Background/Objective: Growing recognition that collaboration among scientists from diverse disciplines fosters the emergence of solutions to complex scientific problems has spurred initiatives to train researchers to collaborate in interdisciplinary teams. Evaluations of collaboration patterns in these initiatives have tended to be cross-sectional, rather than clarifying temporal changes in collaborative dynamics. Mobile health (mHealth), the science of using mobile, wireless devices to improve health outcomes, is a field whose advancement needs interdisciplinary collaboration. The NIH-supported annual mHealth Training Institute (mHTI) was developed to meet that need and provides a unique testbed. Methods: In this study, we applied a longitudinal social network analysis technique to evaluate how well the program fostered communication among the disciplinarily diverse scholars participating in the 2017−2019 mHTIs. By applying separable temporal exponential random graph models, we investigated the formation and persistence of project-based and fun conversations during the mHTIs. Results: We found that conversations between scholars of different disciplines were just as likely as conversations within disciplines to form or persist in the 2018 and 2019 mHTI, suggesting that the mHTI achieved its goal of fostering interdisciplinary conversations and could be a model for other team science initiatives; this finding is also true for scholars from different career stages. The presence of team and gender homophily effects in certain years suggested that scholars tended to communicate within the same team or gender. Conclusion: Our results demonstrate the usefulness of longitudinal network models in evaluating team science initiatives while clarifying the processes driving interdisciplinary communications during the mHTIs.
Structured mixture item response models (StrMixIRMs) are a special type of constrained confirmatory mixture item response theory (IRT) model for detecting latent performance differences in a measurement instrument by characteristic item groups, and classifying respondents according to these differences. In light of limited software options for estimating StrMixIRMs under existing frameworks, this paper proposes reparameterizing it as a confirmatory mixture IRT model using interaction effects between latent classes and item groups. The reparameterization allows for easier implementation of StrMixIRMs with multiple software programs that have mixture modeling capabilities, including open-source ones. This widens the accessibility to these models to a broad range of users and thus can facilitate research and applications of StrMixIRMs. This paper serves two main goals: First, we introduce StrMixIRMs, focusing on the proposed reparameterization based on interaction effects and its various extensions. Second, we illustrate use cases of this novel reparameterization within the mirt 1.41 package in R by employing two empirical datasets. Detailed R code with notes are provided for the applications along with an interpretation of the outputs.
In intelligence research, the sharing of item response data from cognitive ability assessments is often restricted by privacy concerns, while traditional parametric simulation methods frequently fail to capture complex response dependencies. This study proposes a neural network copula (NNC) framework for generating synthetic dichotomous item response data that preserves essential psychometric properties without revealing sensitive examinee information. By decoupling the modeling of marginal item probabilities from the dependence structure using a deep autoencoder and kernel density estimation, the framework accommodates the discrete nature of binary item response data while minimizing distributional assumptions. Validation against large-scale empirical data demonstrated high correspondence across multiple facets. At the data consistency level, the NNC-based synthetic data reproduced total score distributions and inter-item correlations. Psychometrically, the method yielded consistent item characteristic curve parameter estimates, item fit statistics, and test information functions. Furthermore, Monte Carlo replications demonstrated algorithmic stability and inferential precision.
This paper presents a class of higher-order diagnostic classification models (HO–DCMs) capable of capturing complex, nonlinear hierarchical relationships among attributes. Building on and extending prior work, we adopt a nominal response model framework in item response theory and leverage standard maximum likelihood estimation (MLE). In parallel, we demonstrate that sequential HO–DCMs can likewise be implemented within an MLE framework. Furthermore, we introduce a novel chunk-based approach for representing attribute hierarchies, wherein attributes are organized into cognitively coherent subgraphs (chunks) nested within a continuous general ability continuum. The performance of the models is validated through simulation studies evaluating parameter recovery, classification accuracy, and null rejection rates of goodness-of-fit measures. An empirical demonstration showcases how the proposed framework can be applied in practice, highlighting its advantages in model flexibility, interpretability, and the additional diagnostic insights it affords.
Understanding how teaching quality influences student achievement offers insights for the design and implementation of professional development programs. Drawing from an adapted version of the Dynamic Model of Educational Effectiveness for the Taiwanese educational context, this study aims to investigate the relationship between teaching quality and student outcomes in Taiwan. Data for this study were collected through a questionnaire and mathematics tests, involving 55 Taiwanese primary school teachers and 1480 students. Teachers were classified into three stages of teaching quality, which were then linked to students’ mathematics performance. Results showed that students taught by teachers in the highest-quality teaching stage generally achieved higher final mathematics scores, regardless of their initial performance. The study also reveals that students’ initial mathematics performance and socioeconomic status (SES) are significant predictors of their final mathematics outcomes. The outcomes highlight the necessity of considering cultural and contextual factors when identifying stages of teaching quality. Additionally, the findings reaffirm that teaching quality is an important factor in student achievement, as students in high-quality learning environments tend to achieve better academic outcomes. Suggestions related to teacher professional development are proposed, encompassing the development of a student feedback instrument to support teachers in optimising the quality of their teaching and enhancing the overall learning environment.
In item response theory (IRT), the two-tier IRT model can be considered a general confirmatory IRT model, in which multiple correlated primary factors and orthogonal specific factors are loaded onto items. Although full information maximum likelihood estimation combined with the dimension reduction technique has been successful, it becomes computationally infeasible when the number of primary factors increases. In this study, we propose a composite marginal likelihood (CML) estimation via the expectation-maximization algorithm to alleviate this computational demand. We adopt the pairwise fitting approach, where item responses are divided into subsets according to primary factor pairs. Combined with the dimension reduction technique, this approach can, in certain cases, reduce the required numerical integration to as few as three latent variables, regardless of the total number of primary and specific factors in the model. The potential of the CML method for the two-tier IRT model is demonstrated by a simulation study and applications to two empirical datasets.