On this page

Synthesis: Janegizova, Mankesh and Taubaeva (2026) map the research literature on digital and media literacy in pre-service early childhood teacher preparation and find a field organized around one idea: a teacher's capacity to use digital technology well. Media literacy sits beside that center rather than inside it, and AI literacy barely registers at all. The authors' useful move is not the map but what they build on it — a continuity model in which digital competence, media literacy and AI literacy become layers of a single professional digital judgment, offered explicitly as a proposition their data cannot test.

Key Findings

  1. The field's center is tool use, not judgment. Digital competence dominates the keyword network and the most-cited studies, and it is usually defined as the ability to select and use technology in teaching — a capacity that says little about what a four-year-old should encounter.
  2. Media literacy is cited more than it is studied. It attaches to teacher education and digital literacy but stays thin and internally underdeveloped, so a program looking for research on teaching children to read moving-image media will find little to follow.
  3. AI barely registers. AI literacy appears in a couple of keyword entries and never reaches the co-occurrence network; the one highly cited AI study asks what predicts attitudes toward AI education, not what teachers can do with it.
  4. Enthusiasm outruns capability. The best-cited studies report positive attitudes alongside moderate skills and insufficient knowledge, with digital content creation — the work closest to designing a learning experience — the weakest area.
  5. Programs teach the competence and skip the ethics. In Spanish teacher-education study plans in the corpus, digital competence appeared in most programs but the ethical dimension in only a quarter of their teaching guides.
  6. Candidates and programs disagree about readiness. Pre-service teachers called themselves unprepared to use technology in preschool practice; their teacher educators thought the support on offer was adequate. Two perceptions, with no observed practice between them.

What a citation map can and cannot tell a program

A bibliometric map answers what a field has organized itself around and what it has left at the edges — useful for curriculum work, because it shows where a research base exists and where a program will be improvising. It says nothing about study quality: citation counts reward visibility, and a widely cited paper can rest on self-reported confidence as easily as on an experiment. The map is also small and lopsided. The corpus is 86 publications, Spain accounts for a third of them, and fewer than one in ten is co-authored across countries. Early childhood practice is culturally specific — ratios, routines, what counts as a suitable app — so a competence model built inside one system does not travel without translation. Read the growth rate as activity in the field, not as accumulation of evidence.

Three strands that do not add up to one competence

The paper's real contribution is the reading it puts on its own network. Digital competence, media literacy and AI literacy are not rival claims on syllabus time, the authors argue, but layers of one capacity, each preserving what the previous one built: use technology well, judge the media a child meets critically, then reason about what a generative system is doing and whether it belongs in the room. For early childhood the argument is strongest, because the adult mediates almost everything a young child encounters. Judgment there is developmental before it is technical — whether a four-year-old can use a tool without the tool using the interaction, what a generated image teaches about bodies or families, whether a system should be listening at all. The authors are careful about status, and the page should be too: the continuity model is a proposal drawn from the shape of the literature, not a finding, and no study in the corpus tests the sequence.

What the best-cited studies actually show

Content analysis of the ten most-cited papers supplies the human detail. The recurring result is a gap between disposition and capability: candidates like the technology, rate themselves moderately skilled, and know less than they think. One training program did move participants from the lowest competence profiles toward the higher ones, but it measured self-perceived competence and reported nothing about what those teachers later did with children. Where the studies examine skills, content creation is weakest. Where they examine programs, the ethics thin out. The most instructive finding is the disagreement: candidates said they were unprepared, their educators said preparation was adequate, and both were reporting an impression. A program that wants to know which is right has to look at what a candidate does with a resource, a child and a constraint.

What this means for practice

  • Teacher educators: teach the judgment layer explicitly — choosing and critiquing a resource for a specific age group, deciding what a tool may record — rather than adding another software walkthrough to a competence module.
  • Program designers: audit where the ethical dimension actually sits in course guides, not only in program aims.
  • Faculty developers: stop measuring training with confidence surveys. Ask candidates to evaluate a generated storybook, plan a media discussion, or justify refusing a tool, and score the artifact rather than the feeling.
  • Researchers: build a shared instrument for AI literacy in early childhood preparation, and test the continuity model instead of assuming it.
  • Administrators: treat the field's growth as activity rather than evidence. A small, nationally concentrated literature will not settle a procurement or curriculum question on its own.

Limitations

  • Two databases means journals indexed elsewhere and non-English work are largely invisible — a real constraint for a field this nationally concentrated.
  • Bibliometric indicators measure productivity and citation patterns, not study quality, and the content analysis covers only the ten most-cited papers.
  • The window closes at 2025, so the most recent AI-related work is not represented.
  • Keyword co-occurrence reflects authors' and indexers' vocabulary choices; the thematic coding was done without a reported reliability check, and the continuity model is untested.

Citation

Janegizova, U., Mankesh, A., & Taubaeva, G. (2026). Digital and media literacy of pre-service early childhood teachers in the context of digital education: a bibliometric analysis of research trends. Frontiers in Education, 11. https://doi.org/10.3389/feduc.2026.1986795

Embed this page

Copy the code below to embed this page in a learning management system or other website. The embedded version shows only the page content — it hides the site header, navigation, and footer.