The latent space item response model (LSIRM) is a newly-developed approach to analyzing and visualizing conditional dependencies in item response data, manifested as the interactions between respondents and items, between respondents, and between items. This paper provides a practical guide to the Bayesian estimation of LSIRM using three open-source software options, JAGS, Stan, and NIMBLE in R. By means of an empirical example, we illustrate LSIRM estimation, providing details on the model specification and implementation, convergence diagnostics, model fit evaluations and interaction map visualizations.
A growing evidence base suggests that complex healthcare problems are optimally tackled through cross-disciplinary collaboration that draws upon the expertise of diverse researchers. Yet, the influences and processes underlying effective teamwork among independent researchers are not well-understood, making it difficult to fully optimize the collaborative process. To address this gap in knowledge, we used the annual NIH mHealth Training Institutes as a testbed to develop stochastic actor-oriented models that explore the communicative interactions and psychological changes of its disciplinarily and geographically diverse participants. The models help investigate social influence and social selection effects to understand whether and how social network interactions influence perceptions of team psychological safety during the institute and how they may sway communications between participants. We found a degree of social selection effects: in particular years, scholars were likely to choose to communicate with those who had more dissimilar levels of psychological safety. We found evidence of social influence, in particular, from scholars with lower psychological safety levels and from scholars with reciprocated communications, although the sizes and directions of the social influences somewhat varied across years. The current study demonstrated the utility of stochastic actor-oriented models in understanding the team science process which can inform team science initiatives. The study results can contribute to theory-building about team science which acknowledges the importance of social influence and selection.
Random item effects item response theory (IRT) models, which treat both person and item effects as random, have received much attention for more than a decade. The random item effects approach has several advantages in many practical settings. The present study introduced an explanatory multidimensional random item effects rating scale model. The proposed model was formulated under a novel parameterization of the nominal response model (NRM), and allows for flexible inclusion of person-related and item-related covariates (e.g., person characteristics and item features) to study their impacts on the person and item latent variables. A new variant of the Metropolis-Hastings Robbins-Monro (MH-RM) algorithm designed for latent variable models with crossed random effects was applied to obtain parameter estimates for the proposed model. A preliminary simulation study was conducted to evaluate the performance of the MH-RM algorithm for estimating the proposed model. Results indicated that the model parameters were well recovered. An empirical data set was analyzed to further illustrate the usage of the proposed model.
scite is a Brooklyn-based organization that helps researchers better discover and understand research articles through Smart Citations–citations that display the context of the citation and describe whether the article provides supporting or contrasting evidence. scite is used by students and researchers from around the world and is funded in part by the National Science Foundation and the National Institute on Drug Abuse of the National Institutes of Health.