Explore the vast range of titles available.

We discuss prediction of random effects and of expected responses in multilevel generalized linear models. Prediction of random effects is useful for instance in small area estimation and disease mapping, effectiveness studies and model diagnostics. Prediction of expected responses is useful for planning, model interpretation and diagnostics. For prediction of random effects, we concentrate on empirical Bayes prediction and discuss three different kinds of standard errors; the posterior standard deviation and the marginal prediction error standard deviation (comparative standard errors) and the marginal sampling standard deviation (diagnostic standard error). Analytical expressions are available only for linear models and are provided in an appendix . For other multilevel generalized linear models we present approximations and suggest using parametric bootstrapping to obtain standard errors. We also discuss prediction of expectations of responses or probabilities for a new unit in a hypothetical cluster, or in a new (randomly sampled) cluster or in an existing cluster. The methods are implemented in gllamm and illustrated by applying them to survey data on reading proficiency of children nested in schools. Simulations are used to assess the performance of various predictions and associated standard errors for logistic random-intercept models under a range of conditions.

Context effects, where a characteristic of an upper-level unit or cluster (e.g., a country) affects outcomes and relationships at a lower level (e.g., that of the individual), are a primary object of sociological inquiry. In recent years, sociologists have increasingly analyzed such effects using quantitative multilevel modeling. Our review of multilevel studies in leading sociology journals shows that most assume the effects of lower-level control variables to be invariant across clusters, an assumption that is often implausible. Comparing mixed-effects (random-intercept and slope) models, cluster-robust pooled OLS, and two-step approaches, we find that erroneously assuming invariant coefficients reduces the precision of estimated context effects. Semi-formal reasoning and Monte Carlo simulations indicate that loss of precision is largest when there is pronounced cross-cluster heterogeneity in the magnitude of coefficients, when there are marked compositional differences among clusters, and when the number of clusters is small. Although these findings suggest that practitioners should fit more flexible models, illustrative analyses of European Social Survey data indicate that maximally flexible mixed-effects models do not perform well in real-life settings. We discuss the need to balance parsimony and flexibility, and we demonstrate the encouraging performance of one prominent approach for reducing model complexity.

This paper introduces the mini-symposium on Qualitative Comparative Analysis (QCA) and set-theoretic methods, both crisp sets and fuzzy sets, and situates the different contributions in a wider methodological debate concerning cross-case analysis. The paper argues that QCA is not just a set of techniques, but a distinctive research approach, with its own goals and set of assumptions. Concerning the wide methodological debate, special attention is paid to the added value of QCA and specific innovations introduced in the mini-symposium. This essay contrasts the conventional template for conducting social inquiry and the alternate template provided by configurational, case-oriented analytic methods, first formalized in The Comparative Method. The essential contrasts address the fundamental building blocks of social research, ranging from the definition of relevant cases to the understanding of social causation. The alternate template described in this essay provides a much stronger basis for the articulation of within-case and cross-case analysis than is offered by the conventional template. This article provides a first systematic mapping of QCA applications, building upon a database of 313 peer-reviewed journal articles. We find out that the number of QCA applications has dramatically increased during the past few years. The mapping also reveals that csQCA remains the most frequently used technique, that political science, sociology, and management are the core disciplines of application, that macrolevel analyses, medium-N designs, and a mono-method use of QCA remain predominant. A particular focus is also laid on the ratio between the number of cases and number of conditions and the compliance to benchmarks in this respect. QCA's ability of addressing complex theoretical expectations and taking account of configurational relationships is rarely fully exploited. Assessing comparative welfare-state research, which has employed QCA, we find that only about half of the studies reviewed have expressed complex theoretical propositions in set-theoretical terms, revisited cases subsequent to the formal analysis, or subjected findings to robustness checks. We discuss the relevance of each of these three aspects and argue that carefully considering these will improve the quality of QCA applications. Contrasting insights that can be gained from large-N QCA and econometric analysis, we outline two novel ways to integrate both modes of inquiry. The first introduces QCA solutions into a regression model, while the second draws on recent work in lattice theory to integrate a QCA approach with a regression framework. These approaches allow researchers to test QCA solutions for robustness, address concerns regarding possible omitted variables, establish effect sizes, and test whether causal conditions are complements or substitutes, suggesting that an important way forward for set-theoretic analysis lies in an increased dialogue that explores complementarities with existing econometric approaches. This paper discusses five strategies to deal with five types of errors in Qualitative Comparative Analysis (QCA): condition errors, systematic errors, random errors, calibration errors, and deviant case errors. These strategies are the comparative inspection of complex, intermediary, and parsimonious solutions; the use of an adjustment factor, the use of probabilistic criteria, the test of the robustness of calibration parameters, and the use of a frequency threshold for observed combinations of conditions. The strategies are systematically reviewed, assessed, and evaluated as regards their applicability, advantages, limitations, and complementarities. Current standard practices put sufficiency at the core of Qualitative Comparative Analysis (QCA), while the analysis of necessity is limited to the test for necessary conditions. Here, we argue that the possibilities of QCA in the latter domain are much greater. In particular, it can be used to empirically confront theories centered on necessary relations and that involved various conditions. A new operation, labeled the \"systematic necessity assessment,\" is therefore introduced. To show its added value, a published QCA study that confronts theories centered on necessary relations but using the regular minimization is replicated. Limited diversity is among the most understudied methodological challenges. QCA allows for a more conscious treatment of logical remainders than most other comparative methods. The current state of the art is the Standard Analysis (Ragin 2008; Ragin and Sonnett 2004). We discuss two of its pitfalls, both rooted in the primacy given to parsimony. First, the Standard Analysis is no safeguard against untenable assumptions. As a remedy, we propose the Enhanced Standard Analysis (ESA). Second, researchers should consider including theoretically sound counterfactual claims even if they do not contribute to parsimony. We label this Theory-Guided Enhanced Standard Analysis (TESA). This paper aims at strengthening causal inference in necessary condition research. We demonstrate how process tracing based on purposefully selected cases can complement findings on cross-case patterns identified with Qualitative Comparative Analysis (QCA). Using an empirical example, we discuss the meaning of typical and deviant cases in analyses of necessity, develop formulas for identifying the most typical and most deviant cases, and detail the implications of so-called SUIN conditions for meaningful case selection. In addition, we clarify various viable variants of comparative process tracing and formulas for identifying the best-matching pairs of cases. Adapted from the source document.

Medical documentation plays a crucial role in clinical practice, facilitating accurate patient management and communication among health care professionals. However, inaccuracies in medical notes can lead to miscommunication and diagnostic errors. Additionally, the demands of documentation contribute to physician burnout. Although intermediaries like medical scribes and speech recognition software have been used to ease this burden, they have limitations in terms of accuracy and addressing provider-specific metrics. The integration of ambient artificial intelligence (AI)-powered solutions offers a promising way to improve documentation while fitting seamlessly into existing workflows. This study aims to assess the accuracy and quality of Subjective, Objective, Assessment, and Plan (SOAP) notes generated by ChatGPT-4, an AI model, using established transcripts of History and Physical Examination as the gold standard. We seek to identify potential errors and evaluate the model's performance across different categories. We conducted simulated patient-provider encounters representing various ambulatory specialties and transcribed the audio files. Key reportable elements were identified, and ChatGPT-4 was used to generate SOAP notes based on these transcripts. Three versions of each note were created and compared to the gold standard via chart review; errors generated from the comparison were categorized as omissions, incorrect information, or additions. We compared the accuracy of data elements across versions, transcript length, and data categories. Additionally, we assessed note quality using the Physician Documentation Quality Instrument (PDQI) scoring system. Although ChatGPT-4 consistently generated SOAP-style notes, there were, on average, 23.6 errors per clinical case, with errors of omission (86%) being the most common, followed by addition errors (10.5%) and inclusion of incorrect facts (3.2%). There was significant variance between replicates of the same case, with only 52.9% of data elements reported correctly across all 3 replicates. The accuracy of data elements varied across cases, with the highest accuracy observed in the \"Objective\" section. Consequently, the measure of note quality, assessed by PDQI, demonstrated intra- and intercase variance. Finally, the accuracy of ChatGPT-4 was inversely correlated to both the transcript length (P=.05) and the number of scorable data elements (P=.05). Our study reveals substantial variability in errors, accuracy, and note quality generated by ChatGPT-4. Errors were not limited to specific sections, and the inconsistency in error types across replicates complicated predictability. Transcript length and data complexity were inversely correlated with note accuracy, raising concerns about the model's effectiveness in handling complex medical cases. The quality and reliability of clinical notes produced by ChatGPT-4 do not meet the standards required for clinical use. Although AI holds promise in health care, caution should be exercised before widespread adoption. Further research is needed to address accuracy, variability, and potential errors. ChatGPT-4, while valuable in various applications, should not be considered a safe alternative to human-generated clinical documentation at this time.

Missing data are a common occurrence in survey-based research studies in education, and the way missing values are handled can significantly affect the results of analyses based on such data. Despite known problems with performance of some missing data handling methods, such as mean imputation, many researchers in education continue to use those methods as a quick fix. This study reviews the current literature on missing data handling methods within the special context of education research to summarize the pros and cons of various methods and provides guidelines for future research in this area.

The aim of this study was to compare the agreement of the movement demands data during a soccer match (total distance, distance per minute, average speed, maximum speed and distance covered in different speed sectors) between an optical tracking system (Mediacoach System) and a GPS device (Wimu Pro). Participants were twenty-six male professional soccer players (age: 21.65 ± 2.03 years; height: 180.00 ± 7.47 cm; weight: 73.81 ± 5.65 kg) from FC Barcelona B, of whom were recorded a total of 759 measurements during 38 official matches in the Spanish second division. The Mediacoach System and the Wimu Pro were compared using the standardized mean bias, standard error of estimate, intraclass correlation coefficients (ICC), coefficient of variation (%), and the regression equation to estimate data for each variable. In terms of agreement between systems, the magnitude of the ICC was almost perfect (> 0.90-1.00) for all variables analyzed. The coefficient of the variations between devices was close to zero (< 5%) for total distance, distance per minute, average speed, maximum speed, and walking and jogging, and between 9% and 15% for running, intense running, and sprinting at low and at high intensities. It can be observed that, compared to Wimu Pro the Mediacoach System slightly overestimated all the variables analyzed except for average speed, maximum speed, and walking variables. In conclusion, both systems can be used, and the information they provide in the analyzed variables can be interchanged, with the benefits implied for practitioners and researchers.

Taking notes on laptops rather than in longhand is increasingly common. Many researchers have suggested that laptop note taking is less effective than longhand note taking for learning. Prior studies have primarily focused on students' capacity for multitasking and distraction when using laptops. The present research suggests that even when laptops are used solely to take notes, they may still be impairing learning because their use results in shallower processing. In three studies, we found that students who took notes on laptops performed worse on conceptual questions than students who took notes longhand. We show that whereas taking more notes can be beneficial, laptop note takers' tendency to transcribe lectures verbatim rather than processing information and reframing it in their own words is detrimental to learning.

This study investigates the impact of information and communication technology (ICT) on carbon dioxide emissions for a panel of 91 countries over the period 1990 to 2017. The study constructs an ICT index through principal component analysis and tests for the presence of cross-sectional dependence (CSD) in the data. The study employs pooled ordinary least squares, fixed-effects model, and system-generalized method of moments estimation techniques with panel-corrected standard errors (PCSE) to tackle the issues of CSD in the data. The findings of the study show that ICT reduces CO 2 emissions for the full sample of countries. However, the comparative study of developed and developing countries depicts that ICT encourages environmental sustainability in developed countries whereas opposite results are found for developing countries. Moreover, presence of the environmental Kuznets curve is confirmed for the full sample as well as for developed and developing countries. It suggests that with higher levels of development of a country, it would be possible to contribute towards environmental sustainability along with ICT diffusion. Therefore, the outcome of this study may be helpful for policymaker and policies may be designed to encourage ICT investments in developing countries, as ICT will take care of environmental sustainability with higher levels of development.

Disentangling the causal impact of media reporting from the impact of the events being reported is challenging. We solve this problem by comparing the behaviors of investors with access to different media coverage of the same information event. We use zip codes to identify 19 mutually exclusive trading regions corresponding with large U.S. cities. For all earnings announcements of S&P 500 Index firms, we find that local media coverage strongly predicts local trading, after controlling for earnings, investor, and newspaper characteristics. Moreover, local trading is strongly related to the timing of local reporting, a particular challenge to nonmedia explanations.

Epidemiological studies frequently use indices of adiposity related to mortality. However, no studies have validated prediction equations for body composition in adult cancer survivors. We aimed to develop and cross-validate prediction equations for body fat mass (BFM), lean body mass (LBM), trunk fat mass (TFM), and appendicular lean mass (ALM) in adult cancer survivors using sociodemographic, anthropometric, and laboratory test data. This study included adult cancer survivors from the Korean National Health and Nutrition Examination Survey 2008–2011 with complete data on Dual-energy X-ray absorptiometry (DXA) measurements. A total of 310 participants were randomly divided into development and cross-validation groups (5:5 ratio). Age, height, weight, waist circumference, serum creatinine levels, and lifestyle factors were included as independent variables The predictive equations were developed using a multiple linear regression and their predictive performances were primarily evaluated with R 2 and Concordance Correlation Coefficient (CCC). The initial equations, which included age, height, weight, and waist circumference, showed different predictive abilities based on sex for BFM (total: R 2 = 0.810, standard error of estimate [SEE] = 3.072 kg, CCC = 0.897; men: R 2 = 0.848, SEE = 2.217 kg CCC = 0.855; women: R 2 = 0.791, SEE = 2.194 kg, CCC = 0.840), LBM (total: R 2 = 0.736, SEE = 3.321 kg, CCC = 0.838; men: R 2 = 0.703, SEE = 2.450 kg, CCC = 0.774; women: R 2 = 0.854, SEE = 2.234 kg, CCC = 0.902), TFM (total: R 2 = 0.758, SEE = 1.932 kg, CCC = 0.844; men: R 2 = 0.650, SEE = 1.745 kg, CCC = 0.794; women: R 2 = 0.852, SEE = 1.504 kg, CCC = 0.890), and ALM (total: R 2 = 0.775, SEE = 1.726 kg, CCC = 0.876; men: R 2 = 0.805, SEE = 1.320 kg, CCC = 0.817; women: R 2 = 0.726, SEE = 1.198 kg, CCC = 0.802). When additional factors, such as creatinine, smoking, alcohol consumption, and physically inactive were included in the initial equations the predictive performance of the equations were generally improved. The prediction equations for body composition derived from this study suggest a potential application in epidemiological investigations on adult cancer survivors.