A note on computing Louis’ observed information matrix identity for IRT and cognitive diagnostic models

Using Louis’ formula, it is possible to obtain the observed information matrix and the corresponding large-sample standard error estimates after the expectation–maximization (EM) algorithm has converged. However, Louis’ formula is commonly de-emphasized due to its relatively complex integration representation, particularly when studying latent variable models. This paper provides a holistic overview that demonstrates how Louis’ formula can be applied efficiently to item response theory (IRT) models and other popular latent variable models, such as cognitive diagnostic models (CDMs). After presenting the algebraic components required for Louis’ formula, two real data analyses, with accompanying numerical illustrations, are presented. Next, a Monte Carlo simulation is presented to compare the computational efficiency of Louis’ formula with previously existing methods. Results from these presentations suggest that Louis’ formula should be adopted as a standard method when computing the observed information matrix for IRT models and CDMs fitted with the EM algorithm due to its computational efficiency and flexibility.     

Convergence to the Truth Without Countable Additivity

Must probabilities be countably additive? On the one hand, arguably, requiring countable additivity is too restrictive. As de Finetti pointed out, there are situations in which it is reasonable to use merely finitely additive probabilities. On the other hand, countable additivity is fruitful. It can be used to prove deep mathematical theorems that do not follow from finite additivity alone. One of the most philosophically important examples of such a result is the Bayesian convergence to the truth theorem, which says that conditional probabilities converge to 1 for true hypotheses and to 0 for false hypotheses. In view of the long-standing debate about countable additivity, it is natural to ask in what circumstances finitely additive theories deliver the same results as the countably additive theory. This paper addresses that question and initiates a systematic study of convergence to the truth in a finitely additive setting. There is also some discussion of how the formal results can be applied to ongoing debates in epistemology and the philosophy of science.

     

Social value orientations and negotiator outcomes

Social value orientations (SVOs) are known to influence individual behaviour in outcome interdependent settings. By extending these findings to negotiation, this research investigates the relationship between own and partners' SVOs, negotiator strategies and outcomes. Results showed that cooperators, competitors and individualists could be distinguished in terms of initial demands and concessions. Competitors made higher initial demands and larger concessions than individualists or cooperators, suggesting that their ability to maximize outcome differences rests on whether structural features are congruent with this goal. The principal finding of this research was the demonstration that own and partners' SVO interact to determine outcomes. Results showed that the three SVO groups differed in terms of context sensitivity: competitor outcomes were invariant across partners; individualists achieved poor outcomes in negotiations with cooperators and, reciprocally, cooperators attained high outcomes in negotiations with individualists. Additionally, individualist outcomes worsened in their last negotiation, while those of cooperators differed as a function of role and partner's SVO. These results suggest that although the information used by individualists and cooperators differs, for both groups the cognitive representation of negotiations is a further factor influencing their outcomes.     

Readers and service commissioners require clear financial disclosures: Comment on innovation,research integrity,and change: A conflict of interest management framework for program developers (Sanders et al., 2019)

Sanders et al.'s proposal for a management framework for conflicting interests among program developers is very welcome. The underlying principles of such a framework must nevertheless prioritise the need for researchers and commissioners of services to make objective assessments of the impact of interventions reported in journal articles. This is particularly important in the field of randomised trials which may influence public sector expenditure. Using a strict definition derived from known financial conflict of interest, we have demonstrated that child-based effect sizes are much lower for independent studies than for studies with developer involvement. On this basis, we propose that journals publishing evaluations of psychosocial interventions should agree a standardised format for declarations of conflicts of interest based on that recommended by the International Committee of Medical Journal Editors.     

Reducing drug and alcohol use and improving well-being for Indigenous and non-Indigenous Australians using the Community Reinforcement Approach: A feasibility and acceptability study

There is a lack of evidence of effective and appropriate drug and alcohol treatment for Aboriginal and Torres Strait Islander peoples. This paper contributes to addressing the evidence gap by examining the feasibility and acceptability and conducting a pre/post-evaluation of the Aboriginal-adapted Community Reinforcement Approach (CRA) delivered in New South Wales, Australia. Aboriginal and non-Aboriginal clients (n = 55) received tailored CRA delivery between March and November 2013. Compared to the original US version, tailored CRA had reduced technical language, reduced number of treatment sessions, and the addition of group delivery option. An Australian training manual with local case studies was developed. Alcohol, Smoking and Substance Involvement Test (ASSIST), Kessler-5 (K-5) and the Growth Empowerment Measure were used. 58% of participants were followed-up at 3 months. Tailored CRA was feasible to deliver in a rural, community-based health setting, and rated by clients as highly effective and acceptable. CRA was associated with statistically significant reductions in the use of alcohol, tobacco, cannabis, amphetamine and over the counter medication, and levels of psychological distress, and an increase in levels of empowerment for Aboriginal and non-Aboriginal clients. This study provides evidence for the feasibility and acceptability of an Aboriginal-adapted psychological intervention addressing drug, alcohol and mental health outcomes.     

Changes in Religiosity Among Lesbian,Gay, and Bisexual Emerging Adults

Highly religious Americans are relatively likely to oppose lesbian, gay, and bisexual (LGB) rights and many churches are unwelcoming to sexual minorities, which may lead LGB Americans to retreat from religion. To assess this possibility, we investigate trajectories of religious change for sexual minorities and other emerging adults. We use two longitudinal data sources (National Longitudinal Study of Adolescent to Adult Health and the National Study of Youth and Religion) to explore how sexuality predicts the likelihood of decreasing religiosity in emerging adulthood. Results show that three different operationalizations of sexual minority status—attraction, behavior, and identity—are each strongly and consistently associated with disaffiliating from religion and declines in religious service attendance. On the other hand, sexual minority status has inconsistent and relatively small associations with changes in prayer. We conclude by discussing how these results further understanding of religion, sexual identity, and the current generation of emerging adults.