50 Word summary
For some students, handwriting can obstruct rather than express their ideas. Personal accounts and a small study of children with dysgraphia describe how keyboards can
Australia does not entirely lack numbers for typing performance. It lacks something more consequential: an authoritative, developmental and assessable progression describing what students should be able to do, approximately when, and how their performance should be measured.
The national Literacy progression describes students using handwriting and keyboarding “with increasing speed, accuracy and fluency to compose and edit text”. Its terminal keyboarding indicator is that a student “demonstrates automaticity when using keyboarding and screen functions”. But it does not specify how fast, how accurately, under what conditions, or how teachers should recognise automaticity.
The consequence is a familiar circular argument:
Because nobody can establish the perfect threshold, no numbers are published. Because no numbers are published, there is no shared expectation. And because there is no shared expectation, keyboarding is too often left untaught.
The responsible course is not to pretend the research has established an immutable universal threshold. It is to publish a provisional, transparent and revisable progression, evaluate it against students’ actual work, and improve it as better evidence emerges.
The goal is not speed for its own sake. It is to reduce the cognitive effort spent on transcription so students can concentrate on thinking and composing. This article examines the curriculum-writing problem, weighs Australian and international evidence, identifies the important research gaps, and proposes a progression from keyboard awareness to accurate, automatic and productive use.
The national Handwriting and keyboarding sub-element supports teachers to plan, monitor and assess development. It is not a syllabus or a teaching sequence. Its terminal indicator says that a student “demonstrates automaticity when using keyboarding and screen functions” (ACARA, 2022).
It does not provide:
The progression describes handwriting and keyboarding as a constrained skill—one that can be mastered within a bounded period and, once established, should not require continual basic-skills teaching. Yet it provides no common measure of that mastery.
The national English curriculum does not supply the missing progression. Students are asked to plan, create, edit and publish texts without a specified keyboarding pathway.
Handwriting is also commonly described through terms such as legible, fluent, efficient and automatic, without a national letters-per-minute standard. The absence of typing benchmarks is therefore part of a wider problem: curriculum documents describe transcription through qualitative terms but seldom specify how to assess it objectively.
NSW has gone further. Its English K–10 Syllabus includes the focus area Handwriting and digital transcription, with outcomes at Stages 2 and 3. Departmental teaching sequences recommend regular explicit instruction, name touch typing and ask students to monitor typing rate and accuracy against goals and criteria (NESA, 2022; NSW Department of Education, n.d.).
The Stage 3 sequence includes 35 words per minute at 98 per cent accuracy as an example. That is useful guidance, but it is not a mandated state benchmark. Students are also directed to develop individual goals from their own starting points.
NSW therefore provides a named focus, instructional guidance and an illustrative target—but not a common, required progression with consistent assessment and reporting.
Victoria provides a useful contrast with NSW. Victorian Curriculum F–10 Version 2.0 English organises writing under the “Creating texts” strand, but does not establish a discrete handwriting-and-digital-transcription focus area comparable with NSW’s, or prescribe typing-speed and accuracy benchmarks (VCAA, 2024). The point is not that Victoria’s curriculum says nothing about digital writing. It is that teachers are not given a clear, assessed progression for developing keyboard fluency, nor numerical expectations for typing or handwriting. Across jurisdictions, the terminology and detail vary; the common gap is that students and schools lack consistent, measurable transcription goals.
There are reasonable explanations for the reliance on adjectives.
The documents have different purposes. A learning progression describes development across years and learning areas. It is not a testing manual. A number immediately raises questions about task length, text type, keyboard, autocorrect, treatment of errors and whether the measure is copy typing or original writing. Without agreed conditions, a speed figure can seem precise while measuring different things in different schools.
The skill falls between curriculum areas. Keyboarding is simultaneously a literacy skill, a motor skill, a digital-production skill and an assessment-access skill. English specialists focus on composition and language; Digital Technologies specialists focus on systems, data and computational thinking. Keyboarding can end up important to everyone’s work but owned by no one.
The research does not establish a universal cognitive threshold. Automaticity depends not just on speed but on technique, accuracy, text complexity, spelling demands, device familiarity and whether the student is copying or composing. It is prudent not to claim there is one scientifically proven speed at which transcription ceases to consume attention.
These considerations explain why a benchmark should be provisional and carefully defined. They do not justify leaving “automaticity” without observable criteria. Curriculum writers can specify what automaticity looks like, define a consistent task, and set a revisable target.
There are also legitimate equity concerns. A benchmark exposes unequal access to keyboards, instructional time, trained teachers and suitable software. But leaving the skill unmeasured does not protect students facing those disadvantages. It makes the differences harder to see and address.
Finally, numbers can be misused. If speed becomes the sole measure, students may rush, reinforce poor technique and produce inaccurate text. That is why any standard must include technique, accuracy and application—not simply a wpm target.
Cognitive load theory distinguishes between the limited working memory used to process unfamiliar information and organised knowledge stored in long-term memory (Sweller, van Merriënboer, & Paas, 2019).
A novice typist may need to manage several demands at once:
As key locations and motor patterns become familiar, students can rely less on visual search and conscious key-by-key decisions. More attention can then be directed towards ideas, sentences, evidence, audience and revision.
Automaticity does not make writing effortless. Composition remains demanding. It reduces one avoidable demand on attention.
Australian research supports the importance of transcription fluency. A study of 544 Year 2 students in 47 classrooms found that keyboarding automaticity made a distinct contribution to the quality and productivity of computer-based composition, alongside spelling, reading and other factors (Malpique et al., 2025). That is evidence of a relationship, not proof that typing faster by itself causes better writing.
The educational objective should therefore be fluency for composition, not speed as an end in itself.
The NSW Centre for Education Statistics and Evaluation study is the strongest Australian baseline source discussed here. It is a government research report, not a peer-reviewed journal article. It was published in 2021 from research conducted in 2016 (CESE, 2021).
The study used paired paper and computer writing tasks, which is a significant strength: the same students completed writing tasks in both modes. A total of 1,651 students across 69 schools provided paired writing scripts. Typing-test results were available for 1,323 students, and 879 students had both writing and typing data for the analysis of typing speed and writing outcomes (CESE, 2021). The study is a government research report, not a peer-reviewed journal article.
| Year | Median typing speed | Median typing accuracy |
| Year 2 | 5.3 wpm | 93.7% |
| Year 3 | 8.7 wpm | 95.7% |
| Year 4 | 10.3 wpm | 94.2% |
| Year 5 | 15.4 wpm | 94.9% |
CESE reported that, after adjustment for relevant factors, computer-based writing scores were estimated to be 15–20 NAPLAN scaled points lower than paper-based scores, depending on year level. Typing speed was significantly associated with this difference: in the study’s model, each additional 5 wpm was associated with a reduction of about 7.5 scaled points in the computer–paper gap (, ). These are observational associations; the study did not test whether teaching students to type faster causes a specified increase in writing scores.
These are observational associations. They are not proof that an intervention adding 5 wpm will automatically add 7.5 NAPLAN points.
CESE measured students largely as the existing system had left them: keyboard users who, in most trial schools, had not received explicit keyboarding instruction.
It did not study:
This omission does not invalidate CESE’s findings about students’ performance under the conditions studied. It limits what the report can tell policymakers about the potential of computer-based writing after students have been taught to type.
CESE did not determine whether the mode difference persists among students typing at 18–20 wpm, 25–30 wpm or 40 wpm with strong accuracy and technique. Nor did it establish whether handwriting retains an advantage over the keyboard when both modes have been explicitly taught.
The distinction matters. A comparison between generally taught handwriting and largely untaught keyboarding is not a fair test of what computer-based writing could achieve under competent instruction. CESE documents performance in a system where keyboarding was seldom taught; it does not show that teaching keyboarding cannot work.
The EdAlive/Brash analysis is also not peer-reviewed and is developer-funded. It should not be privileged because it is large, nor dismissed solely because of its provenance.
The 2019 analysis compared Year 5 writing results from 99 schools with each school’s own NAPLAN “Similar Schools” benchmark. The Year 5 cohort was approximately 5,350 students. Sixty-nine schools scored at or above their benchmark, with an average margin of 7.1 writing marks (p = 0.000002) (Brash, 2020).
Its limitations matter:
Its reported findings also matter. The analysis provides a large operational dataset, uses a matched school-level benchmark, reports limitations, and offers evidence that classes using the programme regularly can achieve substantially higher average speeds than those found in the CESE baseline. EdAlive materials report class averages above 40 wpm and document a Year 5 class at St Mary’s School in Dubbo increasing from roughly 20 wpm to 43.43 wpm in a year (Alexander & East, 2019; EdAlive, 2019).
These are developer-reported classroom results, not independent intervention evidence. They demonstrate reported attainability under particular conditions; they cannot guarantee equivalent results for all students or schools.
Read together, the two sources are complementary, but neither is conclusive:
That missing comparison is crucial to the policy question.
There is no authoritative international speed staircase. Published expectations combine technique, growth, speed, accuracy and application in different ways.
| Jurisdiction or document | Published expectation | Relevance to Australia |
| Nebraska K–12 Technology Scope and Sequence | Advisory guidance says students should “type 5 words per grade level beginning at 2nd grade”. | Shows how a simple progression can make expecttions visible, but the formula is guidance—not a research-established developmental law. |
| Nebraska model keyboarding sequence | Keyboard awareness in K–3; elementary keyboarding in Grades 3–6; recommends 20–40 days a year of 20–30 minutes a day. | Highlights instruction time, sequencing and application, not just speed. |
| Utah elementary rule | Requires keyboarding proficiency to be assessed by Grade 5 and school-level results reported. Defines proficiency as the ability to key by touch; it does not set a Grade 5 wpm target. | Demonstrates that technique and reporting can be required without inventing a universal speed threshold. |
| Utah Keyboarding Applications course | Students entering the secondary course are expected to have 25 wpm; course guidance expects improvement and sets error limits for timed writing. This is a course standard, not the elementary rule. | Shows how secondary expectations can connect speed, accuracy, document creation and editing. |
| Texas Technology Applications TEKS | Grade 6 students must demonstrate improvement in speed and accuracy as measured by words per minute, using correct technique. No particular target speed is prescribed. | Supports measuring progress rather than imposing an unsupported endpoint. |
| US Common Core writing standards | Grades 4–6 specify increasing page-length expectations for typing in a single sitting. These are output expectations, not speed standards. | Shows that typed production can be part of the expected writing task, while illustrating why fluency matters. |
| England’s 2025 Writing Framework | Says students need to learn to type accurately and speedily, but provides no national wpm target. Its main focus is primary handwriting. | Supports the importance of typing while illustrating that other systems also hesitate to prescribe numbers. |
| New Zealand NZQA Unit Standard 32000 | Basic keyboarding is approximately 15 wpm; touch typing is not required. The finished document must be 100% accurate after proofreading. | Distinguishes initial typing accuracy from the accuracy of a corrected final document. |
| South Africa Information Processing N4 | Sets a learning aim of 35 wpm and a ten-minute accuracy test with an error limit. Its examination production tasks, however, use a much lower prescribed production speed. | A vocational qualification, not a school-age norm; illustrates that timed accuracy, speed development and real production tasks may use different measures. |
A progression should distinguish:
Year levels should be checkpoints, not rigid age-based limits.
CESE’s typing-test medians were around 95 per cent. A provisional progression can therefore use:
Finished documents should be proofread to a much higher standard. First-pass typing accuracy and a corrected final document are different measures.
A Toronto District School Board pilot illustrates why speed alone can mislead: after instruction, students’ measured accuracy rose while their measured wpm fell. The pilot had no control group, so it does not establish the cause. It does show that typing more slowly while making fewer errors should not automatically be judged as regression (Toronto District School Board, 2017).
Keyboarding assessment should examine four things separately:
A student should not advance merely because they hit a speed target.
A workable approach would be:
Speed and accuracy calculations must also be specified. Different tests may count words, characters or errors differently. Results should not be compared unless the measures are compatible. Where gross words per minute is used:
Gross wpm =\frac{\text{typed characters, including spaces}}
{5 \times \text{minutes}}
Accuracy should be defined consistently for that test. CESE’s measure—words typed correctly as a proportion of words entered—should not be silently treated as equivalent to character-level accuracy.
Benchmarks should guide teaching, not rank children. Students with motor impairments or other access needs may require different goals, assistive technology or alternative ways to demonstrate composition. Schools should report progress to students and families and use aggregated results to evaluate instruction, not publish individual speed league tables.
The following are provisional policy recommendations, not research-established thresholds for automaticity. They draw on Australian baseline evidence and overseas curriculum or vocational standards, which differ in purpose and should not be treated as directly equivalent. The proposed figures should be trialled, evaluated against student-level outcomes and revised as stronger evidence becomes available.
The early aim is functional fluency around Year 4, followed by secure automaticity in upper primary, independent curriculum fluency by Year 8, and general digital productivity by Year 10.
| End of year | Provisional copy-typing benchmark | Accuracy target | Observable performance alongside the number |
| Foundation | No speed target | Not applicable | Keyboard and mouse awareness; appropriate seating; letter recognition. Handwriting remains the principal transcription mode. |
| Year 1 | No speed target | Not applicable | Locate letters and basic keys; learn appropriate hand and body position; short guided activities without speed pressure. |
| Year 2 — awareness | 8–10 wpm | 90% | Home-row awareness; increasing use of correct fingers; brief copy tasks; no requirement for full touch mastery. |
| Year 3 — foundations | 12–15 wpm | 92–95% | Alphabet keys increasingly by touch; correct fingering; simple punctuation; handwriting instruction continues. |
| Year 4 — first functional fluency | 18–20 wpm | 95% | Three-minute copy task; eyes principally on copy or screen; can compose and revise a short passage without continual key searching. |
| Year 5 — emerging automaticity | 20–25 wpm | 95% | Sustains a three-minute task, corrects text and begins using common shortcuts. |
| Year 6 — secure primary automaticity | 25–30 wpm | 95–97% | Independent drafting and editing; basic tables and document management. NSW’s 35 wpm at 98% may be an extension goal. |
| Year 7 — application | 30–35 wpm | 97% | Applies typing across subjects; maintains accuracy while researching, taking notes, drafting and revising. |
| Year 8 — independent curriculum fluency | 35–40 wpm | 97% | Sustained typing; produces reports and tables; uses citations, spreadsheets and collaborative documents. |
| Year 9 — subject production | 40–45 wpm | 97% | Produces reports, design folios, presentations, data tables and extended responses efficiently. |
| Year 10 — general digital productivity | 45–50 wpm | 97% | Composes, formats, revises and produces documents under normal assessment conditions. |
| Years 11–12 — general | Maintain 45–50 wpm | 97% | Maintains fluency through curriculum work; no continual basic-skills lessons required. |
| Years 11–12 — extension | 50–60+ wpm | 97–98% | Optional higher target for text-intensive vocational, technical, administrative or programming pathways. Accuracy and application remain more important than speed alone. |
These targets should be tested and revised. They are not a licence to make speed a high-stakes outcome or to treat a child who misses a benchmark as a failed writer.
A student at 40 wpm who can draft, revise, format, enter data and manage documents is better prepared for digital life than one at 60 wpm who can complete only copying drills.
On 2 November 2026, Australia’s education ministers agreed that the Year 5 NAPLAN writing test would return to pen and paper from 2028. New South Wales led the push; ACARA was asked to advise on extending the change to Years 7 and 9.
The decision did not stand alone. It formed part of a wider package addressing screen time, generative AI and curriculum reform, including the establishment of an Australian Schools Digital Detox Taskforce (Duffy, 2026; NSW Department of Education, 2026; Clare, 2026).
The concerns behind that package are genuine. Excessive or poorly structured device use can harm attention and engagement; unproductive screen time should be reduced; and handwriting has well-documented developmental benefits.
But those are general arguments about devices in classrooms. They are not evidence about the specific instrument being altered.
In the public material reviewed for this article, no Australian authority has published an analysis using operational NAPLAN cohorts to determine how typing fluency affected the online writing task. ACARA’s research repository contains studies of test design, automated scoring, device suitability, engagement, readability and accessibility — but none measuring student words per minute against writing marks (ACARA, 2026).
The one departmental study to examine transcription in detail, CESE, recommended developing and monitoring typing fluency rather than abandoning the online mode. The 2020 NAPLAN Review reached the same conclusion.
Section 9 sets out what should be commissioned before the change is treated as settled.
The evidence supports several conclusions:
Important questions remain open:
Before the 2028 reversal is treated as settled, policymakers should commission:
The first could build on a design CESE has already used. The second addresses the central cohort missing from the existing evidence. The third would help refine the proposed benchmarks rather than leave them as conventions.
The onus is on those implementing change to validate their decisions through published research. If ministers, curriculum authorities or researchers disagree with the conclusions or proposed progression here, the appropriate response is to provide the tutored-cohort and student-level evidence that would settle the questions.
Until then, Australia should proceed on the most defensible available course: teach both transcription modes explicitly, measure development carefully, and treat benchmarks as provisional instruments for improving instruction—not immutable laws of cognition.
Australian Curriculum, Assessment and Reporting Authority. (2020). NAPLAN Online: Frequently asked questions. https://www.nap.edu.au/naplan/faqs
Australian Curriculum, Assessment and Reporting Authority. (2022). Literacy: Australian Curriculum Version 9.0. https://www.australiancurriculum.edu.au/curriculum-information/understand-this-general-capability/literacy
Australian Curriculum, Assessment and Reporting Authority. (2026). Understanding online assessment: Research and development. https://www.nap.edu.au/naplan/understanding-online-assessment/research-and-development
Brash, L. (2020). Full research report: Keyboard ability significantly boosts NAPLAN results. EdAlive. https://www.edalive.com/full-research-report-keyboard-ability-significantly-boosts-naplan-results/
Malpique, A. A., Asil, M., Pino-Pasternak, D., Ledger, S., & Teo, T. (2023). The keys of keyboard-based writing: Student and classroom-level predictors of keyboard-based writing in early primary. Contemporary Educational Psychology, 75, Article 102227. https://doi.org/10.1016/j.cedpsych.2023.102227
Malpique, A. A., Asil, M., Pino-Pasternak, D., Ledger, S., & Teo, T. (2025). The contributions of transcription skills to paper-based and computer-based text composing in the early years. Reading and Writing, 38(4), 911–945. Published online 20 April 2024. https://doi.org/10.1007/s11145-024-10543-6
Malpique, A. A., Valcan, D., Pino-Pasternak, D., & Ledger, S. (2025). Teaching computer-based writing: Primary teachers’ preparation, self-efficacy, and instructional practices. The Australian Educational Researcher. https://doi.org/10.1007/s13384-025-00910-9
NSW Department of Education. (n.d.). English Stage 3 instructional sequence: Handwriting and digital transcription. https://education.nsw.gov.au/content/dam/main-education/documents/teaching-and-learning/curriculum/english/english-s3-handwriting-and-digital-transcription-instructional-sequence.pdf
NSW Education Standards Authority. (2022). English K–10 Syllabus: Handwriting and digital transcription. https://curriculum.nsw.edu.au/learning-areas/english/english-k-10-2022/overview
Centre for Education Statistics and Evaluation. (2017). Keyboarding instruction at the Toronto District School Board: Pilot study 2016–17—Recommendations for policy and practice. Toronto District School Board. https://www.tdsb.on.ca/Portals/research/docs/reports/Keyboarding%20Report%202016-17%20-%2026Sep17.pdf
Department for Education. (2025). The writing framework. UK Government. https://assets.publishing.service.gov.uk/media/68bec95444fd43581bda1c86/The_writing_framework_092025.pdf
Department of Higher Education and Training. (2021). Information Processing N4: Subject syllabus (Subject Code 6020254). Republic of South Africa. https://www.dhet.gov.za/Revised%20Report%20191%20Nated%20Syllabus/Business%20Studies/2025-2026/N4%20Information%20Processing.pdf
National Governors Association Center for Best Practices, & Council of Chief State School Officers. (2010). Common Core State Standards for English Language Arts: Writing standards, Grades 4–6. https://www.thecorestandards.org/ELA-Literacy/W/
Nebraska Department of Education. (2018). Nebraska K–12 Technology Scope and Sequence. https://www.education.ne.gov/wp-content/uploads/2018/04/NEK12Tech.pdf
Nebraska Department of Education. (n.d.). Model Scope and Sequence for Keyboarding. https://www.education.ne.gov/wp-content/uploads/2017/07/modelscopeandsequence.pdf
New Zealand Qualifications Authority. (2024). Unit Standard 32000: Enter prescribed text accurately using basic keyboarding skills and identify safe ergonomic practices. https://www.nzqa.govt.nz/nqfdocs/units/pdf/32000.pdf
Texas Education Agency. (2022). 19 Texas Administrative Code §126.17: Technology Applications, Grade 6. https://tea.texas.gov/laws-and-rules/texas-administrative-code/19-tac-chapter-126
Toronto District School Board. (2017). Keyboarding instruction at the Toronto District School Board: Pilot study 2016–17—Recommendations for policy and practice. https://www.tdsb.on.ca/Portals/research/docs/reports/Keyboarding%20Report%202016-17%20-%2026Sep17.pdf
Utah State Board of Education. (2026). R277-700: The Elementary and Secondary School General Core (effective 7 April 2026). https://schools.utah.gov/adminrules/administrativerules/_administrative_rules_/_effective_rules/R277700EffectiveApril2026.pdf
Utah State Board of Education. (2017). Keyboarding Applications: Strands and standards. https://schools.utah.gov/cte/_cte/strands/KeyboardingApplications.pdf
James, K. H., & Engelhardt, L. (2012). The effects of handwriting experience on functional brain development in pre-literate children. Trends in Neuroscience and Education, 1(1), 32–42. https://doi.org/10.1016/j.tine.2012.08.001
Santangelo, T., & Graham, S. (2016). A comprehensive meta-analysis of handwriting instruction. Educational Psychology Review, 28(2), 225–265. https://doi.org/10.1007/s10648-015-9335-1
Skar, G. B. U., Lei, P.-W., Graham, S., Aasen, A. J., Johansen, M. B., & Kvistad, A. H. (2022). Handwriting fluency and the quality of primary grade students’ writing. Reading and Writing, 35(2), 509–538. https://doi.org/10.1007/s11145-021-10185-y
Sweller, J., van Merriënboer, J. J. G., & Paas, F. (2019). Cognitive architecture and instructional design: 20 years later. Educational Psychology Review, 31(2), 261–292. https://doi.org/10.1007/s10648-019-09465-5
For some students, handwriting can obstruct rather than express their ideas. Personal accounts and a small study of children with dysgraphia describe how keyboards can
Don’t Take Our Word for It: Validate the Research Yourself In education, we are often asked to pivot based on the latest policy shift, educational
How falling for the “zero-sum”, “osmosis”, and “all screens are equal” fallacies threatens student literacy—and how to plan for 2027. When policy pivots rapidly, stressed