Diego Gonzalez Campos

• Patsis, Panagiotis; Campos, Diego G. & Scherer, Ronny (2026). Educational inequality across countries: An integrative data analysis of TIMSS 2023 data. Studies in Educational Evaluation. ISSN 0191-491X. 91. doi: 10.1016/j.stueduc.2026.101683. Full text in Research Archive Show summary

  Socioeconomic disparities in educational achievement remain a central challenge for education systems. This study examines cross-national variation in the relation between socioeconomic status (SES) and Grade 8 mathematics achievement, using data from TIMSS 2023. Employing an integrative data analysis approach, we estimated within-school, between-school, and contextual SES effects for 43 countries and applied random-effects meta-regression to examine system-level moderators. Results show that SES is a consistent predictor of achievement within schools, while between-school SES effects were substantially larger, indicating pronounced compositional inequalities. The pooled contextual SES effect was large (ES2 = 0.85, 95% CI [0.74, 0.95]) and highly heterogeneous across countries (τ² = 0.11; Q(42) = 642.40, p < .001), ranging from negligible in Nordic systems to very large in more stratified systems. Random-effects meta-regression analyses revealed that income inequality and between-school segregation strongly predicted contextual SES effects, whereas higher levels of democratic governance were associated with weaker effects.

• Fütterer, Tim; König, Lars; Campos, Diego Gonzalez; Scherer, Ronny; Zitzmann, Steffen & Hecht, Martin (2026). Fast-Track Your Abstract Screening: Mastering ASReview for Accelerating Abstract Screening and Evaluating Decisions From Automatic-Screening Methods. Advances in Methods and Practices in Psychological Science (AMPPS). ISSN 2515-2459. 9(2), p. 1–36. doi: 10.1177/25152459261442150. Full text in Research Archive Show summary

  Research syntheses, such as systematic reviews and meta-analyses, are crucial for synthesizing research to support evidence-based decision-making. However, the abstract-screening phase, during which researchers evaluate titles and abstracts for inclusion, is highly time-consuming and often results in cognitive biases and fatigue. To address these challenges, machine-learning-assisted tools, particularly those using active learning, have gained prominence. One such tool is Active Screening Review (ASReview), an open-source software for semiautomating title and abstract screening in systematic reviews. ASReview incorporates user feedback to prioritize relevant studies, reducing screening time and improving efficiency. Despite its potential, many researchers remain uncertain about integrating ASReview into their workflows and making evidence-based decisions regarding the tool’s configuration, training, and stopping criteria. In this tutorial, we provide a step-by-step guide to using ASReview, including practical examples from psychological research. We demonstrate the software’s application in two use cases: screening unlabeled abstracts using active learning and verifying results from automated-screening methods. In the tutorial, we also offer evidence-based recommendations for selecting stopping rules to balance sensitivity and efficiency. We also outline strategies for prescreening, data-set preparation, model setup, and progress monitoring to ensure that researchers can maximize the tool’s benefits while maintaining scientific rigor. By offering evidence-based guidance at each stage of the process for practitioners without coding skills, in this tutorial, we aim to help researchers harness artificial-intelligence-aided screening to enhance the quality and efficiency of research syntheses across disciplines.

• Scherer, Ronny; Campos, Diego G.; Fütterer, Tim & Siddiq, Fazilat (2026). Unraveling gender disparities in teachers’ technological pedagogical and content knowledge: A large-scale meta-analytic review. Educational Research Review. ISSN 1747-938X. 50. doi: 10.1016/j.edurev.2026.100765. Full text in Research Archive Show summary

  This large-scale meta-analytic review investigates gender disparities in pre- and in-service K-12 teachers' Technological Pedagogical and Content Knowledge (TPACK) self-efficacy and performance. Synthesizing 915 effect sizes from 102 primary studies and 420 teacher samples (N = 681,745 teachers) in 70 countries via multilevel meta-analyses, we found small but significant gender differences favoring male teachers in TPACK self-efficacy (d = −0.17), particularly in technological knowledge (d = −0.31) and technological pedagogical knowledge (d = −0.14). However, no significant gender disparities were observed in TPACK performance (d = −0.08). Our findings revealed considerable heterogeneity in these effects across samples, studies, countries, measures, and over time, which could be partly explained by factors such as educational level, country-level gender inequality, and economic development. These results do not support male teachers’ advantage in all aspects of technology-related teaching competencies. Our meta-analytic review further implies that gender disparities in TPACK vary across contexts or over time.

• Fütterer, Tim; Campos, Diego Gonzalez; Gfrörer, Thomas; Lavelle-Hill, Rosa; Murayama, Kou & Scherer, Ronny (2025). AI tools for systematic literature reviews and meta-analyses in educational psychology: An overview and a practical guide. Learning and Individual Differences. ISSN 1041-6080. 126. doi: 10.1016/j.lindif.2025.102849. Full text in Research Archive Show summary

  The rapid growth of research literature has made systematic reviews and meta-analyses increasingly time-consuming, limiting their utility in fast-evolving fields such as educational psychology. Artificial intelligence (AI) tools have enormous potential to streamline these processes, yet their adoption remains limited due to usability issues and a lack of systematic guidance. Out of 282 tools that we compiled from overviews that listed AI tools for research syntheses, we filtered a subset of 7 AI tools that met quality standards, such as transparency and accessibility. These tools were evaluated for their potential to support systematic reviews and meta-analyses in educational psychology. Our review highlights the tools' strengths, limitations, and ethical considerations for their responsible use by providing practical guidance and coding information.

• Campos, Diego Gonzalez; Koskinen, Antti; Munkácsy, Balázs; Rolfe, Victoria; Patsis, Panagiotis & Tóth, Edit [Show all 12 contributors for this article] (2025). Educational inequalities in Europe: A scoping review of longitudinal studies in K-12 education. Studies in Educational Evaluation. ISSN 0191-491X. 87. doi: 10.1016/j.stueduc.2025.101523. Full text in Research Archive Show summary

  Educational inequalities are a critical challenge. Understanding how these disparities emerge, evolve, and can be addressed requires longitudinal evidence. However, most existing research syntheses rely on cross-sectional data. In this scoping review, we map longitudinal research on educational inequalities among K–12 students in Europe. Drawing on 161 studies published between 1986 and 2024, we examined how disparities related to socioeconomic status, migration or ethnic background, gender, geographical location, or their intersections have been studied over time. Academic achievement and socioeconomic status were the most studied outcome and inequality dimensions, while inequalities in socio-emotional outcomes and geographical location received less attention. Four research clusters were also identified in the literature: individual psychological determinants, mechanisms of inequality, developmental patterns of inequality, and structural determinants of educational transitions. Overall, the findings underscore the heterogeneity of empirical approaches to longitudinal research on educational inequality in Europe and highlight gaps for future research.

• Scherer, Ronny & Campos, Diego Gonzalez (2025). Meta-analyzing correlation matrices in the presence of hierarchical effect size multiplicity. Research Synthesis Methods. ISSN 1759-2879. 16(6), p. 828–858. doi: 10.1017/rsm.2025.10027. Full text in Research Archive Show summary

  To synthesize evidence on the relations among multiple constructs, measures, or concepts, meta-analyzing correlation matrices across primary studies has become a crucial analytic approach. Common meta-analytic approaches employ univariate or multivariate models to estimate a pooled correlation matrix, which is subjected to further analyses, such as structural equation modeling. In practice, meta-analysts often extract multiple correlation matrices per study from various samples, study sites, labs, or countries, thus introducing hierarchical effect size multiplicity into the meta-analytic data. However, this feature has largely been ignored when pooling correlation matrices for meta-analysis. To contribute to the methodological development in this area, we describe a multilevel, multivariate, and random-effects modeling approach, which pools correlation matrices meta-analytically and, at the same time, addresses hierarchical effect size multiplicity. Specifically, it allows meta-analysts to test various assumptions on the dependencies among random effects, aiding the selection of a meta-analytic baseline model. We describe this approach, present four working models within it, and illustrate them with an example and the corresponding R code.

• König, Lars; Zitzmann, Steffen; Fütterer, Tim; Campos, Diego Gonzalez; Scherer, Ronny & Hecht, Martin (2024). An evaluation of the performance of stopping rules in AI-aided screening for psychological meta-analytical research. Research Synthesis Methods. ISSN 1759-2879. doi: 10.1002/jrsm.1762. Full text in Research Archive Show summary

  Several AI-aided screening tools have emerged to tackle the ever-expanding body of literature. These tools employ active learning, where algorithms sort abstracts based on human feedback. However, researchers using these tools face a crucial dilemma: When should they stop screening without knowing the proportion of relevant studies? Although numerous stopping rules have been proposed to guide users in this decision, they have yet to undergo comprehen- sive evaluation. In this study, we evaluated the performance of three stopping rules: the knee method, a data-driven heuristic, and a prevalence estimation technique. We measured performance via sensitivity, specificity, and screening cost and explored the influence of the prevalence of relevant studies and the choice of the learning algorithm. We curated a dataset of abstract collections from meta-analyses across five psychological research domains. Our findings revealed performance differences between stopping rules regarding all perfor- mance measures and variations in the performance of stopping rules across different prevalence ratios. Moreover, despite the relatively minor impact of the learning algorithm, we found that specific combinations of stopping rules and learning algorithms were most effective for certain prevalence ratios of rel- evant abstracts. Based on these results, we derived practical recommendations for users of AI-aided screening tools. Furthermore, we discuss possible impli- cations and offer suggestions for future research.

• Campos, Diego Gonzalez; Fütterer, Tim; Gfrörer, Thomas; Lavelle-Hill, Rosa; Murayama, Kou & König, Lars [Show all 9 contributors for this article] (2024). Screening Smarter, Not Harder: A Comparative Analysis of Machine Learning Screening Algorithms and Heuristic Stopping Criteria for Systematic Reviews in Educational Research. Educational Psychology Review. ISSN 1040-726X. 36(1). doi: 10.1007/s10648-024-09862-5. Full text in Research Archive Show summary

  Abstract Systematic reviews and meta-analyses are crucial for advancing research, yet they are time-consuming and resource-demanding. Although machine learning and natural language processing algorithms may reduce this time and these resources, their performance has not been tested in education and educational psychology, and there is a lack of clear information on when researchers should stop the reviewing process. In this study, we conducted a retrospective screening simulation using 27 systematic reviews in education and educational psychology. We evaluated the sensitivity, specificity, and estimated time savings of several learning algorithms and heuristic stopping criteria. The results showed, on average, a 58% ( SD = 19%) reduction in the screening workload of irrelevant records when using learning algorithms for abstract screening and an estimated time savings of 1.66 days ( SD = 1.80). The learning algorithm random forests with sentence bidirectional encoder representations from transformers outperformed other algorithms. This finding emphasizes the importance of incorporating semantic and contextual information during feature extraction and modeling in the screening process. Furthermore, we found that 95% of all relevant abstracts within a given dataset can be retrieved using heuristic stopping rules. Specifically, an approach that stops the screening process after classifying 20% of records and consecutively classifying 5% of irrelevant papers yielded the most significant gains in terms of specificity ( M = 42%, SD = 28%). However, the performance of the heuristic stopping criteria depended on the learning algorithm used and the length and proportion of relevant papers in an abstract collection. Our study provides empirical evidence on the performance of machine learning screening algorithms for abstract screening in systematic reviews in education and educational psychology.

• Campos, Diego Gonzalez; Cheung, M. W. & Scherer, Ronny (2023). A Primer on Synthesizing Individual Participant Data Obtained From Complex Sampling Surveys: A Two-Stage IPD Meta-Analysis Approach. Psychological methods. ISSN 1082-989X. doi: 10.1037/met0000539. Full text in Research Archive Show summary

  The increasing availability of individual participant data (IPD) in the social sciences offers new possibilities to synthesize research evidence across primary studies. Two-stage IPD meta-analysis represents a framework that can utilize these possibilities. While most of the methodological research on two-stage IPD meta-analysis focused on its performance compared with other approaches, dealing with the complexities of the primary and meta-analytic data has received little attention, particularly when IPD are drawn from complex sampling surveys. Complex sampling surveys often feature clustering, stratification, and multistage sampling to obtain nationally or internationally representative data from a target population. Furthermore, IPD from these studies is likely to provide more than one effect size. To address these complexities, we propose a two-stage meta-analytic approach that generates model-based effect sizes in Stage 1 and synthesizes them in Stage 2. We present a sequence of steps, illustrate their implementation, and discuss the methodological decisions and options within. Given its flexibility to deal with the complex nature of the primary and meta-analytic data and its ability to combine multiple IPD sets or IPD with aggregated data, the proposed two-stage approach opens up new analytic possibilities for synthesizing knowledge from complex sampling surveys.

• Campos, Diego Gonzalez & Scherer, Ronny (2023). Digital gender gaps in Students’ knowledge, attitudes and skills: an integrative data analysis across 32 Countries. Education and Information Technologies: The Official Journal of the IFIP Technical Committee on Education. ISSN 1360-2357. 29, p. 655–693. doi: 10.1007/s10639-023-12272-9. Full text in Research Archive
• Scherer, Ronny & Campos, Diego Gonzalez (2022). Measuring those who have their minds set: An item-level meta-analysis of the implicit theories of intelligence scale in education. Educational Research Review. ISSN 1747-938X. 37. doi: 10.1016/j.edurev.2022.100479. Full text in Research Archive Show summary

  Fixed and growth mindsets represent implicit theories about the nature of one's abilities or traits. The existing body of research on academic achievement and the effectiveness of mindset interventions for student learning largely relies on the premise that fixed and growth mindsets are mutually exclusive. This premise has led to the common practice in which measures of one mindset are reversed and then assumed to represent the other mindset. Focusing on K-12 and university students (N = 27328), we tested the validity of this practice via a comprehensive item-level meta-analysis of the Implicit Theories of Intelligence Scale (ITIS). By means of meta-analytic structural equation modeling and network analysis, we examined (a) the ITIS item-item correlations and their heterogeneity across 32 primary studies; (b) the factor structure of the ITIS, including the distinction between fixed and growth mindset; and (c) moderator effects of sample, study, and measurement characteristics. We found positive item-item correlations within the sets of fixed and growth mindset items, with substantial between-study heterogeneity. The ITIS factor structure comprised two moderately correlated mindset factors ( = 0.63–0.65), even after reversing one mindset scale. This structure was moderated by the educational level and origin of the student sample, the assessment mode, and scale modifications. Overall, we argue that fixed and growth mindsets are not mutually exclusive but correlated constructs. We discuss the implications for the assessment of implicit theories of intelligence in education.

View all works in NVA

• Melby-Lervåg, Monica; Hulme, Charles; snowling, maggie; west, gillian; Crane, claudine & Campos, Diego Gonzalez (2026). The efficacy of early language intervention on reading: an individual participant data (IPD) analysis. Full text in Research Archive
• Campos, Diego Gonzalez (2025). Research Synthesis after the Reproducibility and Open Science Reforms: How do Systematic Reviews and Meta-Analysis Practices in Education Need to Adapt? Full text in Research Archive Show summary

  Reproducibility and open science reforms have become powerful drivers of transparency and rigour in contemporary research, yet systematic reviews and meta-analyses face unique challenges when applied in the social sciences. In this presentation, I will explore the impact of these reforms on research synthesis, identify current barriers to conducting systematic reviews and meta-analyses in an open and reproducible way, and highlight the need for adaptive practices in response to an evolving research landscape. Key findings and takeaways will be highlighted to emphasise the importance of continuing to improve systematic review and meta-analysis methods in the era of reproducibility and open science reforms.

• Campos, Diego Gonzalez (2025). Doing Meta-Analysis: A Meta-Analysis Workshop for the Social Sciences. Full text in Research Archive Show summary

  Meta-analyses are a set of statistical techniques used to combine the results of multiple studies on a specific topic. This workshop will illustrate the steps for synthesising research findings using meta-analytic methods. Specifically, the instructor will focus on identifying the structure of meta-analytic data sets, establishing meta-analytic baseline models, and conducting moderator analyses. Participants will learn about random- and mixed-effects models, heterogeneity, and multilevel meta-analysis and get to know ways of including international large-scale assessment data in the modeling phase of a meta-analysis. The workshop materials will utilise the R package “metafor” and some supplementary R packages. Researchers, statisticians, and students can all benefit from this workshop, which will provide the knowledge and tools to model meta-analytic data.

• Campos, Diego Gonzalez (2025). A Hands-On Introduction to Analyses of Educational Equity Using International Large-Scale Assessment (ILSA) Data. Full text in Research Archive Show summary

  This workshop introduces participants to the use of International Large-Scale Assessments (ILSA) such as PIRLS, TIMSS, and PISA for investigating questions of educational equity. Participants will gain insights into the structure and scope of ILSA data, as well as the opportunities and methodological challenges they present. Aims: • Develop an understanding of the structure, scope, and potential of ILSA datasets for educational equity research • Become familiar with the key design features and measurement frameworks underlying ILSA studies • Gain practical, hands-on experience in accessing, preparing, and analyzing ILSA data with statistical software • Apply appropriate methodological approaches to investigate equity-related questions

• Maurer, Michelle Nina; Campos, Diego Gonzalez; Gowenlock, Anna E.; Norbury, Courtenay Frazier & Melby-Lervåg, Monica (2025). The relationship between socioeconomic factors and oral language in children: A systematic review and meta-analysis. Full text in Research Archive
• Campos, Diego Gonzalez (2024). Approaches to Integrating Large Data into Systematic Reviews and Meta-Analysis in Education. Universitetet i Oslo. Full text in Research Archive Show summary

  Systematic reviews and meta-analyses are crucial in advancing the knowledge base in educational sciences. However, with the increasing amount of literature available, identifying and extracting relevant evidence becomes time-consuming and resource-intensive. Furthermore, the diversity of research techniques, study designs, and reporting practices in the primary literature creates new challenges for analyzing and synthesizing study results. Machine learning and two-stage individual participant data meta-analysis can help reviewers cope with the information overload by streamlining the identification and screening of studies and standardizing the synthesis of quantitative information from large-scale studies. In this doctoral thesis, I explore how these methodological approaches assist reviewers in collecting, analyzing, and integrating information from large-scale data sources in two critical stages of a systematic review and meta-analysis: (a) screening and identifying studies and (b) analyzing and synthesizing study results. This thesis is divided into two main parts. The first part is an extended abstract that explains the background, purpose, and significance of the thesis. This extended abstract is made up of five chapters. Chapter 1 provides an overview of the background, objectives, and research articles. Chapter 2 discusses the theoretical framework, including systematic reviews and metaanalysis, large and big data, and innovative approaches to literature search and data analysis. Chapter 3 outlines the methodology used in the research articles. Chapter 4 presents the main findings of the articles, and Chapter 5 highlights the primary contributions of this thesis. The second part of this thesis is the three co-authored articles. The articles aimed to investigate methodological approaches to cope with the information overflow in the screening, analyses, and integration of large-scale sources of information. Article 1 addresses the challenges of time-consuming and resource-intensive systematic reviews and meta-analyses by empirically testing the performance of active learning algorithms for abstract screening. Article 2 deals with the problems of heterogeneity in study design from large-scale educational assessments by investigating and illustrating a methodological framework to analyze and synthesize individual participant data from large-scale assessments. Finally, article 3 presents an empirical example that analyzes and synthesizes individual participant data from international large-scale educational datasets to investigate the gender gaps in digital competence and the extent to which gender gaps in digital skills and attitudes toward technology are connected. Overall, the findings presented in this thesis provide researchers with methodological approaches to deal with information overload during the screening, analysis, and synthesis of large-scale data sources in systematic reviews and meta-analyses. Machine learning algorithms with heuristic stopping rules expedite the abstract screening process in systematic reviews in education and educational psychology. Similarly, the two-stage individual participant data meta-analysis approach contributes to harmonizing and standardizing the synthesis of quantitative information from large-scale studies. Researchers can use machine learning algorithms and two-stage individual participant data meta-analysis to deal with the growing evidence and the diversity of research techniques, study designs, and reporting practices in the primary literature.

View all works in NVA