Exposure to multiple languages may support the development of Theory of Mind (ToM) in neurotypical (NT) and autistic children. However, previous research mainly applied group comparisons between monolingual and bilingual children, and the underlying mechanism of the observed difference remains unclear. The present study, therefore, sheds light on the effect of bilingualism on ToM in both NT and autistic children by measuring language experiences with a continuous operationalization. We measure ToM with a behavioral, linguistically simple tablet-based task, allowing inclusive assessment in autistic children. Analyses revealed no difference between monolingual and bilingual NT and autistic children. However, more balanced exposure to different languages within contexts positively predicted first-order false belief understanding in NT children but not autistic children. Mediation analysis showed that the impact in NT children was a direct effect and not mediated via other cognitive skills.

In the past decade, researchers have been increasingly interested in understanding the process of language learning, in addition to the effect of instructional interventions on L2 performance gains (i.e., learning products). One goal of such investigations is to reveal the interplay between learning conditions, processes, and outcomes where, for example, certain conditions can promote attention to the learning targets, which in turn facilitates learning. However, the statistical modeling approach taken often does not align with the conceptualization of the complex relationships between these variables. Thus, in this paper, we introduce mediation analysis to SLA research. We offer a step-by-step, contextualized tutorial on the practical application of mediation analysis in three different research scenarios, each addressing a different research design using either simulated or open-source datasets. Our overall goal is to promote the use of statistical techniques that are consistent with the theorization of language learning processes as mediators.

Research shows that meritocratic recruitment (MR) in public administration is positively related to improved government performance and developmental outcomes. However, the mechanisms behind these improvements remain understudied theoretically and empirically. This paper addresses this gap by theorising and testing two simultaneous pathways through which MR influences development outcomes. First, by prioritising competence over nepotism or political expedience, MR enhances the epistemic quality of bureaucratic personnel (the competence mechanism). Second, by creating incentive misalignment between bureaucrats and politicians, it enables bureaucrats to resist undue political influence, prioritise public interests in governance, and ultimately contribute to development (the impartiality mechanism). Applying mediation analysis to fourteen years of cross-national data, we examine whether changes in recruitment systems are associated with competence- and impartiality-laden indicators of government performance and developmental outcomes. The findings provide robust empirical support for these mechanisms, advancing theoretical understanding and empirical insights into the effects of MR.

This study was designed to explore the mediating role of serum 25-hydroxyvitamin D (25(OH) D) in Triglyceride–glucose (TyG) index and hypertension (HTN). Study participants were selected from the 2001 to 2018 National Health and Nutrition Examination Survey. Firstly, we estimated the association between TyG index and serum 25(OH)D with HTN using a weighted multivariable logistic regression model and restricted cubic spline. Secondly, we used a generalised additive model to investigate the correlation between TyG index and serum 25(OH)D. Lastly, serum 25(OH)D was investigated as a mediator in the association between TyG index and HTN. There were 14 099 subjects in total. TyG index was positively and linearly associated with HTN risk, while serum 25(OH)D had a U-shaped relationship with the prevalence of HTN. When the serum 25(OH)D levels were lower than 57·464 mmol/l, the prevalence of HTN decreased with the increase of serum 25(OH)D levels. When serum 25(OH)D levels rise above 57·464 mmol/l, the risk of HTN increases rapidly. Based on the U-shaped curve, serum 25(OH)D concentrations were divided into two groups: < 57·464 and ≥57·464 mmol/l. According to the mediation analysis, when serum 25(OH)D levels reached < 57·464 mmol/l, the positive association between the TyG index and incident HTN was increased by 25(OH)D. When serum 25(OH)D levels reached ≥ 57·464 mmol/l, the negative association between the TyG index and incident HTN was increased by 25(OH)D. There was a mediation effect between the TyG index and HTN, which was mediated by 25(OH)D. Therefore, we found that the association between serum 25(OH)D levels and TyG index may influence the prevalence of HTN.

Extended redundancy analysis (ERA) is a statistical approach to component-based multivariate regression modeling that explores interrelationships among multiple sets of while incorporating regression with a data-reduction technique. The extant models that utilize ERA have assumed the outcome variables with the same data type. Also, ERA models focused on estimating direct pathways only without explicitly addressing mediation effects. In this paper, ERA is extended to handle multiple mediators and mixed types of outcome variables by adopting a Bayesian framework, taking into account correlation structure among all of the outcome variables. The proposed method develops an algorithm that derives the joint posterior distribution of parameters using a Markov chain Monte Carlo algorithm. Simulations and an empirical dataset are provided to illustrate the usefulness of the proposed method.

Mediation analysis constitutes an important part of treatment study to identify the mechanisms by which an intervention achieves its effect. Structural equation model (SEM) is a popular framework for modeling such causal relationship. However, current methods impose various restrictions on the study designs and data distributions, limiting the utility of the information they provide in real study applications. In particular, in longitudinal studies missing data is commonly addressed under the assumption of missing at random (MAR), where current methods are unable to handle such missing data if parametric assumptions are violated.

In this paper, we propose a new, robust approach to address the limitations of current SEM within the context of longitudinal mediation analysis by utilizing a class of functional response models (FRM). Being distribution-free, the FRM-based approach does not impose any parametric assumption on data distributions. In addition, by extending the inverse probability weighted (IPW) estimates to the current context, the FRM-based SEM provides valid inference for longitudinal mediation analysis under the two most popular missing data mechanisms; missing completely at random (MCAR) and missing at random (MAR). We illustrate the approach with both real and simulated data.

A social network comprises both actors and the social connections among them. Such connections reflect the dependence among social actors, which is essential for individuals’ mental health and social development. In this article, we propose a mediation model with a social network as a mediator to investigate the potential mediation role of a social network. In the model, the dependence among actors is accounted for by a few mutually orthogonal latent dimensions which form a social space. The individuals’ positions in such a latent social space are directly involved in the mediation process between an independent and dependent variable. After showing that all the latent dimensions are equivalent in terms of their relationship to the social network and the meaning of each dimension is arbitrary, we propose to measure the whole mediation effect of a network. Although individuals’ positions in the latent space are not unique, we rigorously articulate that the proposed network mediation effect is still well defined. We use a Bayesian estimation method to estimate the model and evaluate its performance through an extensive simulation study under representative conditions. The usefulness of the network mediation model is demonstrated through an application to a college friendship network.

This research concerns a mediation model, where the mediator model is linear and the outcome model is also linear but with a treatment–mediator interaction term and a residual correlated with the residual of the mediator model. Assuming the treatment is randomly assigned, parameters in this mediation model are shown to be partially identifiable. Under the normality assumption on the residual of the mediator and the residual of the outcome, explicit full-information maximum likelihood estimates of model parameters are introduced given the correlation between the residual for the mediator and the residual for the outcome. A consistent variance matrix of these estimates is derived. Currently, the coefficients of this mediation model are estimated using the iterative feasible generalized least squares (IFGLS) method that is originally developed for seemingly unrelated regressions (SURs). We argue that this mediation model is not a system of SURs. While the IFGLS estimates are consistent, their variance matrix is not. Theoretical comparisons of the FIMLE variance matrix and the IFGLS variance matrix are conducted. Our results are demonstrated by simulation studies and an empirical study. The FIMLE method has been implemented in a freely available R package iMediate.

Despite wide applications of both mediation models and missing data techniques, formal discussion of mediation analysis with missing data is still rare. We introduce and compare four approaches to dealing with missing data in mediation analysis including listwise deletion, pairwise deletion, multiple imputation (MI), and a two-stage maximum likelihood (TS-ML) method. An R package bmem is developed to implement the four methods for mediation analysis with missing data in the structural equation modeling framework, and two real examples are used to illustrate the application of the four methods. The four methods are evaluated and compared under MCAR, MAR, and MNAR missing data mechanisms through simulation studies. Both MI and TS-ML perform well for MCAR and MAR data regardless of the inclusion of auxiliary variables and for AV-MNAR data with auxiliary variables. Although listwise deletion and pairwise deletion have low power and large parameter estimation bias in many studied conditions, they may provide useful information for exploring missing mechanisms.

To investigate the associations between dietary patterns and biological ageing, identify the most recommended dietary pattern for ageing and explore the potential mediating role of gut microbiota in less-developed ethnic minority regions (LEMRs). This prospective cohort study included 8288 participants aged 30–79 years from the China Multi-Ethnic Cohort study. Anthropometric measurements and clinical biomarkers were utilised to construct biological age based on Klemera and Doubal’s method (KDM-BA) and KDM-BA acceleration (KDM-AA). Dietary information was obtained through the baseline FFQ. Six dietary patterns were constructed: plant-based diet index, healthful plant-based diet index, unhealthful plant-based diet index, healthy diet score, Dietary Approaches to Stop Hypertension (DASH), and alternative Mediterranean diets. Follow-up adjusted for baseline analysis assessed the associations between dietary patterns and KDM-AA. Additionally, quantile G-computation identified significant beneficial and harmful food groups. In the subsample of 764 participants, we used causal mediation model to explore the mediating role of gut microbiota in these associations. The results showed that all dietary patterns were associated with KDM-AA, with DASH exhibiting the strongest negative association (β = −0·91, 95 % CI (–1·19, −0·63)). The component analyses revealed that beneficial food groups primarily included tea and soy products, whereas harmful groups mainly comprised salt and processed vegetables. In mediation analysis, the Synergistetes and Pyramidobacter possibly mediated the negative associations between plant-based diets and KDM-AA (5·61–9·19 %). Overall, healthy dietary patterns, especially DASH, are negatively associated with biological ageing in LEMRs, indicating that Synergistetes and Pyramidobacter may be potential mediators. Developing appropriate strategies may promote healthy ageing in LEMRs.

The Thinking Healthy Program (THP) is a multicomponent low-intensity cognitive behavioral therapy-based psychosocial intervention. This intervention has been shown to be clinically effective in perinatal depression (PND) and feasible for implementation in low-resourced settings. It has also been shown to work universally for different phenotypes of PND. However, the mechanism through which THP resolves different phenotypes of PND are unclear. The present investigation presents secondary mediation analyses of a dataset curated from a cluster randomized controlled trial conducted in Pakistan assessing the effectiveness of the THP. Women aged 16–45 years in their third pregnancy trimester, with a diagnosis of PND, underwent 16 sessions of the intervention. The severity of depression was assessed using the Hamilton Depression Rating Scale (HDRS). 2-1-1 mediation models revealed that social support exerted significant mediation in driving the intervention effects for improving the severity of depressive symptoms on the HDRS [B (SE) = 0.45 (0.09), 95% CI: 0.30–0.60] and its symptom dimensions of core emotional symptoms [B (SE) = 0.27 (0.06), 95% CI: 0.18–0.37], somatic symptoms [B (SE) = 0.24 (0.04), 95% CI: 0.16–0.31] and insomnia symptoms [B (SE) = 0.04 (0.02), 95% CI: 0.02–0.07].