There are various statistical models with different features and assumptions in meta-analysis. Popular models include the fixed-effect (or common effect), random-effects, and mixed-effects models. Apart from these models, several alternative models, such as the multiplicative error model (also known as the unrestricted weighted least squares model), the hybrid of additive and multiplicative error models, and the location-scale model, have been proposed in the literature. Understanding these models can be challenging for researchers without a solid mathematical background. Implementing or modifying these models is even more challenging for researchers unless they have advanced statistical and programming knowledge. This tutorial elucidates a structural equation modeling (SEM) framework to understand these models. Several R packages are introduced to facilitate the specification and generation of graphical models for these meta-analytic models. Researchers may fit these models with the full information maximum likelihood estimation method. This holds significant potential across two domains. First, it can be used as an educational tool in teaching and learning meta-analytic models. Second, the framework supports the development and evaluation of novel meta-analytical models, which have not yet been implemented in current meta-analysis software, thereby fostering advancements in meta-analytic methods. This tutorial also demonstrates, using two real datasets, how to extend it to address research questions involving complex meta-analytic data. Limitations and future directions to extend the SEM-based meta-analysis are discussed.

Due to the indeterminacy problem in factor analysis, factors are not uniquely determined by observed data. We investigate the consequences of this problem for the interpretation of simulation studies and for linear structural equation models. For simulation procedures, factor indeterminacy implies that as long as the simulated data have the correct covariance matrix, there exists a possibly large space of factors and error terms that can generate the simulated data via the model equations. For linear structural equation models, the linearity of the structural part of the model is only partly testable by the distribution of the data. Practical recommendations for conducting simulation studies are provided. We give numerical illustrations, including a case where the observed data are normal and fit a linear structural equation model perfectly, yet are generated by a highly non-linear structural equation model.

The autoregressive latent trajectory (ALT) model was introduced at the 1999 Psychometric Society meeting. Over the past 25 years, ALT models have become widely used for studying longitudinal dynamics, especially changes in multiple variables over time. Yet, their popularity has also generated confusion. Some published claims are contradictory, others are factually incorrect, and many emphasize the need for clearer explanations of the classic and latent-variable ALT (LV-ALT) models. This article explains the assumptions and structure of ALT models and clarifies how their parameters should be interpreted. It also supports both between- and within-component interpretations—contrary to assertions that such interpretations are impossible. We review prior literature on ALT properties and assess their accuracy. Further, we show the broad range of longitudinal models that are special cases of ALT models, enabling researchers to test and select the most suitable models for their data. An empirical example, used in multiple studies, illustrates these points. Overall, ALT and LV-ALT models provide a unifying framework for comparing many common longitudinal models. By clarifying their properties, correcting misconceptions, and demonstrating their applications, we aim to help researchers choose models that best reflect their theories and data.

This quantitative study develops and validates a novel Culturally Responsive Blended Learning Success (CR-BLS) model for evaluating technology-enhanced EFL instruction effectiveness in collective educational contexts. Conducted at a Chinese university, the research examined an intelligent blended learning ecosystem integrating three widely adopted platforms (U-Campus, GradingNet, and Learning Hub) for content delivery, assessment, and collaboration in College English instruction. The CR-BLS model extends a widely recognized seven-factor quality framework by integrating three culturally grounded moderators: hierarchical teacher–student relationships, collective learning orientation, and examination-driven motivation. The proposed model was empirically validated through data collection from 1,816 non-English major undergraduates engaged with the blended learning ecosystem using partial least squares structural equation modeling (PLS-SEM). Results revealed that the proposed quality factors explained 71.4% of the variance in perceived satisfaction, 54.2% in perceived usefulness, and 34.1% in system use. Learner quality and instructor quality significantly influenced both perceived satisfaction and usefulness, while educational system quality primarily affected system use. Furthermore, perceived usefulness, satisfaction, and system use collectively accounted for 64.7% of the variance in perceived learning benefits. Cultural moderators demonstrated significant influence on the relationships between quality dimensions and learning outcomes, enhancing the proposed model’s contextual validity for collective educational environments. This study contributes to both theoretical advancement and practical implementation of blended learning systems in collective educational settings.

Parthenium weed (Parthenium hysterophorus L.) is rapidly invading southern Oman, posing growing challenges to agricultural productivity and rural livelihoods in Dhofar’s crop–livestock systems. This study assesses its agronomic and socioeconomic impacts using field surveys and a stratified household survey of 40 farms conducted between June and August 2022. Data were analyzed using partial least-squares structural equation modeling (PLS-SEM) to examine the relationships between farmer characteristics, infestation levels, weed management practices, and farm revenue. Results indicate that while most farmers are aware of P. hysterophorus, limited recognition at early growth stages delays control and increases management costs. Education (β = 1.009, P = 0.06), cultivated area (β = 1.033, P = 0.003), and chemical control expenditures (β = 0.295, P = 0.05) were positively associated with gross revenue, whereas age had a negative effect (β = −0.762, P = 0.08). Infestation significantly increased labor-intensive weeding costs (β = 2.072, P = 0.07) but reduced chemical use (β = −1.303, P = 0.032), indicating substitution toward manual control. Although time spent uprooting parthenium reduced infestation levels (β = −0.128, P < 0.001), it also increased weeding and chemical control costs, highlighting the financial burden of relying on manual methods. Crop-specific analysis showed heterogeneous vulnerability, with peas (Pisum sativum L.) more affected than wheat (Triticum aestivum L.) and tomato (Solanum lycopersicum L.). Overall, the findings demonstrate that parthenium imposes measurable income and welfare risks on farming households. These impacts can be mitigated through farmer education, early detection, and judicious chemical use within an integrated weed management framework. Strengthening extension services and promoting crop-specific interventions are essential for protecting rural livelihoods and food security in arid, invasion-prone farming systems.

Structural equation modeling (SEM) is a flexible statistical technique with multiple applications, including behavioral genetics and social sciences. Building on the original design of the umx package, which improved accessibility to OpenMx by specifying a concise syntax, umx v4.5 extends functionality for longitudinal and causal twin designs while improving interoperability with graphical modeling tools such as Onyx. New capabilities include: classic and modern cross-lagged panel models; Mendelian Randomization Direction-of-Causation (MR-DoC) twin models incorporating polygenic scores as instruments; support for definition variables directly in umxRAM(); a workflow for importing paths from Ωnyx; a dedicated function for incorporating censored variables’ data into models, particularly valuable in biomarker research; improved covariate placeholder handling for definition variables; sex-limitation modeling across five twin groups, accommodating quantitative and qualitative sex differences; and covariate residualization in wide- or long-format data. These new functionalities accelerate reproducible, reliable, publication-ready twin and family modeling, and integrated journal-quality reporting, thereby lowering barriers to genetic epidemiological analyses

Structural equation modeling (SEM) is a powerful and flexible modeling framework for testing complex relationships among observed and latent variables. However, its methodological complexity and analytical flexibility also increase the risk of questionable research practices (QRPs), especially in fields like applied linguistics, where training in advanced statistics may be limited. This article synthesizes the literature on QRPs and applies it to SEM by identifying seven categories of problematic practices: not checking assumptions, not validating a measurement model, not testing competing models, not sufficiently justifying modeling decisions, relying on post hoc model modification, overemphasizing global fit indices, and incomplete or nontransparent reporting. Each practice is described with examples and linked to broader issues in research ethics and transparency. The paper concludes with concrete recommendations for improving the credibility and reproducibility of SEM research, emphasizing the integration of best practices with the principles of open science.

Due to certain developments in the business of fundraising, there exists an increasing necessity to theoretically and empirically analyze donor behavior. This article examines donor retention from the donor’s point of view. It concretizes four antecedents and their relationships with one another. The influencing variables investigated are commitment, trust, satisfaction, and involvement. The empirical analysis conducted in Germany among donors of four representative social non-profit organizations shows that all variables have an influence on donor retention, although some of them only indirectly.

Eating disorders (EDs) are particularly prevalent among young adult females. Previous research has shown that childhood trauma and reduced mentalizing abilities are involved in ED symptoms. The current study was aimed at testing the mediating effects of failures in mentalizing on the relationship between childhood trauma and ED risk among young adult females. The sample consisted of 409 Caucasian young adult females, aged between 18 and 30 years old (M = 23.45, SD = 2.76). The reported mean body mass index was within the normal range (M = 22.62; SD = 4.35). Self-report instruments were administered to assess the variables of interest. Structural equation modeling revealed that childhood trauma predicted increased failures in mentalizing (β = .36) and ED risk (β = .30), that failures in mentalizing predicted an increased ED risk (β = .35), and that the positive association between childhood trauma and ED risk was partially mediated by failures in mentalizing (indirect effect: β = .13). These findings suggest that ED symptoms might result from unprocessed and painful feelings embedded in child abuse and neglect. Clinical interventions focused at improving mentalizing abilities might reduce the ED risk among young adult females who have been exposed to childhood trauma.

The current meta-analytic structural equation modeling (MASEM) techniques cannot properly deal with cases where there are multiple effect sizes available for the same relationship from the same study. Existing applications either treat these effect sizes as independent, randomly select one effect size amongst many, or create an average effect size. None of these approaches deal with the inherent dependency in effect sizes, and either leads to biased estimates or loss of information and power. An alternative technique is to use univariate three-level modeling in the two-stage approach to model these dependencies. These different strategies for dealing with dependent effect sizes in the context of MASEM have not been previously compared in a simulation study. This study aims to compare the performance of these strategies across different conditions; varying the number of studies, the number of dependent effect sizes within studies, the correlation between the dependent effect sizes, the magnitude of the path coefficient, and the between-studies variance. We examine the relative bias in parameter estimates and standard errors, coverage proportions of confidence intervals, as well as mean standard error and power as measures of efficiency. The results suggest that there is not one method that performs well across all these criteria, pointing to the need for better methods.