Walter Lana Leite
Current Organizations / Institutes
The organizations/institutes from the scholar's latest article
Previous Organizations / Institutes
The organizations / institutes from the scholar's other articles in the past ten years
Walter Lana Leite has a total of 155 co-authors. Here are the top co-authors by number of shared publications:
- Yongseok Lee
Department of Human Development and Family Science, Purdue University, West Lafayette, USA
8 shared publications A. Corinne Huggins-Manley
University of Florida, Gainesville, USA
6 shared publicationsJames Algina
University of Florida, Gainesville, FL, USA
6 shared publications- Huan Kuang
Florida State University
5 shared publications - Huibin Zhang
Tennessee Reading Research Center University of Tennessee Knoxville Tennessee USA
5 shared publications - Wanli Xing
University of Miami
5 shared publications
Latest Publications
- 6 November 2026
- Published by Taylor & Francis in Journal of Research on Educational Effectiveness
- p. 1-36
Do Intelligent Tutoring Systems Benefit K-12 Students? A Meta-Analysis and Evaluation of Heterogeneity of Treatment Effects in the U.S.
Strong interest in applying artificial intelligence (AI) to education resulted in multiple annual conferences and a large stream of publications (Korkmaz & Correia, 2019). Among these applications ...
- 6 November 2026
- Published by Taylor & Francis in Journal of Research on Educational Effectiveness
- p. 1-36
Do Intelligent Tutoring Systems Benefit K-12 Students? A Meta-Analysis and Evaluation of Heterogeneity of Treatment Effects in the U.S.
Strong interest in applying artificial intelligence (AI) to education resulted in multiple annual conferences and a large stream of publications (Korkmaz & Correia, 2019). Among these applications ...
- Open Access
- 8 June 2026
- Published by Springer Nature in Behavior Research Methods
- Vol. 58 (7) , 188
Publisher Correction: Enhancing propensity score analysis with data missing not at random: Introducing dual-forest proximity imputation
- 8 June 2026
- Published by Springer Nature in Behavior Research Methods
- Vol. 58 (7) , 188
Open AccessPublisher Correction: Enhancing propensity score analysis with data missing not at random: Introducing dual-forest proximity imputation
- Open Access
- 29 May 2026
- Published by Springer Nature in Behavior Research Methods
- Vol. 58 (7) , 182
Enhancing propensity score analysis with data missing not at random: Introducing dual-forest proximity imputation
Researchers using propensity score analysis (PSA) to estimate treatment effects using secondary data may have to handle data that are missing not at random (MNAR). Existing methods for PSA with MNAR data use logistic regression to model the missing data mechanisms, thus requiring manual specification of functional forms, and are difficult to implement with a large number of covariates. To overcome these limitations, this study proposes alternatives to existing methods by replacing logistic regression with a random forest. Also, it introduces the dual-forest proximity imputation method, which leverages two types of proximity matrices of random forest techniques and incorporates missingness pattern information in each matrix. Results from a Monte Carlo simulation show dual-forest proximity imputation’s enhanced bias reduction with various types of MNAR mechanisms as compared to existing and alternative methods. A case study is also provided using data from the National Longitudinal Survey of Youth (NLSY79).
- 29 May 2026
- Published by Springer Nature in Behavior Research Methods
- Vol. 58 (7) , 182
Open AccessEnhancing propensity score analysis with data missing not at random: Introducing dual-forest proximity imputation
Researchers using propensity score analysis (PSA) to estimate treatment effects using secondary data may have to handle data that are missing not at random (MNAR). Existing methods for PSA with MNAR data use logistic regression to model the missing data mechanisms, thus requiring manual specification of functional forms, and are difficult to implement with a large number of covariates. To overcome these limitations, this study proposes alternatives to existing methods by replacing logistic regression with a random forest. Also, it introduces the dual-forest proximity imputation method, which leverages two types of proximity matrices of random forest techniques and incorporates missingness pattern information in each matrix. Results from a Monte Carlo simulation show dual-forest proximity imputation’s enhanced bias reduction with various types of MNAR mechanisms as compared to existing and alternative methods. A case study is also provided using data from the National Longitudinal Survey of Youth (NLSY79).
- 15 May 2026
- Published by Taylor & Francis in Structural Equation Modeling: A Multidisciplinary Journal
- Vol. 33 (6) , 967-974
Sensitivity Analyses for Omitted Confounders in Structural Equation Models with Tabu Search Optimization
- 15 May 2026
- Published by Taylor & Francis in Structural Equation Modeling: A Multidisciplinary Journal
- Vol. 33 (6) , 967-974
Sensitivity Analyses for Omitted Confounders in Structural Equation Models with Tabu Search Optimization
- 2 January 2026
- Published by Taylor & Francis in Professional Development in Education
- Vol. 52 (1) , 378-420
Understanding the heterogeneity of effects of teacher professional development: a systematic meta-review
Professional development (PD) programs for in-service teachers play a critical role in providing opportunities to deepen content knowledge, refine pedagogical skills, and adapt to evolving standard...
- 2 January 2026
- Published by Taylor & Francis in Professional Development in Education
- Vol. 52 (1) , 378-420
Understanding the heterogeneity of effects of teacher professional development: a systematic meta-review
Professional development (PD) programs for in-service teachers play a critical role in providing opportunities to deepen content knowledge, refine pedagogical skills, and adapt to evolving standard...
- 1 January 2026
- Published by Psychometric Society
Cost-Efficient Sampling Strategies for Cluster-Randomized Trials Detecting Moderation and Main Effects
Moderation analyses can answer questions about where, for whom, and under what conditions intervention effects are most salient. Together with the main effect analyses, which assess whether interventions work, the results from the moderation and main effect analyses provide the evidence base for effective policy development. The literature has developed statistical power formulas for the main and moderation effects. However, literature has not offered a holistic framework that simultaneously considers both cost efficiency and statistical power for experimental studies investigating these effects. The present study develops a jointly optimal design framework for cluster-randomized trials (CRTs) investigating both moderation and main effects. Such a framework enables researchers to identify optimal sampling allocations by accounting for sampling costs and statistical power for both effects. The results show that the optimal sample allocations in CRTs are the same for the average treatment effect and the moderated treatment effect with cluster-level moderators. The proposed methods, including optimal sample allocation calculations and power analyses for both effects, have been implemented in the R package odr.
- 1 January 2026
- Published by Psychometric Society
Cost-Efficient Sampling Strategies for Cluster-Randomized Trials Detecting Moderation and Main Effects
Moderation analyses can answer questions about where, for whom, and under what conditions intervention effects are most salient. Together with the main effect analyses, which assess whether interventions work, the results from the moderation and main effect analyses provide the evidence base for effective policy development. The literature has developed statistical power formulas for the main and moderation effects. However, literature has not offered a holistic framework that simultaneously considers both cost efficiency and statistical power for experimental studies investigating these effects. The present study develops a jointly optimal design framework for cluster-randomized trials (CRTs) investigating both moderation and main effects. Such a framework enables researchers to identify optimal sampling allocations by accounting for sampling costs and statistical power for both effects. The results show that the optimal sample allocations in CRTs are the same for the average treatment effect and the moderated treatment effect with cluster-level moderators. The proposed methods, including optimal sample allocation calculations and power analyses for both effects, have been implemented in the R package odr.
- Open Access
Do intelligent tutoring systems benefit K-12 students? A meta-analysis and evaluation of heterogeneity of treatment effects in the U.S
To expand the use of intelligent tutoring systems (ITS) in K-12 schools, it is essential to understand the conditions under which their use is most beneficial. This meta-analysis evaluated the heterogeneity of ITS effects across studies focusing on elementary, middle, and high schools in the U.S. It included 18 studies with 77 effect sizes across 11 ITS. Overall, there was a significant positive effect size of ITS on U.S. K-12 students' learning outcomes (g=0.271, SE=0.011, p=0.001). Furthermore, effect sizes were similar across elementary and middle schools, and for low-achieving students, but were lower in studies including rural schools. A MetaForest analysis showed that providing worked-out examples, intervention duration, intervention condition, type of learning outcome, and immediate measurement were the most important moderators of treatment effects.
Open AccessDo intelligent tutoring systems benefit K-12 students? A meta-analysis and evaluation of heterogeneity of treatment effects in the U.S
To expand the use of intelligent tutoring systems (ITS) in K-12 schools, it is essential to understand the conditions under which their use is most beneficial. This meta-analysis evaluated the heterogeneity of ITS effects across studies focusing on elementary, middle, and high schools in the U.S. It included 18 studies with 77 effect sizes across 11 ITS. Overall, there was a significant positive effect size of ITS on U.S. K-12 students' learning outcomes (g=0.271, SE=0.011, p=0.001). Furthermore, effect sizes were similar across elementary and middle schools, and for low-achieving students, but were lower in studies including rural schools. A MetaForest analysis showed that providing worked-out examples, intervention duration, intervention condition, type of learning outcome, and immediate measurement were the most important moderators of treatment effects.
- Open Access
- 16 October 2025
- Published by American Psychological Association (APA) in Psychological Methods
Machine learning for propensity score estimation: A systematic review and reporting guidelines.
Machine learning (ML) has become a common approach for estimating propensity scores (PSs) for quasi-experimental research using matching, weighting, or stratification on the PS. This systematic review examined 179 applications of ML for PS estimation across different fields, such as health, education, social sciences, and business over 40 years. The results show that the gradient boosting machine (GBM) is the most frequently used method, followed by random forest. Classification and regression trees, neural networks, and the super learner were also used in more than 5% of studies. The most frequently used packages to estimate PSs were twang, gbm, and randomforest in the R statistical software. The review identified that critical steps of the propensity score analysis are frequently underreported. Covariate balance evaluation was not reported by 48.04% of articles. Also, improper use of p values for covariate balance evaluation was identified in 13.97% of the studies. Only 22.8% of studies performed a sensitivity analysis. Many hyperparameter configurations were used for ML methods, but only 46.9% of studies reported the hyperparameters used. A set of guidelines for reporting the use of ML for PS estimation is provided. (PsycInfo Database Record (c) 2025 APA, all rights reserved).
- 16 October 2025
- Published by American Psychological Association (APA) in Psychological Methods
Open AccessMachine learning for propensity score estimation: A systematic review and reporting guidelines.
Machine learning (ML) has become a common approach for estimating propensity scores (PSs) for quasi-experimental research using matching, weighting, or stratification on the PS. This systematic review examined 179 applications of ML for PS estimation across different fields, such as health, education, social sciences, and business over 40 years. The results show that the gradient boosting machine (GBM) is the most frequently used method, followed by random forest. Classification and regression trees, neural networks, and the super learner were also used in more than 5% of studies. The most frequently used packages to estimate PSs were twang, gbm, and randomforest in the R statistical software. The review identified that critical steps of the propensity score analysis are frequently underreported. Covariate balance evaluation was not reported by 48.04% of articles. Also, improper use of p values for covariate balance evaluation was identified in 13.97% of the studies. Only 22.8% of studies performed a sensitivity analysis. Many hyperparameter configurations were used for ML methods, but only 46.9% of studies reported the hyperparameters used. A set of guidelines for reporting the use of ML for PS estimation is provided. (PsycInfo Database Record (c) 2025 APA, all rights reserved).
- Open Access
- 18 July 2025
- Published by Center for Open Science in OSF Preprints
Enhancing Propensity Score Analysis with data Missing Not at Random: Introducing Dual-Forest Proximity Imputation
Researchers using propensity score analysis (PSA) to estimate treatment effects using secondary data may have to handle data that is missing not at random (MNAR). Existing methods for PSA with MNAR data use logistic regression to model the missing data mechanisms, thus requiring manual specification of functional forms, and are difficult to implement with a large number of covariates. To overcome these limitations, this study proposes alternatives to existing methods by replacing logistic regression with a random forest. Also, it introduces the Dual-Forest Proximity imputation method, which leverages two types of proximity matrices of random forest techniques and incorporates missing pattern information in each matrix. Results from a Monte Carlo simulation show Dual-Forest Proximity imputation’s enhanced bias reduction with various types of MNAR mechanisms as compared to existing and alternative methods. A case study is also provided using data from the National Longitudinal Survey of Youth 1979 (NLSY79).
- 18 July 2025
- Published by Center for Open Science in OSF Preprints
Open AccessEnhancing Propensity Score Analysis with data Missing Not at Random: Introducing Dual-Forest Proximity Imputation
Researchers using propensity score analysis (PSA) to estimate treatment effects using secondary data may have to handle data that is missing not at random (MNAR). Existing methods for PSA with MNAR data use logistic regression to model the missing data mechanisms, thus requiring manual specification of functional forms, and are difficult to implement with a large number of covariates. To overcome these limitations, this study proposes alternatives to existing methods by replacing logistic regression with a random forest. Also, it introduces the Dual-Forest Proximity imputation method, which leverages two types of proximity matrices of random forest techniques and incorporates missing pattern information in each matrix. Results from a Monte Carlo simulation show Dual-Forest Proximity imputation’s enhanced bias reduction with various types of MNAR mechanisms as compared to existing and alternative methods. A case study is also provided using data from the National Longitudinal Survey of Youth 1979 (NLSY79).
- 31 May 2025
- Published by SAGE Publications in Educational and Psychological Measurement
- Vol. 85 (6) , 1180-1199
Evaluating the Performance of a Regularized Differential Item Functioning Method for Testlet-Based Polytomous Items
This study investigated the effect of testlets on regularization-based differential item functioning (DIF) detection in polytomous items, focusing on the generalized partial credit model with lasso penalization (GPCMlasso) DIF method. Five factors were manipulated: sample size, magnitude of testlet effect, magnitude of DIF, number of DIF items, and type of DIF-inducing covariates. Model performance was evaluated using false-positive rate (FPR) and true-positive rate (TPR). Results showed that the simulation had effective control of FPR across conditions, while the TPR was differentially influenced by the manipulated factors. Generally, the small testlet effect did not noticeably affect the GPCMlasso model’s performance regarding FPR and TPR. The findings provide evidence of the effectiveness of the GPCMlasso method for DIF detection in polytomous items when testlets were present. The implications for future research and limitations were also discussed.
- 31 May 2025
- Published by SAGE Publications in Educational and Psychological Measurement
- Vol. 85 (6) , 1180-1199
Evaluating the Performance of a Regularized Differential Item Functioning Method for Testlet-Based Polytomous Items
This study investigated the effect of testlets on regularization-based differential item functioning (DIF) detection in polytomous items, focusing on the generalized partial credit model with lasso penalization (GPCMlasso) DIF method. Five factors were manipulated: sample size, magnitude of testlet effect, magnitude of DIF, number of DIF items, and type of DIF-inducing covariates. Model performance was evaluated using false-positive rate (FPR) and true-positive rate (TPR). Results showed that the simulation had effective control of FPR across conditions, while the TPR was differentially influenced by the manipulated factors. Generally, the small testlet effect did not noticeably affect the GPCMlasso model’s performance regarding FPR and TPR. The findings provide evidence of the effectiveness of the GPCMlasso method for DIF detection in polytomous items when testlets were present. The implications for future research and limitations were also discussed.