-
Research Matters 40 - Foreword
Welcome to this fortieth issue of Research Matters, marking 20 years of publication. A big number anniversary provides an opportunity to reflect on achievements and to speculate on what the future may hold. The foreword from issue 20 considered early Research Matters history. It is a real privilege to pick up 10 years later and take stock of the next decade of published research – and what a decade it has been!
Download
-
Research Matters 40 - Editorial
Our first article explores teachers’ views on the literary canon, analysing teachers’ definitions of “literary canon”, and teachers’ ratings of 22 prose texts for how canonical they were and for their aesthetic quality. Our second article describes research comparing exams where students do or do not have access to a literary text they have studied. Our third article summarises findings from research about the June 2020 student cohorts who were unable to take exams and compares the progression outcomes of these students to those for a pre-pandemic cohort of students. Our fourth article considers the issues around the potential use of automarkers based on Large Language Models (LLMs). Our final article describes analysis of the thinking skills assessed in exam papers for several Cambridge International A Level subjects, with a particular focus on the higher-order skills of problem-solving, metacognition and systems thinking.
Download
-
Wanted dead or alive: Canonical authors for literature in English curricula
Brady, J., Majewska, D., & Greatorex, J. (2025). Wanted dead or alive: Canonical authors for literature in English curricula. Research Matters: A Cambridge University Press & Assessment publication, 40, 7-27. https://doi.org/10.17863/CAM.121934
Our study explored 250 international teachers’ definitions of the English-language literary canon. Questionnaire respondents defined the literary canon, rated the aesthetic value of 22 prose texts, and gave the 22 texts a “canon rating” to indicate the extent to which they thought the text was canonical. We found that teachers find some value in the literary canon, as defined by them, whereas others called for the canon to be expanded or disbanded. We report that perceived aesthetic quality appears to be a dimension of canon, and that authors seem to be admitted to the canon posthumously. The work provides a rare insight into teachers’ definitions of canon – information that helps those who set curricula, and those in the wider education ecosystem, to understand the complex meanings underpinning requests for students to study more or fewer canonical texts.
Download
-
Comparing open-book and closed-book exams in international literature in English assessments
Walland, E. (2025). Comparing open-book and closed-book exams in international literature in English assessments. Research Matters: A Cambridge University Press & Assessment publication, 40, 28-49. https://doi.org/10.17863/CAM.121935
When designing assessments such as examinations, it is important to consider which materials students should be permitted to access during the assessment, for example, whether to allow students access to a relevant book, such as a literary text. Debates among teachers and assessment professionals highlight differing views about the advantages and disadvantages of open-book and closed-book exams, and how each approach could influence the constructs being assessed, student performance, exam preparation and students’ experiences. However, the effects of open- and closed-book exams vary based on how they are implemented and enacted. This small-scale study explored differences between open-book and closed-book exams in international secondary school literature in English qualifications. Variations in the nature, extent, and quality of textual references in samples of essays written under different conditions were explored. Additionally, the research explored a small group of assessors’ perspectives on these differences and analysed the differences in student marks in some regions based on examination format. This study provides valuable insights into the ongoing debate about open- and closed-book exams within this context, informing syllabus development and future research. The findings highlight the complexities and interactions among various assessment principles.
Download
-
Highlights from research on the progression of Key Stage 4 and 5 cohorts impacted by the COVID-19 pandemic
Vidal Rodeiro, C. L., & Lim, C. H. J. ( (2025). Highlights from research on the progression of Key Stage 4 and 5 cohorts impacted by the COVID-19 pandemic. Research Matters: A Cambridge University Press & Assessment publication, 40, 50-71. https://doi.org/10.17863/CAM.121936
The Research Division at Cambridge University Press & Assessment has an ongoing programme of research that tracks the Key Stage 4 and Key Stage 5 cohorts of students impacted by the COVID-19 pandemic. The findings from this work can help inform us regarding whether these students were negatively affected by the pandemic and whether they might require any further support while in education.
In this article, we present a summary of findings from research conducted to date using data from the June 2020 student cohorts, drawn from the National Pupil Database, a longitudinal dataset maintained by the Department for Education in England. We examined various progression outcomes and compared them with those of a pre-pandemic cohort of students to assess whether, and in what ways, these outcomes have changed for students affected by the pandemic. Findings from the research have suggested that the students from the June 2020 cohorts were not disadvantaged in their transitions to post-16 study or to higher or further education. However, there was some evidence suggesting that different subgroups of students had progressed slightly differently. This implies that while overall the pandemic cohorts were not disadvantaged, the actual experience of different groups of students may have differed.
Download
-
The ethics of deploying Large Language Models in high-stakes automarking
Morley, F., & Walland, E. (2025). The ethics of deploying Large Language Models in high-stakes automarking. Research Matters: A Cambridge University Press & Assessment publication, 40, 72-92. https://doi.org/10.17863/CAM.121937
The recent development of Large Language Models (LLMs) such as Claude, Gemini, and GPT has led to widespread attention on potential applications of these models. Marking exams is a domain which requires the ability to interpret and evaluate student responses (often consisting of written text), and the potential for artificial intelligence (AI) tools based on LLMs to contribute to the marking process has been noted and researched.
This article imagines three scenarios where LLM-based automarking is applied to a high-stakes exam context, and examines three ethical issues identified based on current LLM technology as of June 2025 (Aloisi, 2023; Morley et al., 2025). These scenarios allow us to discuss: (1) explainability, whether automarking decisions can be explained; (2) bias, whether automarkers could make biased decisions disadvantaging certain demographic groups; and (3) adversarial attacks, how vulnerable automarkers might be to actors exploiting their vulnerabilities in order to produce a higher mark. Within the scenarios, we compare automarkers to human examiners in the context of well-functioning human marking in high-stakes settings (based on research regarding exam board marking in England). This allows us to explore ethical dimensions of automarkers based on what we know about automatic and human marking. In particular, this article argues that human examiners have a higher potential for trustworthiness over LLM-based automarkers.
Download
-
Filled with skills: An analysis of four International AS and A Levels
Suto, I., Nelson, S., Roberts, J., Sidhu, A., & Spence, L. (2025). Filled with skills: An analysis of four International AS and A Levels. Research Matters: A Cambridge University Press & Assessment publication, 40, 93-120. https://doi.org/10.17863/CAM.121938
A holistic education should nurture a range of skills that are essential for thriving in a fast-changing world, including higher-order thinking skills. These skills are deeply intertwined with knowledge and many in the assessment community agree they should not be assessed in isolation. Curricula structured around long-established subject disciplines and assessed via written examinations are often considered “knowledge rich”. This article reports on how they also foster a range of higher-order skills that are important to students’ futures.
We conducted a systematic analysis of Cambridge International AS & A Levels in English Language, Geography, Physics, and Psychology. To suites of specimen examination papers, we applied a skills coding framework based on Marzano and Kendall’s (2007) educational taxonomy, which encompasses problem-solving and metacognitive skills among other thinking skills. We incorporated additional codes for systems thinking components from an environmental sustainability framework.
The analysis revealed a broad and rich coverage of higher-order thinking skills, as well as lower-order thinking skills such as retrieval, with variations across subjects. For example, while systems thinking is not mentioned in formal assessment objectives, it is present in Geography, Physics, and Psychology examinations. Although we did not explore this comprehensively, the study also suggested the examinations demand exam techniques that include metacognitive and problem-solving skills, further demonstrating higher-order thinking skills’ integration within these qualifications.
Download
-
Research Matters 40 - Research News
Bowett, L. (2025). Research News. Research Matters: A Cambridge University Press & Assessment publication, 40, 121-123.
A summary of recent conferences, reports, blogs and research articles published since the last issue of Research Matters.
Download