Skip to content
Interface language
  1. Wonkhe
    InternationalCommentary
    Topic illustration of a school counselor listening to a studentIllustration

    234 invited, only 16 replies: who is missing from a student-belonging survey?

    A student-success adviser at a UK university reflects on a belonging survey: 16 of 234 invited students responded, approximately 6.84%. Rather than describing the 218 non-respondents as alienated, the author raises possible explanations such as work, caring responsibilities, commuting or mistrust that would need interviews to investigate, and argues for sustained conversations alongside questionnaires.

    Why read: The reflection encourages student-support teams to address non-response bias and design interviews or diary methods that can hear from students who remain silent.

  1. UNESCO
    InternationalResearch
    Topic illustration of education researchers examining charts and notesIllustration

    UN agencies report that more than half of Afghanistan's primary-school-age children are outside formal education

    A new UNESCO and UNICEF report says 53% of Afghanistan's primary-school-age children are outside formal general education and describes restrictions on girls' education and problems with learning quality. The material checked so far is the agencies' press release; the underlying survey methods still need to be checked against the full report.

    Why read: A major new report addressing both access to education and learning quality.

  1. Tes Magazine
    InternationalResearch
    Source image: Mid-year moves between attainment groups in UK schools: observational findings do not show that moving caused a declineSource

    Mid-year moves between attainment groups in UK schools: observational findings do not show that moving caused a decline

    Tes interviews a research team about an observational analysis of more than ten thousand pupils in 21 secondary schools belonging to one school group. Moving to a higher set mid-year was not followed by a clear improvement in attainment in most subjects. The researchers explicitly caution against concluding that the move caused the outcome.

    Why read: Independent education reporting that specifies the sample and methodological limits can help schools examine grouping decisions.

  1. Education Week
    InternationalResearch
    Source image: Education Week Investigates AI Education Reviews: Study Populations Are Mixed in Large-Effect ConclusionsSource

    Education Week Investigates AI Education Reviews: Study Populations Are Mixed in Large-Effect Conclusions

    Education Week interviewed researchers about an audit of 14 meta-analyses on AI in education, which found methodological problems such as inconsistent definitions of interventions and outcomes. The report does not reject all trials of AI in teaching; the audit paper itself still needs to be examined separately for its methods.

    Why read: Independent interviews and comparisons on the quality of evidence about AI in education, addressing questions left unanswered by product and institutional press releases.

  2. The Hechinger Report
    InternationalResearchPreprint · not peer reviewed
    Source image: Why Did Two AI Studies Find a Positive Association in One Case and a Negative Trial Effect in the Other?Source

    Why Did Two AI Studies Find a Positive Association in One Case and a Negative Trial Effect in the Other?

    A reporter juxtaposes two October working papers involving the same researcher: use of a middle-school math platform in Arizona was positively associated with MAP scores, while a university randomized students by instructor to receive a course AI assistant and found lower grades and engagement with the learning platform. The tools, educational levels, and comparison methods differ, so the findings cannot directly cancel each other out or be generalized as a single overall effect of AI.

    Why read: Use the contrasting results to practice reading research: first distinguish an association with use from randomized assignment, then examine implementation conditions, engagement, and outcome measures.

  1. World Bank
    InternationalResearch
    Topic illustration of education researchers examining charts and notesIllustration

    World Bank Uses Implementation Science to Track Education Reform, Distinguishing Reach from Learning Outcomes

    The World Bank describes a remedial-teaching model piloted in 400 Rwandan schools, along with related baseline and midline surveys, and notes that learning outcomes still require future evaluation. A separate teacher-training project has reached many schools, but the number reached cannot substitute for improvements in student skills.

    Why read: Clearly illustrates the distinction between evidence about implementation and evidence about outcomes.

  2. Education Endowment Foundation
    InternationalResearch
    Source image: UK Randomized Trial: AI Lesson-Planning Tool Saves Time, but Willingness to Use It VariesSource

    UK Randomized Trial: AI Lesson-Planning Tool Saves Time, but Willingness to Use It Varies

    EEF reports an NFER randomized trial involving 464 teachers at 108 schools: the Aila group spent about 76% as much time planning as the control group on average, but use declined over ten weeks. The study measured planning time and resource quality; it does not show that student achievement improved.

    Why read: An empirical case for distinguishing tool efficiency, adoption behavior, and learning outcomes.

  1. OECD
    InternationalResearch
    Topic illustration of education researchers examining charts and notesIllustration

    OECD Education at a Glance Focuses on Teacher Shortages: Cross-National Averages Need to Be Read in Context

    The OECD's 2026 Education at a Glance highlights teacher shortages, reporting that about 9.1% of primary and secondary teachers in the sampled countries were not fully qualified. The indicator is comparable across countries, but national definitions, missing values, and reference years differ; the average cannot be used directly to assess a particular country's schools.

    Why read: A foundational data entry point for comparing education systems internationally.

  1. Education Endowment Foundation
    InternationalResearch
    Source image: Large-Scale UK Trial: Mathematics Reasoning Program Yields an Average Gain Equivalent to About One MonthSource

    Large-Scale UK Trial: Mathematics Reasoning Program Yields an Average Gain Equivalent to About One Month

    EEF reports a randomized trial involving 6,291 Year 2 pupils in 238 schools. On average, the mathematics reasoning program group made progress equivalent to about one month more than the usual-teaching group. The “month” is an EEF equivalence metric; it does not mean each student actually received one additional month of instruction.

    Why read: A classroom intervention trial with an independent evaluation and a relatively large sample.

  1. UNICEF
    InternationalResearch
    Source image: UNICEF Estimates Climate Disasters Disrupted Learning for More Than 171 Million Students in 2025Source

    UNICEF Estimates Climate Disasters Disrupted Learning for More Than 171 Million Students in 2025

    In an analysis published in September 2026, UNICEF estimated that climate disasters disrupted the education of more than 171 million students worldwide in 2025. This is a retrospective estimate, not a count of disruptions that have already occurred in 2026; the numbers affected by different disasters cannot simply be added together.

    Why read: Climate risk is an important global issue for continuity of education.

  1. Education Endowment Foundation
    InternationalResearch
    Source image: Schools in England Report: Literacy and Attendance Are the Main Barriers for Disadvantaged PupilsSource

    Schools in England Report: Literacy and Attendance Are the Main Barriers for Disadvantaged Pupils

    An EEF-commissioned analysis of more than 1,200 Pupil Premium strategy statements from schools in England found that 91% mentioned literacy and English issues and 83% mentioned attendance. It reflects challenges identified by schools in their documents; it is not an independent measure of students' skills or absences.

    Why read: Shows how schools define problems and reminds readers to distinguish self-reports from outcomes.

  1. Journal of East China Normal University
    ChinaResearch
    Topic illustration of education researchers examining charts and notesIllustration

    Longitudinal Survey Examines the Limits of Gaokao Identification: The University Development Process Still Matters

    Using a longitudinal survey of students at Double First-Class universities, the authors argue that scores offer an initial signal for identification, but have limits in identifying broader competencies; university courses, student–faculty interaction, and participation also shape development. The abstract cannot establish causal effects across all universities.

    Why read: Brings selection and subsequent development into the same research question, avoiding a discussion limited to exam scores.

  1. Economic Policy Institute Education
    InternationalResearch
    Topic illustration of education researchers examining charts and notesIllustration

    The U.S. teacher pay gap remains large: how was 25.2% calculated?

    The Economic Policy Institute used Bureau of Labor Statistics data to estimate that in 2025, weekly wages for U.S. public school teachers were 25.2% below those of comparable college graduates. After employer-paid benefits are included, the estimated total compensation gap was 14.5%. These are adjusted differences between occupational groups; they do not show the causal gain an individual would receive by changing careers.

    Why read: A closer look at an older report’s methods: how wage, benefit and geographic definitions shape discussion of the teacher workforce. The estimates can be compared with recruitment and retention data.

  1. Journal of East China Normal University
    ChinaResearch
    Topic illustration of education researchers examining charts and notesIllustration

    Survey of 3,483 teachers finds an association between time pressure and knowledge sharing

    The study analyzed 3,483 questionnaire responses from primary and secondary school teachers using structural equation modeling to examine knowledge sharing and its associations with sharing capacity, collaborative lesson planning and time pressure. The survey model describes patterns of association; by itself, it cannot prove that reducing workload will increase knowledge sharing.

    Why read: The findings offer a mechanism to examine in the organization of school-based professional learning.

  1. Journal of East China Normal University
    ChinaResearch
    Topic illustration of education researchers examining charts and notesIllustration

    Comparing teaching texts from six large language models: strength in higher-order questions does not mean classroom effectiveness

    The study compared multidisciplinary teaching texts generated by six large language models with teachers’ instructional materials, using three dimensions: value guidance, knowledge construction and thinking development. The objects of evaluation were texts, not student learning outcomes; the findings cannot establish that large language models teach better than teachers in classrooms.

    Why read: Provides a concrete research design and clear boundaries around what is evaluated in discussions of large language models for education.

  1. Journal of East China Normal University
    ChinaResearch
    Topic illustration of a school counselor listening to a studentIllustration

    An education ecosystem index seeks to measure equity, burden and student development together

    The paper constructs an index with five dimensions and reports a 2024 assessment based on 10,651 household-survey responses across 31 provincial-level regions. This is an evaluation framework proposed by the authors; scores and regional clusters depend on the indicators, weights and sampling design and should not be treated as an official ranking.

    Why read: Adds to discussion of education assessment beyond a single examination score.

  1. OECD
    InternationalResearch
    Topic illustration of a teacher and students learning with digital tools and booksIllustration

    OECD Digital Education Outlook: better assignment performance with generative AI does not necessarily mean learning

    The OECD’s 2026 Digital Education Outlook discusses the use of generative AI in teaching, learning and governance. It cautions that when a task is handed directly to AI, assignment performance may improve without a corresponding gain in learning. The report offers policy and research recommendations; it does not mean every tool has been evaluated for causal effects.

    Why read: Offers international evidence and a clear distinction between performance and learning for discussions of AI in classrooms.

You have reached the end