About

James S. Kim is the Juliana W. and William Foss Thompson Professor of Education and Society at Harvard Graduate School of Education. His expertise encompasses literacy intervention and experimental design, with a professional mission focused on conducting policy-relevant research aimed at improving literacy outcomes for low-income students and struggling readers at scale. He leads the READS Lab (Research Enhances Adaptations Designed for Scale in Literacy), a research team dedicated to building long-term partnerships with school districts to address significant challenges in literacy and education. Kim's research prioritizes understanding how building students’ domain knowledge and reading engagement can foster long-term improvements in reading comprehension. His work emphasizes open science principles, promoting open data, materials, preregistration, and replication data. He co-developed the Model of Reading Engagement (MORE), a spiraled and sustained content literacy intervention shown to improve reading comprehension among first to third-grade students across various subjects. His research has been recognized for meeting rigorous standards, with long-term follow-up indicating persistent impacts through fourth grade. Kim’s core value is to serve as a servant leader, aiming to inspire all learners to appreciate the written word and to find beauty, goodness, and truth in the natural and social worlds.

Research topics

• Pedagogy
• Mathematics education
• Psychology
• Linguistics
• Computer Science

Selected publications

• Mapping the Mechanisms of Interdisciplinary Learning Transfer from Reading to Math Achievement: Evidence from a Large-Scale Randomized Controlled Trial

  Brown Digital Repository · 2026-01-03

  articleOpen accessSenior author
  PublisherDOI

• Asset‐Based Implementation of Structured Adaptations in an Online Third‐Grade Content Literacy Intervention

  Reading Research Quarterly · 2025-08-21 · 2 citations

  articleOpen accessSenior author

  ABSTRACT Scaling up evidence‐based educational interventions presents challenges, particularly in adapting to new contexts while maintaining fidelity. Structured adaptations that integrate the strengths of experimental science (high fidelity) and improvement science (high adaptation) represent a novel design framework for supporting the equitable implementation of research‐based practices and programs. This preregistered study examined the effectiveness of structured adaptations to a Tier 1 content literacy intervention, Modeling of Reading Engagement (MORE), on Grade 3 students' ( N = 1914) engagement in asynchronous digital app and print‐based reading activities, the quality of synchronous student–teacher interactions during Zoom‐delivered lessons, and student learning outcomes during the COVID‐19 school closures. Using a cluster randomized trial design, 95 teachers and their students in 26 elementary schools were randomly assigned to either a high‐fidelity core treatment or a structured adaptation condition. In the latter, teachers participated in Team‐Based Learning activities that tightly coupled knowledge acquisition and application, with a focus on improving student engagement. Students in the structured adaptations condition outperformed students in the core treatment condition on science reading comprehension (ES = 0.07) and science background knowledge (ES = 0.09). Implementation analyses of the synchronous lessons indicate that the structured adaptations also improved norms of social interaction between students and teachers, resulting in stronger engagement, better feedback, and dialogic questioning. These findings suggest the importance of intentionally building in opportunities for teachers to adapt instruction within a clear framework. When teachers have support to work together and combine research‐based strategies with classroom knowledge, they can more effectively engage students and improve learning outcomes.

  PublisherOA PDFDOI

• Estimating causal effects on psychological networks using item response theory.

  Psychological Methods · 2025-06-02 · 5 citations

  articleOpen accessSenior author

  Network models in which each variable interacts with the others in a complex system have emerged as an important alternative to latent variable models in psychometric research. However, confirmatory methods for group network comparison can be limited by practical constraints, such as the computational intractability of the Ising model in large networks. In this study, we demonstrate how to estimate causal effects on network state and strength when direct network estimation is not feasible by leveraging the mathematical equivalencies between the Ising model and item response theory (IRT) models. We demonstrate through simulation that a two-parameter logistic explanatory IRT model can simultaneously recover causal effects on network state and strength. We first apply the method to a single empirical example of a vocabulary assessment from a content literacy intervention to demonstrate model building and interpretation strategies. We then replicate our approach with 72 empirical data sets from randomized controlled trials with item-level outcome data in education, economics, health, and related fields. Our results show that causal effects on network strength are both common and uncorrelated with effects on network state, suggesting that causal network models can provide new insight into the impact of interventions in the social and behavioral sciences. (PsycInfo Database Record (c) 2026 APA, all rights reserved).

  PublisherDOI

• Estimating Causal Effects on Psychological Networks Using Item Response Theory

  2025-03-20 · 4 citations

  preprintOpen accessSenior author

  Network models in which each variable interacts with the others in a complex system have emerged as an important alternative to latent variable models in psychometric research. However, confirmatory methods for group network comparison are limited by practical constraints, such as the computational intractability of the Ising model in large networks. In this study, we demonstrate how to estimate causal effects on network state and strength when direct network estimation is not feasible by leveraging the mathematical equivalencies between the Ising model and item response theory (IRT) models. We demonstrate through simulation that a two-parameter logistic (2PL) explanatory IRT model can simultaneously recover causal effects on network state and strength. We first apply the method to a single empirical example of a vocabulary assessment from a content literacy intervention to demonstrate model building and interpretation strategies. We then replicate our approach with 72 empirical datasets from randomized controlled trials with item-level outcome data in education, economics, health, and related fields. Our results show that causal effects on network strength are both common and uncorrelated with effects on network state, suggesting that causal network models can provide new insight into the impact of interventions in the social and behavioral sciences.

  PublisherOA PDFDOI

• Standing Still: A Case of Stiff Person Syndrome and Common Variable Immunodeficiency

  Cureus · 2025-11-13

  articleOpen access

  Stiff-person syndrome (SPS) is a rare autoimmune neurologic disorder characterized by progressive rigidity and spasms, while common variable immunodeficiency (CVID) features hypogammaglobulinemia and recurrent infections. Their coexistence complicates management by requiring autoimmune suppression without further compromising host defense. A 54-year-old man with CVID, diagnosed based on hypogammaglobulinemia (IgG 386 mg/dL) and recurrent sinopulmonary infections, subsequently developed SPS with progressive truncal and limb stiffness, causing gait impairment and dyspnea. He experienced a severe aspergillus pneumonia, and examination showed finger clubbing and resting hypoxemia. Pulmonary function testing demonstrated moderate airflow obstruction with distal airway involvement. Chest radiography later revealed an elevated left hemidiaphragm, consistent with respiratory muscle dysfunction in SPS. Combined therapy with intravenous immunoglobulin (IVIG) and rituximab was initiated, yielding meaningful improvement in stiffness and a reduction in respiratory infections. After a motor vehicle accident with spinal fusion, the patient reported worsening exertional dyspnea and variable IgG troughs; treatment was transitioned to subcutaneous immunoglobulin (SCIG) 10% with rHuPH20 to maintain steadier IgG levels. This case emphasizes the importance of immunologic evaluation in refractory SPS and demonstrates that combined IVIG and rituximab can provide functional benefit while addressing CVID. Transition to SCIG may stabilize IgG exposure and sustain clinical gains. A timeline of diagnoses and treatments highlights the interplay between autoimmunity and hypogammaglobulinemia and supports a tailored, multidisciplinary strategy.

  PublisherOA PDFDOI

• Time to transfer: Long-term effects of a sustained and spiraled content literacy intervention in the elementary grades.

  Developmental Psychology · 2024-02-26 · 18 citations

  articleOpen access1st authorCorresponding

  = 2,870 students) were randomized to a treatment or control condition. In the treatment condition (i.e., full spiral curriculum), students participated in content literacy lessons from Grades 1 to 3 during the school year and wide reading of thematically related informational texts in the summer following Grades 1 and 2. In the control condition (i.e., partial spiral curriculum), students participated in lessons in only Grade 3. The Grade 3 lessons for both conditions were implemented online during the COVID-19 pandemic school year. Results reveal that treatment students outperformed control students on science vocabulary knowledge across all three grades. Furthermore, intent-to-treat analyses revealed positive transfer effects on Grade 3 science reading (ES = .14), domain-general reading comprehension (ES = .11), and mathematics achievement (ES = .12). Treatment impacts were sustained at 14-month follow-up on Grade 4 reading comprehension (ES = .12) and mathematics achievement (ES = .16). Findings indicate that a content literacy intervention that spirals topics and vocabulary across grades can improve students' long-term academic achievement outcomes. (PsycInfo Database Record (c) 2024 APA, all rights reserved).

  PublisherDOI

• Effects of tier 1 content literacy intervention on early-grade English learners’ reading and writing: Exploring the mediating roles of domain-specific vocabulary and oral language proficiency.

  Journal of Educational Psychology · 2024-07-18 · 9 citations

  article
  PublisherDOI

• Leveraging Item Parameter Drift to Assess Transfer Effects in Vocabulary Learning

  Applied Measurement in Education · 2024-07-02 · 6 citations

  article

  Longitudinal models typically emphasize between-person predictors of change but ignore how growth varies within persons because each person contributes only one data point at each time. In contrast, modeling growth with multi-item assessments allows evaluation of how relative item performance may shift over time. While traditionally viewed as a nuisance under the label of "item parameter drift" (IPD), IPD may be of substantive interest if it reflects how learning manifests on different items or subscales at different rates. In this study, we apply the Explanatory Item Response Model to assess IPD in a causal inference context. Simulation results show that when IPD is not accounted for, both parameter estimates and standard errors can be affected. We illustrate with an empirical application to the persistence of transfer effects from a content literacy intervention , revealing how researchers can leverage IPD to achieve a more fine-grained understanding of how vocabulary learning develops over time.

  PublisherDOI

• Improving second-grade reading comprehension through a sustained content literacy intervention: A mixed-methods study examining the mediating role of domain-specific vocabulary.

  Journal of Educational Psychology · 2024-05-01 · 7 citations

  articleOpen accessSenior author

  This mixed-methods study explores the relationship between early elementary students’ domainspecific vocabulary knowledge and their ability to comprehend grade-level reading passages on unfamiliar science topics. Specifically, this study used (a) structural equation modeling (SEM) to examine the extent to which students’ networks of domain-specific vocabulary knowledge in Grades 1 and 2 mediated the effects of a Tier 1 content-based literacy intervention on domainspecific reading comprehension scores in Grade 2 (N = 2,156); and (b) quantitative survey and qualitative interview data from teachers (N = 48) to surface new themes about teacher vocabulary instruction that might suggest potential explanations for the SEM findings. SEM analysis revealed that students’ domain-specific vocabulary knowledge in first and second grade explained 69% of the treatment effect on a domain-specific reading comprehension outcome. Results from the quantitative survey also indicated that treatment group teachers reported providing more incidental exposures to vocabulary than control teachers (Effect Size [ES] = .54) and qualitative analyses revealed that teachers with high incidental exposures tended to provide expanded opportunities for their students to engage with words and to connect words to topics. Findings from this mixed-method study paint a more complete picture of (a) the important role domain-specific vocabulary knowledge plays in facilitating reading comprehension transfer in the domain of science, and (b) what teachers do during vocabulary instruction to promote transfer in domain-specific reading comprehension.

  PublisherOA PDFDOI

• Estimating Causal Effects on Psychological Networks Using Item Response Theory

  2024-11-13 · 1 citations

  preprintOpen accessSenior author

  Network models in which each variable interacts with the others in a complex system have emerged as an important alternative to latent variable models in psychometric research. However, confirmatory methods for group network comparison are limited by practical constraints, such as the computational intractability of the Ising model in large networks. In this study, we demonstrate how to estimate causal effects on network state and strength when direct network estimation is not feasible by leveraging the mathematical equivalencies between the Ising model and item response theory (IRT) models. We demonstrate through simulation that a two-parameter logistic (2PL) explanatory IRT model can simultaneously recover causal effects on network state and strength. We first apply the method to a single empirical example of a vocabulary assessment from a content literacy intervention to demonstrate model building and interpretation strategies. We then replicate our approach with 72 empirical datasets from randomized controlled trials with item-level outcome data in education, economics, health, and related fields. Our results show that causal effects on network strength are both common and uncorrelated with effects on network state, suggesting that causal network models can provide new insight into the impact of interventions in the social and behavioral sciences.

  PublisherOA PDFDOI

Frequent coauthors

• Jonathan Guryan

  28 shared

• David M. Quinn

  20 shared

• Mary Burkhauser

  18 shared

• Jackie Eunjung Relyea

  North Carolina State University

  16 shared

• Kyung Park

  16 shared

• Ethan Scherer

  Harvard University

  13 shared

• Joshua B. Gilbert

  11 shared

• Catherine A. Asher

  University of Michigan–Ann Arbor

  10 shared

Labs

• READS LabPI

Education

• EdD

  Harvard University