Data gaps in Education

Representative open questions

Showing 30 of 625 — one per source paper, highest-quality first.

• Sex Differences in Mathematics: A Meta-analytic Review Across Content, Grades, and Geographic Regions (2026) · doi

  The GGGI scores of countries with lower scores were underrepresented in the sample, making it difficult to fully disambiguate between the role of GGGI and other cultural factors.

• Power-Optimized AI-Enhanced Telepresence Robots: A Validated Multi-Modal Framework for Sustainable Remote Learning in Higher Education (2026) · doi

  The 315-participant sample provided cross-sectional engagement and usability data but lacks longitudinal assessment; sustained participation rates, learning outcome retention, and adaptive AI effectiveness over extended deployment periods (semester-long or academic-year-long studies) are unmeasured.

• A systematic review of generative artificial intelligence in education: Pedagogical impacts, ethical risks, and future directions (2026) · doi

  Administrative applications using AI chatbots and virtual assistants for admissions, enrollment, and academic advising claim to enhance efficiency and reduce operational costs, but lack specific metrics comparing processing time, user satisfaction, and error rates for AI-automated versus human-handled student requests across institutions with different enrollment scales.

• Ethical challenges of artificial intelligence in education: A systematic literature review on bias, privacy, and academic integrity (2026) · doi

  Speech recognition systems struggle with regional accents, multilingual learners, and speech impairments in language learning applications, but the paper does not specify what inclusive dataset composition (percentage representation of accents, languages, speech disorders) or fairness-aware training procedures should be implemented to achieve equitable performance across these user populations.

• Error Patterns and Predictors in Solving Algebraic Word Problems: Evidence from Secondary School Students (2026) · doi

  The study identifies reading errors as the least common (5-7%) despite evidence that proficient readers misinterpret critical algebraic terms; however, it does not investigate whether reading error rates differ when linear equation word problems incorporate domain-specific terminology versus everyday language, or test targeted reading comprehension interventions.

• ChatGPT in computer programming education: A review of current literature and applications (2026) · doi

  Research into ChatGPT adoption in programming education is concentrated in developed regions, with under-representation from Africa, the Middle East, and South America. Comparative studies examining regional differences in access, perception, and pedagogical integration are needed to reveal cross-cultural adoption patterns and inform localised AI system customisation.

• Coding Science: The Role of Block-Based Programming Activities in Enhancing Computational Thinking and Science Achievement (2026) · doi

  Targeted subgroup analyses are needed to examine how gender, prior science achievement, and previous technology exposure mediate the impact of block-based programming integration on computational thinking and science achievement, as aggregate group-level reporting may mask individual differences.

• When teachers learn by observation: Vicarious reinforcement and student response systems in teacher education (2026) · doi

  The paper identifies that English language barriers exist when students use written communication through SRS platforms (given Arabic is their first language), but does not investigate how language mode effects (verbal vs. written responses in student response systems) impact content understanding and vicarious learning outcomes in multilingual teacher education contexts.

• A(n) (re)awakening: preservice teachers connecting their culture to science teaching along the US–México border (2026) · doi

  Longitudinal studies are needed that follow preservice teachers from their teacher preparation programs through their induction years to investigate how they maintain culturally relevant science instruction practices when navigating institutional constraints and administrative pressure to comply with traditional curricula in K-12 settings.

• Digital Platforms and Relational Practice: A Netnographic Study of Educator-Family Communication (2026) · doi

  Only three of five interviewed families reported using digital platform posts as bridges to build intersetting knowledge connecting online updates with real-world conversations. Research is needed to identify which characteristics of educator posts (specificity, visual content, activity suggestions, learning domain) most effectively function as bridges between digital and physical settings in educator-family communication, and whether post characteristics vary by child age group in ECEC contexts.

• When faces trigger feelings: multi-stakeholder experiences, use, and expectations of facial-emotion learning technologies and materials for autistic people (2026) · doi

  The paper documents that some autistic children refuse to engage with emotions labeled as negative (anger, sadness) in facial-emotion apps because the emotional displays provoke anxiety or distress. Research is needed to determine which specific emotions and facial expressions trigger rejection versus acceptance in autistic populations, and how to design curricula that safely introduce challenging emotions.

• Segmental and suprasegmental pronunciation training via AI: An exploration of university students’ perceptions and attitudes (2026) · doi

  The paper examines university students' perceptions of AI-based pronunciation training tools, but does not compare effectiveness across different proficiency levels (beginner, intermediate, advanced) or investigate whether segmental and suprasegmental training outcomes differ by initial English proficiency of learners.

• On the development and potential influential aspects of prospective science teachers’ competencies for the use of explainer videos for learning in physics education (2026) · doi

  Content knowledge (CK) related to electrical circuits was not collected in this mixed-methods study despite empirically proven positive relationships between CK and pedagogical content knowledge (PCK); a follow-up study should include CK measures to explain variance in prospective teachers' professional vision of learning support features.

• Producing concept-motivated signs supports learning of STEM in American Sign Language (2026) · doi

  Participants' ASL age of acquisition ranged widely (see Table 3) but the analysis does not examine whether age of ASL acquisition or native versus non-native ASL status moderated the effects of concept-motivated versus engagement-motivated signing on STEM learning outcomes, despite acknowledging variation in language backgrounds across the sample.

• Adequacy and Utilization of Physical Resources: Implications for Effective Teaching and Learning (2026) · doi

  The paper documents that inadequacy stems from insufficient government funding, poor maintenance, and mismanagement, but provides no empirical quantification of how much each factor contributes to resource shortfalls in Delta State schools. Research should conduct cost-benefit analysis and resource allocation audits to determine whether prioritizing maintenance improvement versus infrastructure expansion would yield greater learning outcome gains.

• Digital Disruption of Academic Integrity (2026) · doi

  The gap between plagiarism awareness and plagiarism practice among postgraduate students remains unexplained; research is needed to investigate whether this knowledge-ethics behavior reconciliation gap varies by discipline, institutional plagiarism policy stringency, or student demographic characteristics in digital scholarship environments.

• The Role of Artificial Intelligence in the Career Expectations of Ukrainian Students: Implications for Higher Education (2026) · doi

  No longitudinal data exists tracking how Ukrainian students' attitudes toward AI evolve over time in relation to technological advancements, educational program changes, and acquisition of practical professional experience. Longitudinal studies must be designed to examine the dynamics of ChatGPT perception and career expectations as students gain actual hands-on experience with AI tools and progress through their academic programs.

• The Effects of Handwriting and Typing on Chinese Character Learning in the Digital Age: Evidence from Arab Adolescent Learners (2026) · doi

  The paper recommends integrating gamified typing activities and game-based learning apps for character instruction but provides no empirical data on the specific game design parameters (difficulty progression, feedback mechanisms, reward schedules) that maximize long-term retention versus perceptual memory in adolescent learners.

• No one-size-fits-all: a study of prompt techniques and large language models to enhance AI’s mathematics educational quality (2026) · doi

  The study uses only six problem-solving tasks focused on number and algebra domains. The transferability of findings regarding prompt techniques and model architecture effectiveness to geometry, data, and mathematical modelling domains remains unexplored and requires empirical validation.

• Pre-service Teachers' Self-Directed Learning in a Blended Learning Environment: A Study on Scale Development and Affecting Factors (2026) · doi

  The model explains 67.4% of SDL variance, leaving 32.6% unexplained; additional predictors beyond student awareness, technological ability, self-monitoring, and autonomy need identification and measurement, particularly institutional factors (course design structure, instructor training, LMS affordances) that may influence SDL in blended learning contexts.

• Development of User Story and Design Thinking Integration Teaching Model for Software Engineering Education (2026) · doi

  The paper identifies that session time constraints negatively impacted PEOU but does not specify which subsections or DT framework phases (from the d.school framework) were most time-inefficient or which could be reduced without compromising learning outcomes. Future research should conduct detailed time-motion analysis of each US-DT teaching model component to identify specific optimization targets for session restructuring.

• Does “Maniplo-Spatial Arty” Really Work? A Study of the Effectiveness of “Maniplo-Spatial Arty” as a Teaching and Learning in Science Pedagogy (2026) · doi

  The study's evidence for Level 2 effectiveness relies on two weak sources: short-term knowledge reported within lessons or immediate post-lesson interviews, and retrospective learner annotations on previous maniplo-spatial tasks conducted in some cases a year prior. Systematic measurement of retention intervals (e.g., 1 week, 4 weeks, 8 weeks post-lesson) using consistent assessment instruments across multiple maniplo-spatial tasks is absent.

• Quality Evaluation of Indonesian Student-generated User Stories: Insight from Human and ChatGPT Evaluation (2026) · doi

  The study evaluated user stories exclusively in the Indonesian language context with a small number of human evaluators; future research must expand to multilingual and cross-cultural datasets with more diverse evaluator groups to establish whether the identified quality issues (Uniform, Conceptually Sound, Atomic, Ambiguity) generalize across different linguistic and educational contexts.

• Engagement in Exploratory Action Research to Improve Teaching Practice in EFL Writing (2026) · doi

  The writing study plan implementation showed promise but the paper does not disaggregate which components (pre-writing activities, increased writing practice frequency, or reading material integration) contributed most significantly to student writing improvements. Comparative action research examining each component's isolated effect within EFL writing instruction would clarify implementation priorities.

• Mind the gap: Learning the surface forms of movement dependencies (2026) · doi

  Studies on functional elements in infant processing (Höhle et al. 2004, Kim & Sundara 2021) demonstrate sensitivity to morphology and determiners by 6-19 months, but the excerpt does not indicate whether functional morphemes specifically signal filler-gap dependency structure in child-directed speech. Corpus-based analysis of morphosyntactic marking of movement dependencies in naturalistic input to children is needed.

• Washington University in St. Louis School of Medicine (2010) · doi

  The expansion of ambulatory experiences over the past decade is noted, but the paper lacks specific documentation of which subspecialty electives incorporate ambulatory components, what proportion of students participate in primary care electives, and whether differential outcomes exist between students with varying ambulatory versus inpatient experience ratios.

• Research on the Teaching Model of AIGC-Empowered Interdisciplinary Competence Cultivation for Postgraduates in Biomedical Fields (2026) · doi

  The 'Four-Wing Intelligent Integration' teaching model's teacher training includes 'AIGC Tool Workshop' and 'Interdisciplinary Teaching Design Seminar,' but no data is provided on teacher competency assessment outcomes, training duration, or the specific AIGC tool platforms used in the biomedical engineering interdisciplinary curriculum.

• Academic Medicine in a Transformational Time (2015) · doi

  The paper acknowledges increased reliance on international medical graduates in primary care but lacks specific analysis of how many IMG positions could be replaced by domestically trained physicians under different medical school expansion and specialty preference scenarios, or what threshold of IMG dependence creates exacerbated health challenges in source countries.

• The Effect of Videoscribe Media Assisted by the Guided Inquiry Learning Model on Students’ Physics Learning Outcomes (2026) · doi

  The research was conducted on physics students described as 'relatively less independent' learners, but does not stratify results by learner independence levels or test whether the medium effect of Videoscribe-Guided Inquiry is replicated in highly independent, moderately independent, or diverse independence-level student populations.

• John McCutcheon (2020) · doi

  While Pauly's work emphasizes total fish and invertebrate catch as a crucial impact measure, the review does not detail specific protocols for estimating unintended bycatch and discards in small-scale subsistence fisheries where catch reporting is minimal or absent.