The strategy of tuberculosis (TB) contact investigation is essential for enhancing disease detection. We conducted a cross-sectional study to evaluate the yield of contact investigation for new TB cases, estimate the prevalence of TB, and identify characteristics of index cases associated with infection among contacts of new cases notified between 2010 and 2020 in São Paulo, Brazil. Out of 186466 index TB cases, 131055 (70.3%) underwent contact investigation. A total of 652286 contacts were screened, of which 451704 (69.2%) were examined. Of these, 12243 were diagnosed with active TB (yield of 1.9%), resulting in a number needed to screen of 53 and a number needed to test of 37 to identify one new TB case. The weighted prevalence for the total contacts screened was 2.8% (95% confidence interval [CI]: 2.7%–2.9%), suggesting underreporting of 6021 (95% CI: 5269–6673) cases. The likelihood of TB diagnosis was higher among contacts of cases identified through active case-finding, abnormal chest X-ray, pulmonary TB, or drug resistance, as well as among children, adults, women, individuals in socially vulnerable situations, and those with underlying clinical conditions. The study highlights significant TB underreporting among contacts, recommending strengthened contact investigation to promptly identify and treat new cases.

We propose a hierarchical Bayesian model for analyzing multi-site experimental fMRI studies. Our method takes the hierarchical structure of the data (subjects are nested within sites, and there are multiple observations per subject) into account and allows for modeling between-site variation. Using posterior predictive model checking and model selection based on the deviance information criterion (DIC), we show that our model provides a good fit to the observed data by sharing information across the sites. We also propose a simple approach for evaluating the efficacy of the multi-site experiment by comparing the results to those that would be expected in hypothetical single-site experiments with the same sample size.

Human performance in cognitive testing and experimental psychology is expressed in terms of response speed and accuracy. Data analysis is often limited to either speed or accuracy, and/or to crude summary measures like mean response time (RT) or the percentage correct responses. This paper proposes the use of mixed regression for the psychometric modeling of response speed and accuracy in testing and experiments. Mixed logistic regression of response accuracy extends logistic item response theory modeling to multidimensional models with covariates and interactions. Mixed linear regression of response time extends mixed ANOVA to unbalanced designs with covariates and heterogeneity of variance. Related to mixed regression is conditional regression, which requires no normality assumption, but is limited to unidimensional models. Mixed and conditional methods are both applied to an experimental study of mental rotation. Univariate and bivariate analyzes show how within-subject correlation between response and RT can be distinguished from between-subject correlation, and how latent traits can be detected, given careful item design or content analysis. It is concluded that both response and RT must be recorded in cognitive testing, and that mixed regression is a versatile method for analyzing test data.

Classical factor analysis assumes a random sample of vectors of observations. For clustered vectors of observations, such as data for students from colleges, or individuals within households, it may be necessary to consider different within-group and between-group factor structures. Such a two-level model for factor analysis is defined, and formulas for a scoring algorithm for estimation with this model are derived. A simple noniterative method based on a decomposition of the total sums of squares and crossproducts is discussed. This method provides a suitable starting solution for the iterative algorithm, but it is also a very good approximation to the maximum likelihood solution. Extensions for higher levels of nesting are indicated. With judicious application of quasi-Newton methods, the amount of computation involved in the scoring algorithm is moderate even for complex problems; in particular, no inversion of matrices with large dimensions is involved. The methods are illustrated on two examples.

Van Breukelen offers a promising method for modeling both response speed and response accuracy. However, the underlying conception of both dependent measures is somewhat flawed, leading the author to conclude that the approach possesses limitations that, under revised assumptions, may not hold. The central misconception, and a set of related misconceptions, is addressed, and it is suggested that this approach holds a good deal of promise for application in the perceptual and cognitive sciences.

We extend dynamic generalized structured component analysis (GSCA) to enhance its data-analytic capability in structural equation modeling of multi-subject time series data. Time series data of multiple subjects are typically hierarchically structured, where time points are nested within subjects who are in turn nested within a group. The proposed approach, named multilevel dynamic GSCA, accommodates the nested structure in time series data. Explicitly taking the nested structure into account, the proposed method allows investigating subject-wise variability of the loadings and path coefficients by looking at the variance estimates of the corresponding random effects, as well as fixed loadings between observed and latent variables and fixed path coefficients between latent variables. We demonstrate the effectiveness of the proposed approach by applying the method to the multi-subject functional neuroimaging data for brain connectivity analysis, where time series data-level measurements are nested within subjects.

This study employs social cognitive theory to examine the dynamics of ethical climate, environmental passion, and low-carbon behaviours among Malaysian public servants based on data from 407 employees across 37 departments. Although ethical climate did not have a direct impact on low-carbon behaviour, a significant association with environmental passion was observed. Additionally, environmental passion exhibited a noteworthy relationship with low-carbon behaviour, and emerged as a mediator between ethical climate and low-carbon behaviour, with green mindfulness moderating this relationship. These findings underscore the importance of nurturing environmental passion and green mindfulness to promote low-carbon behaviour among employees and aid organisations in addressing environmental challenges. By addressing these empirical gaps, this study contributes to the literature on low-carbon behaviour and offers both theoretical insights and practical implications for sustainability initiatives.

Trust in national and local institutions is an essential component of democracy. The literature has dealt mainly with the former, while less attention has been given to the latter. This paper advances a novel theoretical approach to inquire about trust in local institutions, which is also used to test national ones. We posit that trust is affected by the perceptions individuals have of the physical space where they live. Both a) the perceived quality of life in the neighbourhood where individuals live and b) the neighbourhood (perceived) peripherality are hypothesized to affect trust in local (and to a lesser extent) national institutions. We test our hypotheses in Italy, over a large representative sample of more than 40.000 respondents. We show that both variables are crucial predictors of local trust, but only the perceived quality of life predicts national trust. Equally important, social, cultural and economic individual capital does not modify the relation.

Although the older adult population faces a higher risk of poverty compared to others, there is no clear picture of their poverty in Iran. The aim of this study was to measure multidimensional poverty and its related factors among Iranian older adults. This cross-sectional study was conducted from July to November 2019 and collected data by interviewing 1,280 participants in Tehran, Iran. To compute multidimensional poverty, four dimensions were used: health (disabilities), education, housing and standard of living. Single and multidimensional poverty and the joint distribution of deprivation were calculated. Multilevel logistic regression models were used to determine the relationship between predictor variables and outcome (multidimensional poverty). Multidimensional poverty among Tehran's older people was 59.0 per cent. The prevalence of health, housing, education and standard of living deprivations were 15.4, 25.3, 29.5 and 29.9 per cent, respectively. Furthermore, multivariate analysis shows that living with a spouse, being employed, and having health and social insurance coverage were protective factors, while being female was a risk factor for multidimensional poverty. Approximately 21 per cent of multidimensional poverty variance was attributed to the district level and the remaining was assigned to individual-level factors. This study showed that the older adults living in different areas of Tehran experience different aspects of poverty. So paying attention to the dimensions of multidimensional poverty can play an important role in customising the policies of each district. Also, the findings of this study on risk and protective factors of multidimensional poverty can be effective in designing and implementing interventions to mitigate poverty among the older adults.

The co-existence of under- and overweight at population level around the globe is well documented. However, this has yet to be explored using suitable statistical techniques in the context of Bangladesh. This study aimed to examine the prevalence and risk factors for being underweight and overweight or obese compared with normal weight in ever-married non-pregnant women aged 15–49 years in Bangladesh using data from the most recent Bangladesh Demographic and Health Survey conducted in 2014. Multilevel multinomial logistic regression (MLMLR) and quantile regression models were fitted to examine the associations of socioeconomic and individual-, household- and community-level factors on the nutritional status of women as measured by BMI. Overall, the prevalences of underweight, normal weight, overweight and obese women were 19%, 58%, 19% and 4%, respectively, in 2014. The MLMLR analysis revealed that women of young age, widowed/divorced/separated, having a larger family size and children aged ≤5 years in the household, currently amenorrhoeic and members of non-government organizations were at significantly increased risk of being underweight; those of older age, having higher parity, more educated, frequently watched TV and non-poor were more likely to be overweight or obese relative to normal BMI. Women from more affluent communities and urban areas were more likely to be overweight or obese relative to normal BMI than their counterparts from less-affluent and rural communities. Women’s nutritional status was found to be heterogeneous across the regions of the country. The findings indicate that, along with individual-level factors, community-level characteristics are also important in explaining women’s BMI in Bangladesh. The issue of under- and overweight or obesity among women in Bangladesh requires the immediate adoption of a public health policy for its mitigation. When developing intervention programmes, important determinants and uniform development of regions should be taken into consideration to combat the dual burden of under- and overweight among women in Bangladesh.

Economic progress in India over the past three decades has not been accompanied by a commensurate improvement in the nutritional status of children, and a disproportionate burden of undernutrition is still focused on socioeconomically disadvantaged populations in the poorest regions. This study examined the nutritional status of children under 3 years of age using data from the fourth round of Indian National Family Health Survey conducted in 2015–2016. Child undernutrition was assessed in a sample of 126,431 under-3 children using the anthropometric indices of stunting, underweight and wasting (‘anthropometric failure’) across 640 districts, 5489 primary sampling units and 35 states/UTs of India. Descriptive statistics were used to examine the regional pattern of childhood undernutrition. Multilevel logistic regression models were fitted to examine the adjusted effect of social group (tribal vs non-tribal) and economic, demographic and contextual factors on the risks of stunting, underweight and wasting accounting for the hierarchical nature of the data. Interaction effects were estimated to model the joint effects of socioeconomic position (household wealth, maternal education, urban/rural residence and geographical region) and social group (tribal vs non-tribal) with the likelihood of anthropometric failure among children. The burden of childhood undernutrition was found to vary starkly across social, economic, demographic and contextual factors. Interaction effects demonstrated that tribal children from economically poorer households, with less-educated mothers, residing in rural areas and living in the Central region of India had elevated odds of anthropometric deprivation than other tribal children. The one-size-fits-all approach to tackling undernutrition in tribal children may not be efficient and could be counterproductive.

Quantitative comparative social scientists have long worried about the performance of multilevel models when the number of upper-level units is small. Adding to these concerns, an influential Monte Carlo study by Stegmueller (2013) suggests that standard maximum-likelihood (ML) methods yield biased point estimates and severely anti-conservative inference with few upper-level units. In this article, the authors seek to rectify this negative assessment. First, they show that ML estimators of coefficients are unbiased in linear multilevel models. The apparent bias in coefficient estimates found by Stegmueller can be attributed to Monte Carlo Error and a flaw in the design of his simulation study. Secondly, they demonstrate how inferential problems can be overcome by using restricted ML estimators for variance parameters and a t-distribution with appropriate degrees of freedom for statistical inference. Thus, accurate multilevel analysis is possible within the framework that most practitioners are familiar with, even if there are only a few upper-level units.