns is not expected to be high; this is largely due to the inclusion of an RCT filter, which significantly narrows down the potential number of studies for inclusion.
The review is being carried out as a Doctoral Dissertation with supervisors as supporting co-authors, therefore there is limited capacity within the team. Should the search return 1000 records or less, dual screening of all records will take place at title and abstract stage. The following process will be followed to ensure robustness of the process.
A test batch of the same 50 records will be allocated to all four reviewers for screening, then Cohen's κ coefficient (k) calculated to measure inter-rater reliability across these records. This process will be repeated, allocating further batches of 50 records for screening by the full team, until satisfied that decisions are consistent across the team and that the screening questions are appropriate. As per Cohen's original discussion, a k value of 0.41 or greater will be considered acceptable. Once satisfied with decision making, the lead author will screen all remaining records, and in addition, distribute them among co-authors (K. W., L. O.'H., T. L.) for independent screening, meaning that all records receive dual screening. Should more than 1000 records be returned from the search process, the lead author will screen all records, and distribute a random sample among co-authors to ensure 1000 records have been independently screened by two authors.
Full text of remaining studies deemed relevant, or where it is unclear as to relevance, will then be retrieved and each record will be screened in duplicate at this stage. The lead author will screen all records, and will distribute all records among the three co-authors to ensure each record is screened by two independent reviewers.
The following screening questions are proposed: 1. Does the study consider use of mobile devices in the classroom?
2. Are study participants in the correct age group (4-11) and within a primary school (or equivalent) class setting?
3. Is the intervention aimed at improving literacy and/or numeracy and related skills?
4. Do pupils use the device themselves (rather than the teacher)?

Is an RCT used?
Where the answer is no to any one of the above questions, the study will be eliminated and no further questions need be answered.
Throughout the process, details of any studies identified which have not yet been completed or reported on will be recorded and revisited before publication. The screening process will also be fully documented using a PRISMA Flow Diagram as specified in chapter 4 of the Cochrane Handbook (Lefebvre et al., 2019). A list of 'characteristics of excluded studies' will be compiled for those studies which appear to meet the eligibility criteria but have been excluded for a specific reason.

| Data extraction and management
When the final list of eligible studies has been identified, coding will take place. A coding framework (see Appendix D) has been adapted from guidelines set out in "Chapter 5: Collecting data" from the Cochrane Handbook . This will be refined following identification and review of the final studies. As the team has limited resources, it is not anticipated that all studies will be dual coded.
Instead, all studies will be coded by the lead reviewer, while a 20% sample will be independently coded by a second reviewer. Again where possible, Cohen's k will be calculated to measure inter-rater reliability, and if not satisfactory (k = 0.41 or greater), a further sample of 20% will be independently double coded. Throughout the search, screening and coding processes, any disagreements will be resolved through discussion with all reviewers until consensus is reached.
Coding will focus on the following information: • Study identifiers and background information (e.g., authors, geography, year, ID, source) • Characteristics of sample/participants (e.g., age, gender, country, ethnicity, sample size, demographics) • Intervention details (such as setting information, location, type of device, activity, frequency, outcome of interest, delivery approach, assessment of SAMR classification (substitution, augmentation, modification, redefinition)). The Template for Intervention Description and Replication (TIDieR) (Hoffman et al., 2014) will be used to guide the information recorded.
• Study design (e.g., specific outcomes measured, tools or instruments used, methods of data collection, timing of data collection, effect sizes) Where the population studied includes children outside of the specified age group (4-11), contact will be made with the author/s to determine if disaggregated data is available. If not, the study will be considered for inclusion if the majority of the study population is within the specified age group. If this is the case, the review team will discuss the implications for the study, consult with the Expert Advisory Group, and undertake sensitivity analysis (discussed below) to assess potential impact on findings, ensuring they "keep faith with the objectives of the review" (Cochrane Handbook, section 3, McKenzie et al., 2019). Where decisions such as these have been made, justification will be fully documented in the final review.

| Assessment of risk of bias in included studies
As this review will only include randomised studies, the latest version of the Cochrane "Risk of Bias" (RoB2) tool will be used to assess for bias in individually randomised trials, including the variant tool for cluster-randomised trials (Sterne et al., 2019). Studies will be rated as low, high or unclear risk of bias. Risk of bias will be independently rated by two reviewers, and disagreements resolved in discussion with a third reviewer.

| Measures of treatment effect
Summary data will be collected from each included study, and metaanalysis will be carried out if sufficient and appropriate studies are identified. It is assumed that there will be variability across the studies, for example in population or implementation of interventions, as well as through sampling error, therefore a random effects model will be used throughout. While the outcomes of interest (literacy and numeracy) have been stated, the ways in which these have been measured will differ across studies. Results will therefore be standardised to allow comparison. Where the dependent variable is continuous, standardised mean difference (d) or correlation coefficient (r) will be calculated, and where there is a dichotomous dependent variable, odds ratio will be calculated. Primary information to be collected will include mean value of the outcome measure, standard deviation for each intervention group, and number of participants in each group.

| Unit of analysis issues
Studies, rather than reports of studies, will be the desired unit of analysis, however the search may return a number of reports of various aspects of the same study. During the screening process, the review team will manually identify and link multiple reports of the same study to avoid "double counting". All reports of the same study will be reviewed to determine the most appropriate information for inclusion. Any additional reports of the study will be used to glean background information.
Should any study have two intervention groups with one control group, the intervention groups will be combined (if similar), otherwise the Robust Variance Estimation (RVE) method will be adopted to deal with nonindependent effect sizes (Hedges et al., 2010). Similarly, RVE will also be used to perform meta-regression where there are multiple outcome measures reported for the same outcome domain (correlated effects) within a single study.
Where studies report multiple outcome measures for the same construct at different follow up periods, the main meta-analysis will focus on outcomes measures immediately posttest. Any follow up data will be grouped into similar time periods and separate metaanalyses carried out. Consideration will also be given to the intensity/ DORRIS ET AL. Given that the included studies will be set in schools, it is likely that cluster randomised trials will be included. In this case, the unit of allocation will be a group or cluster. In such cases, we will assess whether the study has been appropriately adjusted for clustering (for example through the use of multi-level modelling), then follow procedures to estimate effective sample size using an estimate of intracluster correlation coefficient (ICC) as outlined in the Cochrane Handbook Chapter 23 .
The review team will discuss and document any decisions made in selecting the primary data for inclusion.

| Dealing with missing data
Where the study report is missing key data, the reviewers will attempt to calculate the required measures from reported data (e.g., calculating standard error from confidence intervals or p value).
However, if this is not possible, the author will be contacted to request data. Should the data not be accessible the study will not be included in the meta-analysis (however will still be reported).

| Assessment of heterogeneity
Heterogeneity refers to variation in effect sizes across studies. Cochran's Q will be calculated to assess whether any differences between studies are due to chance alone. As Q has low power when the number of studies is low (as expected in this meta-analysis), I 2 will also be calculated and reported, with I > 50% considered moderate heterogeneity and I > 75% considered large heterogeneity.

| Assessment of reporting biases
A number of different reporting biases will be assessed throughout the review process, including: Publication and time-lag bias: The search strategy reported above is constructed to minimise risk of publication bias, including multiple publication, or nonpublication. A funnel-plot will also be constructed (study precision against effect size) and inspected for symmetry, however if the number of studies is low, bias may remain unclear. If appropriate, Egger's regression test may also be used (Egger et al., 1997).
Outcome reporting bias: There may also be bias in terms of the specific outcomes reported on in a study, with data only partially reported, particularly if one or more outcome areas or subsets produce more significant findings. As above, the RoB2 tool will be used to assess potential bias in this regard.
Location and language bias: language and location will not be used to limit searches, and translations will be sought where studies are not presented in English. As noted above, where a translation is not available, the study will be included as 'unclassified' and potential bias assessed and discussed (see sensitivity analysis below).

| Data synthesis
If there are two or more studies with common characteristics which can be meaningfully and logically grouped together, meta-analysis will be carried out. Rev Man will be used to synthesise the main effects across all identified studies, and for each outcome area (literacy and numeracy). This will include weighted mean effect size, standard error and confidence interval. Forest plots will be used to display findings. In the event that there are not sufficient studies to undertake meta-analysis, a narrative synthesis will be undertaken and reported.

| Subgroup analysis and investigation of heterogeneity
Again where meaningful we will undertake subgroup analysis to identify any specific characteristics that may have a greater or lesser effect; this will be done using meta-analytic regression on the following moderating factors: • Activity classification via the SAMR framework; that is, activities are judged to reflect substitution, augmentation, modification or redefinition as per Puentedura's definitions.
• Screen size: as discussed earlier, research suggests a number of ways in which screen size may impact usage and outcomes, therefore it will be important to determine if this is a moderating factor. This will have important implications for future practice, particularly since the number of children with their own smartphone (with typically smaller screens than tablets) is increasing (OFCOM, 2019).
• Gender: In the early days of computers, boys were considered to be more enthusiastic users (Bergin et al., 1993). Recent research concludes that girls and boys now spend similar amounts of time using technology, and are equally proficient, however their activities differ, with girls more likely to use computers for homework or social media, while boys are much more likely to play computer games (Mullan, 2018). Any potential difference in impact across the genders in terms of educational outcomes will have implications for practice.
• Intervention frequency: defined as low, medium or high (these will be more clearly defined by reviewing the included studies following coding).

| Sensitivity analysis
Although effort will be made throughout the review process to remain objective, there are various stages at which decisions made may impact final conclusions (decision nodes), as per Cochrane Handbook Chapter 10 (Deeks et al., 2019). Sensitivity analyses will therefore be undertaken to demonstrate that the review is robust despite any decisions made or eligibility criteria employed: an example is the discussion above where a sample may include some children outside of the specified age range. A subjective decision will be required from the review team as to inclusion.
Sensitivity analysis will also be conducted to determine if any of the following unduly influence the findings: • Studies with a high or unclear risk of bias • Studies with incomplete data • Studies with outlier effect sizes (identified through a funnel plot) All decisions will be fully documented in a summary table, and steps taken to resolve any issues that may adversely impact the strength of conclusions drawn.

CONTRIBUTIONS OF AUTHORS
This systematic review and meta-analysis will form the basis of Claire Winter and Dr. O'Hare will advise on the wider review process including information retrieval, outcome measures and statistical analysis. Prof. Lwoga will advise and support in developing the search strategy and in information retrieval.

DECLARATIONS OF INTEREST
There are no conflicts of interest noted for any of the authors. Bebell and Pedulla (2015) A quantitative investigation into the impacts of 1:1 iPads on early learner's ELA and math achievement. This review considered both primary and post-primary schools, did not include smartphones, and focused on wider learning outcomes. There is no subgroup analysis completed, either across age groups or specific learning outcomes, therefore the team feels our proposed review It will be useful to revisit this review following completion of the proposed review to identify any similarities or differences in findings. Proposed review will also seek to add detail to the discussion presented in this review Talan  AB RCT OR "randomised control trial" OR "randomised controlled trial" OR "randomized control trial" OR "randomized controlled trial" OR randomised OR randomized OR placebo* OR (random* AND (allocat* OR assign*)OR (blind* AND (single OR double OR treble OR triple)) S8 S6 AND S7 S7 AB iPad OR tablet OR "tablet computer*" OR touchscreen OR app OR "handheld device" OR "handheld computer" OR "PDA" OR "personal digital assistant" OR "portable computer" OR '"e-book" OR ebook OR "e-reader" OR ereader OR "electronic storybook" OR "game* console" OR "digital game" OR smartphone OR "smart phone" OR "mobile phone" OR iPhone OR "cell* phone" OR "portable cell* phone" OR "mobile telephone" OR "cell* telephone" OR "transportable Cell* Phone" OR "1:1 comput*" OR "online instruction" OR "mobile learn*" OR M-learning S6 S4 OR S5 S5 AB numeracy OR number* OR math* OR arithmetic OR sums OR calculat* OR addition OR adding OR subtract* OR multiplication OR division OR count* OR algebra OR fractions OR decimal* OR geometr* OR statistic* OR "problem solving" OR "shape sort*" OR "mathematical literacy" OR "quantitative literacy" S4 AB Literacy OR Reading OR Writing OR Handwriting OR Transcription OR "Verbal reasoning" OR "Critical thinking" OR Comprehension OR Notetaking OR Composition OR "Listening skills" OR Phonic* OR "phonemic awareness" OR Alphabet OR Spell* OR Vocabulary OR Punctuation OR Grammar OR "Reading fluency" OR "Speaking skills" OR "Spoken language" OR "Critical literacy" OR "literacy skills" S3 S1 OR S2 S2 AB "Primary school" OR "Elementary school" OR "Junior school" OR Kindergarten OR "grade 1" OR "grade 2" OR "grade 3" OR "grade 4" OR "grade 5" OR "First Grade" OR "Second Grade" OR "Third Grade" OR "Fourth Grade" OR "Fifth Grade" OR "1st grade" OR "2nd grade" OR "3rd Grade" OR "4th Grade" OR "5th grade" S1 AB ("Age* 4" OR "age* 5" OR "age* 6" OR "age* 7" or "age* 8" OR "age* 9" OR "age* 10" OR "age* 11" OR "4 year* old*" OR "5 year* old*" OR "6 year* old*" OR "7 year* old*" OR "8 year* old*" OR "9 year* old*" OR "10 year* old*" OR "11 year* old*" OR Child* OR boy OR girl) AND (school OR class* OR classroom)

Journal of Information
The following tool has been adapted from guidelines set out in "Chapter 5: Collecting data" from the Cochrane Handbook for Systematic Reviews of Interventions. The tool will be piloted and refined before final data extraction.

APPENDIX D: DATA EXTRACTION FRAMEWORK
Responses/notes (free text unless options given) Location in text or source (e.g., pg, figure no.)

General information
Name of data extractor: Date of data extraction Study ID or reference Report ID (if more than one report relating to the same study) Any other results reported (e.g., odds ratio, risk difference, CI or P value) No. missing participants

Reasons missing
No. participants moved from other group

Reasons moved
Unit of analysis (by individuals, cluster/groups) Statistical methods used and appropriateness of these (e.g., adjustment for correlation) Reanalysis required? (specify, e.g., correlation adjustment) Y/N/unclear (specify) Reanalysis possible? Y/N/unclear (specify) Reanalysed results