Computers and Education Submission Guide

Source: Computers and Education on Elsevier, Clarivate JCR 2024, accessed May 2026.

For a side-by-side read on the axes authors weigh, this peer-comparison table sets Computers & Education against its closest sibling venues:

Source: Clarivate JCR 2024 and publisher author pages, accessed June 2026. Acceptance and timing figures are approximate community estimates.

Three operational signals govern editorial assessment:

1. Educational-technology substance. The journal requires substantive edtech contribution.

2. Learning-outcomes connection. Top-tier edtech papers connect technology to learning outcomes empirically or theoretically.

3. Methodological rigor. Empirical, design-based, or qualitative methods must be top-tier.

4. Wider education community. The official scope says small-scale evaluations of specific software or systems in specialist domains are not normally published unless wider relevance is explicitly drawn out. That is the filter many AI-in-education and course-tool papers miss.

Recent issues span:

• AI in education (LLMs, ChatGPT, intelligent tutoring)

• Learning analytics and educational data mining

• Online and distance learning post-COVID

• Game-based learning and gamification

• VR/AR for education and training

• Computational thinking and CS education

• Digital literacy and ICT competency

• Mobile and ubiquitous learning

For specific recent papers and DOIs, see Computers and Education on ScienceDirect. Current article anchors we checked included:

• 10.1016/j.compedu.2026.105563

• 10.1016/j.compedu.2026.105593

• 10.1016/j.compedu.2026.105592

The recent pattern is clear: AI, VR, algorithmic advice, peer feedback, and data literacy papers work best when they are written as education research first and technology demonstrations second.

Computers & Education sets no rigid body word cap for Articles (most run 8,000 to 10,000 words), and Editorial Manager accepts source files up to roughly 50 MB per file, with figures supplied as separate high-resolution files.

Required-artifacts checklist (have these ready before you open Editorial Manager):

• Cover letter stating the educational-technology contribution and the learning-outcomes connection

• Author contributions and a fixed, anonymized authorship list

• Conflicts of interest declaration (required even when none exist)

• Funding statement naming grants and funding bodies

• Data availability statement and an ethics-approval statement for human-subjects (student) data

• ORCID iDs for the submitting authors

• Suggested reviewers who are genuine edtech experts, not collaborators

Before upload, run your manuscript through a Computers & Education submission readiness check if the technology is strong but the learning-outcomes, theory, or broader education contribution is still uncertain.

If you are still comparing C&E with BJET, ETR&D, JECR, ET&S, or Computers & Education: Artificial Intelligence, start with the general Computers and Education manuscript fit check.

The editorial triage runs on a predictable clock. These markers describe a typical Computers & Education submission from upload through the first editorial decision:

• Day 0: submission lands in Editorial Manager and the technical check runs (anonymization, highlights file, declarations)

• Day 1-7: the handling editor screens for broad educational-technology relevance and learning-outcomes connection

• Day 7-14: desk decision (reject, redirect to a sister edtech venue, or send to double-anonymized review) for most submissions

• Day 14-48: external review for papers that clear the desk, ending in the first decision after review

Longer-horizon expectations from current ScienceDirect journal insights:

• Submission to first decision: 7 days

• Submission to decision after review: 48 days

• Submission to acceptance: 190 days

• Acceptance to online publication: 2 days

Treat those as journal-level indicators, not a promise for one manuscript. The submission decision you control is whether the study is broad educational-technology research before the editor evaluates fit.

Official sources set the requirements, but the remaining question is manuscript fit. The review tells you whether your paper clears the Computers and Education fit check before upload, especially around tool-evaluation paper without learning-outcome evidence beyond satisfaction and intention-to-use, aI-in-education paper where the novelty is the AI architecture rather than the educational consequence, and single-context deployment study without external-generalizability discussion or theoretical contribution. Paid Manusights reviews include a 60-day money-back guarantee, and we do not train models on submitted manuscripts.

Across edtech manuscripts targeting Computers & Education, three recurring decision risks matter most across submissions that C&E editors filter out at the desk-screen stage. (Per Elsevier published medians, C&E issues 7-day first decisions and 48-day post-review decisions, with 190 days submission-to-acceptance; papers that do not show learning, teaching, usability, or user-experience consequences can be returned before external review.) These are testable against your own manuscript before you commit to the Elsevier submission route.

A Manusights review checks whether your paper clears the Computers & Education-specific readiness checks that official Elsevier instructions cannot evaluate from a generic Editorial Manager checklist. Paid Manusights reviews include a 60-day money-back guarantee; submitted manuscripts are not used for model training.

Tool-evaluation paper without learning-outcome evidence beyond satisfaction and intention-to-use

Across C&E-targeted manuscripts, we consistently see authors submit small-scale evaluations of a single LMS feature, tutoring system, VR module, or classroom platform that measure satisfaction scales (System Usability Scale, UEQ, custom Likert satisfaction), intention to use (UTAUT2 / TAM behavioral-intention items), self-reported perceived learning, time-on-task, or click-stream engagement metrics, and stop there.

C&E editors specifically check whether the outcome battery includes at least one validated learning-outcome measure: pre/post conceptual-knowledge test with established reliability (Cronbach's alpha >= 0.7 with item-source citation), validated educational-psychology construct (self-regulated learning via OSLQ, motivation via IMI / MSLQ, metacognition via MAI, computational thinking via CTt), behavioral evidence of learning transfer at delay (2-week or longer retention test, near-transfer task, far-transfer task), or rigorous performance evidence (grade, course outcome, certification pass-rate, work-product quality rubric).

Manuscripts where the outcome battery is entirely perceptual or behavioral-intent can be redirected to discipline-specific venues (Computers in Human Behavior for the affective/intention side, BJET / ETR&D for design-research framing, Journal of Educational Computing Research for narrower platform studies, Education and Information Technologies for descriptive deployments) during the 7-day first-decision window.

The fix is to add one validated learning-outcome measure with delayed-retention design, report effect sizes with confidence intervals (Cohen's d or Hedges' g, not just p-values), and front-load the learning result in the abstract before the satisfaction/intention findings.

Check whether your Computers & Education manuscript shows learning-outcome evidence →

AI-in-education paper where the novelty is the AI architecture rather than the educational consequence

In Manusights reviews, we observe that C&E manuscripts featuring generative AI, intelligent tutoring systems, automated feedback, learning analytics, or LLM-tutor designs frequently lead with the AI architecture novelty (transformer fine-tuning approach, RAG pipeline design, prompt-engineering scheme, multi-agent orchestration, embedding-similarity scoring) and treat the educational contribution as a downstream demonstration.

C&E's editorial culture treats this inversion as a publication-venue mismatch: the journal's published criterion is that technology must change learning, teaching, feedback, self-regulation, assessment, or educational decision-making, not that the technology is novel in CS terms.

Manuscripts where the contribution statement names a model / architecture / prompt-scheme as the primary novelty get flagged at desk and either rejected (to redirect to TKDE, ACM Learning@Scale, ACM TOCHI, EMNLP-Findings, NeurIPS-Workshop, AIED main track) or returned for major revision asking for an educational reframing that the authors may not be positioned to deliver.

The fix is to restructure the contribution statement so the educational change is the load-bearing claim (e.g., "we show that AI feedback restructures self-regulated learning patterns for novice programmers" rather than "we present an LLM-based feedback system"), demote the AI architecture details to methods (or supplementary), include a learning-theory grounding for why the AI design should produce the educational effect (cognitive load theory, ICAP, self-determination theory, ZPD, productive failure, desirable difficulties), and report educational rather than ML-benchmark metrics in the headline results.

Check whether your Computers & Education AI contribution is educational rather than just technical →

Single-context deployment study without external-generalizability discussion or theoretical contribution

The third recurring pattern in C&E-targeted manuscripts is a deployment study at a single institution, single course, single grade-level cohort, or single national/cultural context that reports findings without addressing what readers outside that setting should take from the work.

C&E reviewers (drawn from international edtech research community) specifically check whether the discussion section addresses: external generalizability (which course features, instructional conditions, learner populations, cultural contexts the findings should or should not transfer to, with explicit boundary conditions), theoretical contribution (which prior framework the findings extend, refute, or refine; not just citing the framework but moving it forward), educational-design implications (specific patterns practitioners outside the study setting can adopt), and a limitations section that names the generalizability limits without retreating to "more research is needed" boilerplate.

Manuscripts that read as a local case study (well-executed but parochial) face revision requests demanding these additions or get redirected to regional/specialty venues (Education and Information Technologies for descriptive deployments, regional journals, conference proceedings).

The fix is to add an external-generalizability paragraph naming 2-3 specific contexts where the design should and should not transfer, name a specific theoretical contribution beyond "we extend X" (one concrete way the framework moves forward), include design implications stated as patterns rather than as recommendations, and own the single-context limitation honestly in limitations rather than hiding it in the discussion.

Check whether your Computers & Education generalizability argument is strong enough →

• Tool evaluation without broader education relevance. A study of one LMS feature, tutoring system, VR module, or classroom platform can be well executed and still too narrow if it does not explain what educators outside that setting learn.

• AI novelty without learning theory. Generative AI, intelligent tutoring, automated feedback, and analytics papers need more than novelty of the model. The paper has to show how the technology changes learning, teaching, feedback, self-regulation, assessment, or educational decision-making.

• Outcome measures too shallow. Satisfaction, intention to use, clicks, time-on-task, or perceived usefulness are rarely enough alone. Strong C&E submissions usually connect technology use to learning outcomes, conceptual understanding, behavior, or validated educational constructs.

• Implementation architecture crowding out education. The official scope warns that computers as a delivery platform only is insufficient. Detailed architecture should not dominate the paper unless it supports a clear educational contribution.

• Pre-submission review for education research

• If the manuscript is mainly an applied AI decision-support paper with education as one testbed, compare Expert Systems with Applications before choosing Computers & Education; if it is already submitted there, use the Expert Systems with Applications Under Review status guide for portal timing and follow-up thresholds.

If the paper is already in the portal, use the Computers & Education Under Review status guide to interpret the live status label, decide when to follow up, and prepare the reviewer-risk map before a decision arrives.

Last verified: May 2026 against Computers and Education editorial pages.