For large-scale surveys such as the National Health and Aging Trends Study (NHATS), investigators may wish to combine data from two (or more) cohorts in a single analysis to obtain larger sample sizes. Unfortunately, it is not possible to combine the 2011 and 2015 NHATS cohorts while retaining the sample weights. We applied Bayesian hierarchical models with poststratification as an alternative strategy for obtaining population-based estimates from NHATS. As proof of principle, we compared prevalence estimates of frailty obtained from our Bayesian approach with those obtained from the 2011 and 2015 cohorts using the NHATS sample weights. Once validated, we applied our strategy to combine the cohorts into a single analytical dataset without overlap of participants, and generated Bayesian estimates of frailty for the combined cohort. Estimates from the Bayesian model closely matched the weighted NHATS estimates. The ability to combine cohorts while generating population-based estimates will allow investigators to address questions that require larger sample sizes, thereby enhancing the value of NHATS to the scientific community.

Objective: This study presents the "Post-Separation Family" (PSF) module within the panel study "Growing Up in Germany" (German title: Aufwachsen In Deutschland: Alltagswelten [AID:A]) and discusses its contribution to providing information on diverse post-separation family constellations in Germany. Background: Rates of separation and divorce are persistently high in Western societies, and post-separation families are increasingly diverse and complex. However, official statistics and large-scale surveys in Germany often lack detailed data on post-separation family constellations, particularly regarding non-resident parents. Method: The PSF module in AID:A collects comprehensive information on everyday practices, conflicts, and parental care involvement of a wide range of family constellations. It enables the classification of diverse family constellations, such as single parents and stepfamilies, and includes data on non-resident parents. Results: In 2019, about 23% of families with minors were post-separation families with a non-resident parent (i.e., 17% single-parents, 6% stepfamilies). About 12% of minors in post-separation families practiced a shared care arrangement according to the PSF module data. Key measures on post-separation care arrangements can be further linked to the broad spectrum of outcomes collected in AID:A (e.g., economic hardship, subjective well-being, and parental education). Conclusion: The PSF module represents a concise instrument for analyzing post-separation family diversity in Germany, which could lay the groundwork for national and international comparisons of diverse family constellations if adopted in other surveys.

Estimating the risk of re-identification probabilistically is well-developed for the case of a random representative sample drawn from the general population, such as large-scale government surveys conducted regularly at National Statistical Institutes. Recent work extended this procedure to assess the risk of re-identification in non-probability subpopulation registers such as a cancer register. In this paper, we extend this work further to the case of samples drawn from registers or more generally to non-probability samples, such as those used in opt-in panels at survey organizations. The assumption is that membership to the subpopulation register is not known and the sampling mechanism is also unknown. We show how to assess the risk of re-identification for these types of non-probability samples using a probability-based reference sample to infer population parameters under the probabilistic modeling framework. We demonstrate with a simulation study and a real application on the 2021 Survey of Doctoral Recipients drawn from a subpopulation register of all PhD recipients from an accredited US institution.

Objective Nutrition literacy is critical for establishing healthy dietary behaviors during childhood, yet research on this topic among primary school students in rapidly urbanizing China remains limited. The aim of this large-scale survey is to assess the current status and identify key influencing factors of nutrition literacy among primary school students in Nanshan District, Shenzhen City. Methods A cross-sectional survey was conducted in October 2024 utilizing a cluster random sampling method. A total of 2,423 students from 21 public primary schools participated. Data were collected using the validated “Nutrition Literacy and Dietary Behavior Questionnaire for School-aged Children”, which evaluated four dimensions: nutrition knowledge and concepts, food selection, food preparation, and food intake. Statistical analyses included descriptive statistics, correlation analysis, and binary logistic regression. Results Participants achieved a mean total nutrition literacy score of 69.93 ± 8.75, with 30.38% meeting the criterion for adequate nutrition literacy (score ≥75). Interdimensional analysis revealed statistically significant positive correlations among all four domains ( r = 0.198 ~ 0.363, p < 0.001). Multivariable logistic regression identified grade level—representing individual-level factors (Grade 5 vs. Grade 3 OR = 0.626, 95% CI: 0.500–0.783), high household economic status—family-level factors (OR = 1.389, 95% CI: 1.139–1.649), and participation in school activities including nutrition-related activities (OR = 1.346, 95% CI: 1.125–1.611) and regular weight monitoring (OR = 1.346, 95% CI: 1.125–1.611) as key predictors of adequate nutrition literacy. Conclusion Nutrition literacy among primary school students in Nanshan District requires substantial improvement and is influenced by factors at individual, familial, and institutional levels. These findings suggest the necessity of developing a comprehensive, student-centered intervention model that integrates family-school collaboration to effectively enhance nutritional literacy.

The rise of Generative Artificial Intelligence (GAI) is reshaping higher education by influencing how university students access, process, and apply knowledge. This study explores university students’ perceptions and usage patterns of GAI, with a specific focus on ChatGPT, through a large-scale survey of 1,366 students from 24 Italian universities. Employing cluster analysis, the research identifies four distinct student groups based on their digital skills and attitudes toward GAI: Indifferent, Fearful, Pragmatics, and Enthusiasts. Findings indicate significant variations in students’ engagement with GAI, with factors such as gender, degree programs, and technological proficiency shaping their perspectives. While some students embrace AI as a valuable educational tool, others express ethical and academic integrity concerns. Notably, digital competence correlates with AI adoption, with STEM students exhibiting higher engagement. The study highlights the need for universities to address both the opportunities and challenges of GAI, proposing evidence-based policies to foster responsible integration into academic settings. By providing a nuanced understanding of student interactions with AI, this research aims to inform pedagogical strategies and regulatory frameworks to balance innovation with ethical considerations in higher education.

Tropical coastal seas are overlooked hotspots of Pseudo-nitzschia diversity.

Who uses AI in research, and for what? Large-scale survey evidence from Germany

Parenting behavior is a central determinant of childhood development and is thus deserving of more scientific attention. In the present article, we constructed an ultra-short scale for the assessment of parenting styles, the Parenting Scale 4 (PS-4). To this end, we analyzed large samples of parent–child dyads—one representative of the German general population (Sample 1), the other representative for the German federal state Lower Saxony (Sample 2). We applied an algorithm-based scale-shortening technique in Sample 1 and confirmed the resulting model in Sample 2, finding excellent model fit and—given the extreme brevity—acceptable reliability. Furthermore, we show that the model is invariant across parent and child genders. Correlations with the Strengths and Difficulties Questionnaire remain virtually unchanged compared to a longer version of the Parenting Scale, which is evidence for the PS-4’s validity. Overall, the PS-4 can be recommended for the assessment of parenting behavior, particularly in large-scale surveys with time constraints.

In democracies, public support for policies is crucial to their legitimacy, and the absence can impede necessary reforms, which are needed to keep most disastrous effects of climate change at bay. This study emphasizes the role of policy goal communication as an important but often overlooked dimension of climate policy discourse, arguing that how policy proposals are linked to their intended goals during political debates directly influences public support. We present findings from a novel large-scale (n = 5,655) survey with an embedded randomized experiment that systematically manipulates the type of goal communication and the level of policy goal ambition. Unlike previous studies, the expected policy effectiveness was generated and agreed upon through an iterative process of expert elicitation to provide respondents with a most accurate statement. The results highlight that the presentation of information on policy effectiveness as an inherent element of policy design and not as the larger context in which the policy is proposed significantly increases support for highly ambitious policies renewable energy policies. This study implies that policymakers seeking to promote ambitious climate policies should focus on directly linking proposed policies with the goals these policies should reach.

High-quality anthropometric data are critical for accurately monitoring child nutritional outcomes and informing policy decisions, yet inconsistencies in measurement and reporting across large-scale surveys continue to challenge data reliability. This research assesses the quality of height-for-age (HAZ), weight-for-age (WAZ), and weight-for-height (WHZ) z-scores based on a repeated cross-sectional analysis of two rounds of the National Family Health Survey (NFHS-4, 2015-2016 and NFHS-5, 2019-2021), examining improvements, persistent gaps, and regional disparities. We have used WHO-recommended diagnostics including digit preference, age-heaping, completeness of measurements, biologically implausible values, and distributional properties of z-scores to evaluate the plausibility of anthropometric data and generate state-level rankings to compare transitions across rounds. The results indicate modest national-level improvements in data quality in NFHS-5, particularly reductions in digit preference and implausible values; however, substantial inter-state variation remains, with some states demonstrating clear progress while others continue to exhibit measurement anomalies. The completeness of date of birth improved from 99.0% in NFHS-4 to 99.9% in NFHS-5, while completeness of anthropometric measurements declined from 98.5% to 96.6%. Digit preference for height decreased from 15.2% to 14.4%, and the proportion of biologically implausible HAZ values declined from 3.4% to 2.3%. However, the standard deviation of HAZ increased from 1.77 to 1.85 and that of WHZ from 1.40 to 1.50, indicating persistent measurement variability. Transitions in HAZ rankings further reveal mixed patterns of advancement and stagnation, with regional clustering of improvements more evident in certain parts of the country. Overall, while NFHS-5 reflects progress in anthropometric data quality, key challenges persist related to inconsistent adherence to measurement protocols, variable field performance, and inadequate supervisory oversight. Strengthening training, standardizing procedures, and reinforcing monitoring mechanisms are essential for achieving more reliable anthropometric data, thereby enhancing the accuracy of child nutrition estimates and supporting more evidence-based policy interventions in India.

1. DNA metabarcoding-high-throughput sequencing of barcode regions from bulk samples-has become a key tool for insect biodiversity assessment. Yet, how methodological choices affect the accuracy of metabarcoding data remains insufficiently explored. In this paper, we ask: (1) How does the lysis method (non-destructive lysis vs. destructive homogenization) affect community recovery? (2) How comprehensively does metabarcoding capture species richness? (3) To what extent can spike-ins improve abundance estimates? (4) How accurately can species abundances be estimated?2. We evaluated the accuracy of insect metabarcoding using 4,749 bulk samples from a large-scale biodiversity survey subjected to mild lysis. Of these samples, 856 were also homogenized, allowing a systematic comparison of the effect of alternative treatments. To potentially improve abundance estimates, we added six biological spike-ins (i.e., foreign insects) to all samples, and two synthetic spike-ins (artificial DNA fragments) to the homogenization treatment. In addition, we established the contents of 15 samples by individually barcoding all specimens, enabling direct assessment of occurrence and abundance estimates.3. Our results revealed consistent differences between destructive and non-destructive treatments. While both methods reliably detected the majority of species, small and soft-bodied taxa were more often recovered after mild lysis than after homogenization, while the reverse was true for heavily sclerotized, hairy, and large taxa. Using biological spike-ins for calibration reduced the variance in read numbers per specimen considerably, especially in homogenized samples, while synthetic spike-ins were less effective. In a Bayesian analysis, where species data were matched to the best-fitting spike-in calibration curve, accurate abundance estimates (+/-1 individual) were obtained for 72.9% of species occurrences.4. Our results show that it is possible to obtain reasonably accurate abundance estimates from metabarcoding data, and that mild lysis and homogenization result in different taxon-specific biases in terms of occurrence data, with neither method outperforming the other. Accuracy is improved by homogenization rather than mild lysis of samples, and by the use of biological rather than synthetic spike-ins. Together, these findings provide a major step towards robust, quantitative biodiversity monitoring using DNA-metabarcoding.

Purpose This study aims to explore the preference differences for two mainstream types of tourism digital information, artificial intelligence-generated content (AIGC) and user-generated content (UGC), between tourists of different age groups and highlights the key role of episodic future thinking in shaping tourists’ preferences. Design/methodology/approach A sequential explanatory mixed-methods design was used. It included a scenario-based experiment, a large-scale survey and adapted autobiographical interviews. Findings Older tourists show a stronger preference for AIGC, while young tourists prefer UGC. Age-related differences in episodic future thinking (contextual clarity and self-engagement level) lead to varying information demands in terms of characteristics and requirements. These demands, in turn, shape tourists’ overall information preferences. Additionally, travel experience further moderates these information preference effect. Originality/value This study offers new insights into information preference mechanisms and enriches age-based tourist information behavior theory. The findings also provide practical guidance for the development of age-appropriate tourism information products and services.

ABSTRACT Despite several decades of large-scale surveys in educational contexts, our understanding of what teachers and students do in classrooms in the process of teaching and learning different school subjects is still limited. In this study, we developed and used a quantitative approach inspired by the Q-method to map how Danish teachers (N = 289) and lower secondary students (grades 5, 6, 7, and 8, N = 3025) experience teaching practices in three school subjects: science, Danish, and mathematics. Based on a Q-method-inspired inter-person correlation factor analysis of the data, we identified different teacher and student profiles for each subject. The results show that it is possible to identify common practices among teachers and among students, but they also indicate important discrepancies between how teachers and students perceive the same practices. We discuss our findings and suggest ideas for further research.

Background: Measuring glycated hemoglobin (HbA1c) is essential for assessing long-term glycemic control, yet direct testing remains expensive and underutilized in many large-scale health surveys and resource-constrained settings. This study aims to (i) deliver a highly accurate and interpretable ML model for predicting HbA1c from routinely collected clinical, biochemical, and demographic data, (ii) reduce dependency on extensive laboratory panels by identifying a compact, cost-efficient subset of key predictors, and (iii) establish a transferable, explainable modeling framework applicable across chronic disease biomarkers. Unlike prior HbA1c prediction studies that focus primarily on classification or accuracy-driven models, this work introduces a unified framework for continuous HbA1c regression that jointly integrates cost-oriented feature parsimony, stratified regression validation, and explainability by design. Methods: We aggregated data from the National Health and Nutrition Examination Survey (NHANES) cycles 2007-2020, encompassing 66,148 records and 224 candidate features. We implemented a two-stage feature selection pipeline: Incremental Correlation Selection (ICS) to narrow the variable space, followed by Recursive Feature Elimination with Cross-Validation (RFECV) to isolate the most informative features. Model interpretability was assessed using partial dependence plots and feature importance analysis. Results: The optimal model, LightGBMRegressor with most-frequent imputation, achieved R2 = 0.7161, MAE = 0.334, MSE = 0.304, and MAPE = 5.56%, while using only 40 selected features. Interpretability analysis revealed clinically coherent relationships that align with physiological expectations. Discussion: The proposed framework maintains robust predictive performance while substantially reducing the number of required input features, enabling cost-efficient HbA1c estimation together with transparent, physiologically coherent model insights. By consolidating continuous HbA1c prediction, cost-aware feature selection, stratified evaluation, and explainability within a single pipeline are enhanced. Conclusions: This study advances beyond existing approaches and offers a practical blueprint for scalable biomarker estimation in population health and clinical decision-support applications. Its explainable, efficient, and generalizable design positions it as a strong candidate for clinical decision-support and population-health applications.

Within emergency management, few studies have analyzed the shifting landscape of the workforce in the United States. As emergency management is an evolving field, it is important to note changes in the profession. The purpose of this study was to examine the current emergency management workforce, specifically analyzing its demographic breakdowns, organizational concerns, and community-based dynamics. The intent was to determine if the workforce is representative of the communities they work in. Disasters are known to disproportionately impact vulnerable populations, making it more important for emergency managers to be demographically and functionally diverse to effectively reach and prepare these communities. Using a mixed-methods approach, data were gathered through five focus group sessions, and using themes from the focus groups, a large-scale survey was designed and disseminated to seven emergency management organizations across the country. The survey collected responses for three weeks, and participants were offered the opportunity to enter a raffle for a $100 gift card. The focus group data were analyzed using Atlas.ti, and the survey data were analyzed using Microsoft Excel. In total, 20 emergency managers participated in the focus groups, and 232 participants completed the survey. Our analysis showed high levels of concern regarding an overall lack of funding and resources within organizations. Other concerns included insufficient representation of historically underrepresented populations, limiting emergency managers' capacity to effectively engage their communities. This study identifies key challenges within the evolving emergency management workforce, most notably limited funding, the risk of burnout, underrepresentation of minority groups in leadership, and lack of standardization. At the same time, encouraging trends are emerging, including greater gender diversity and growing participation from younger professionals. The findings provide a foundational overview to guide future research on strengthening and supporting the workforce and the communities they serve.

Purpose This study investigates the relationship between artificial intelligence (AI)-related system characteristics and two interpersonal states commonly associated with effective teamwork, namely employee well-being and mutual trust. While generative AI has shown potential to improve organizational performance, its specific effects on internal team-based working relationships remain underexplored. Design/methodology/approach A theoretical model is developed to explore the influence of three antecedent variables, quality of information, system quality and generative AI use, on collaboration within teams. Collaboration is operationalized using two key constructs: employee well-being and mutual trust. The model is empirically tested using data from a large-scale survey of 208 professionals working in team-based environments. Data analysis is conducted using partial least squares structural equation modeling (PLS-SEM). Findings The results confirm that all three antecedent variables positively influence team-based collaboration dynamics. Specifically, the use of generative AI chatbots, such as ChatGPT, is shown to enhance employee well-being and foster mutual trust within teams, both of which act as interpersonal enablers of team collaboration. These outcomes suggest that the integration of high-quality AI tools can meaningfully support collaborative processes in professional settings. Originality/value This study contributes to the emerging field of generative AI research by shifting the focus from performance outcomes to collaboration mechanisms within teams. It offers practical implications for managers seeking to optimize teamwork in AI-enabled environments, including investing in system quality, redesigning workflows to integrate AI effectively and promoting a culture of trust and transparency around AI adoption.

Escherichia albertii and Escherichia fergusonii have recently been recognized as emerging pathogens in humans and animals. E. albertii is a causative agent of foodborne infections in humans, but many aspects of its transmission routes and its prevalence in farm animals remain unclear. E. fergusonii has been isolated mainly from extraintestinal infections in humans and various animal species, but no large-scale surveys of its prevalence in farm animals have been reported. In this study, we isolated E. albertii and E. fergusonii from 3,975 fecal samples collected from farm animals throughout Japan. The prevalence of E. albertii was 3.1% (57/1,838) in swine, 2.2% (12/548) in poultry, and 0.4% (7/1,589) in cattle, while that of E. fergusonii was 37.4% (688/1,838) in swine, 46.2% (253/548) in poultry, and 17.1% (272/1,589) in cattle. We also isolated E. coli from fecal samples from which E. albertii or E. fergusonii was isolated and compared the antimicrobial resistance profiles of the E. albertii/E. fergusonii strains with those of the E. coli strains that were presumed to have experienced the same antimicrobial selection pressure in the farm animals. The antimicrobial resistance profiles of the E. fergusonii and E. coli strains isolated from the same swine fecal samples were similar, but those of the E. albertii and E. coli strains differed significantly (the former strains were highly susceptible). This suggests that E. albertii has not coexisted with E. coli for as long as E. fergusonii.

The concept of the ecological niche, defined as the basic habitat requirements for a species, is central to understanding species geographic distributions and predicting their responses to environmental change. However, identifying the essential niche for large regional communities remains a challenge because niche axes can be "hidden" by the complexity of the underlying ecological processes. Here, applying advanced species distribution modelling to nationwide environmental DNA survey data, we identified hidden niche axes of the Japanese coastal fish community and investigated the response diversity to these axes. Our survey detected 1,220 coastal fish species. The hidden niche axes collectively explained most of the variation in fish biodiversity and revealed five biogeographic boundaries for the regional community. These niches of the Japanese fish community may primarily relate to several processes due to ocean currents, such as environmental filters, transport from source areas and dispersal barriers. We also found that the response diversity to niche axes was positively correlated with species richness, although local communities with particularly high response diversity were geographically biased. A better understanding of the niche axes of the regional ecological community should help to mitigate the loss of biodiversity and ecosystem services caused by ongoing environmental change.

The successful integration of autonomous vehicle (AV) technologies into future mobility systems depends not only on technological maturity but also on user acceptance and perceived value. While existing research has identified several demographic determinants of AV acceptance, the role of educational background—particularly differences between humanities and STEM graduates—has received limited attention within the context of user-centred mobility research. This study examines how educational background and gender influence attitudes toward autonomous vehicle technologies using a large-scale survey conducted in Hungary (N = 8663). The analysis combines non-parametric statistical tests with effect size measures, exploratory factor analysis, and structural equation modelling (SEM) to capture both group differences and underlying attitudinal mechanisms. The results indicate no meaningful differences between humanities and STEM graduates in overall acceptance of autonomous vehicles or trust in the technology. Statistically significant differences are observed only in two dimensions: willingness to spend on autonomous driving features and expectations regarding improved travel speed. However, effect size analyses reveal that these differences are negligible in practical terms, indicating substantial overlap in user attitudes. SEM results show that educational background does not directly determine acceptance of autonomous vehicle technologies. Instead, its influence is mediated through three latent attitude dimensions relevant for electric and autonomous mobility adoption: willingness to invest, functional expectations (e.g., time savings and convenience), and safety orientation. Humanities graduates—especially men—exhibit slightly higher financial openness toward autonomous features, whereas STEM graduates place greater emphasis on functional performance. Safety-related attitudes play a central mediating role, with gender-specific patterns. By integrating large-sample effect size interpretation with SEM-based modelling, this study provides a nuanced understanding of user acceptance of autonomous vehicle technologies. The findings suggest that differences between educational groups reflect variations in attitudinal emphasis rather than fundamental divides, offering relevant insights for user-centred AV development, mobility policy design, and communication strategies in the transition toward automated and electric mobility systems.

Assessments of extreme psychological constructs often face a persistent challenge: the ceiling effect, in which a significant proportion of respondents select the highest score on a scale, thus obscuring meaningful variation within the population. This effect may have profound consequences in studies of extreme psychological constructs. To address this limitation, we present a novel framework that integrates Multistage Testing (MST) with open-ended questions that are automatically analyzed by large language models (LLMs). This hybrid approach adapts the survey questions to the respondent while leveraging LLMs to efficiently and reliably interpret free-text answers from large-scale online surveys. Using a case study on aversion toward cockroaches, we show how our method can effectively eliminate extreme ceiling effects, revealing hidden data distributions that are often obscured by extreme responses to conventional Likert-type survey questions. We also validate our method by comparing LLM performance to expert human annotations. This demonstrates the consistency and reliability of LLMs in evaluating free-text answers. This framework offers a generalizable methodology that enables more precise and sensitive quantitative measurement of extreme psychological constructs, allowing researchers to study topics that until now were inaccessible due to significant, inherent ceiling effects.