Item response theory as a feature selection and interpretation tool in the context of machine learning.

Optimizing the number and utility of features to use in a classification analysis has been the subject of many research studies. Most current models use end-classifications as part of the feature reduction process, leading to circularity in the methodology. The approach demonstrated in the present research uses item response theory (IRT) to select features independent of the end-classification results without the biased accuracies that this circularity engenders.

He said, she said: work, biopsychosocial, and lifestyle contributions to coronary heart disease risk.

Objective: To test a model incorporating job characteristics, biopsychosocial, lifestyle, and nonmodifiable factors as they relate to coronary heart disease (CHD). Specifically, job characteristics and nonwork social ties (NWST) were examined as predictors of biopsychosocial health (BPSH), which was, in turn, expected to predict CHD directly and indirectly through influencing lifestyle. We also examined how age and family history of premature heart disease predicted objectively measured CHD risk.

Social workers and satisfaction with child welfare work: aspects of work, profession, and personal life that contribute to turnover.

Social workers practicing in government-mandated child welfare programs experience several unique challenges and workplace stressors that can contribute to social worker workplace dissatisfaction and higher rates of turnover. Most research on workplace wellbeing primarily focuses on workplace characteristics rather than on other variables, such as personal and professional life factors. From a sample of child welfare workers (n = 145), and following a model of subjective well-being, our findings show that three factors--work, profession, and personal life--significantly predict overall social worker satisfaction and intention to leave, confirming previous research on the multiple aspects of a social worker's life that contributes to his or her subjective well-being.

Determinants of adverse events in hospitals--the potential role of patient safety culture.

This study investigated two contributing factors in predicting adverse events in hospital settings. We approached this issue using the statistical procedure of hierarchical linear modeling to test for multilevel relationships. We found that the resource intensity of the presenting case was related to the severity level of negative incidents in hospital settings in a large metropolitan center.

Four multi-item interrater agreement options: comparisons and outcomes.

This study compared four multi-item indices of interrater agreement: (a) intraclass correlation coefficient, (b) within-group interrater agreement, (c) modified within-group interrater agreement, and (d) average deviation index. Findings included (a) that the different indices of interrater agreement provided different information about the agreement across raters, (b) standards for acceptable agreement are inconsistent for the several indices, (c) the average deviation index showed most groups had acceptable agreement, and (d) removing data from analyses based on unacceptable interrater agreement values does little in the way of affecting overall outcomes. Implications include that interrater agreement indices are not interchangeable and that for research purposes the concern about interrater agreement may be overstated.

Patient satisfaction: evaluating the success of hospital ward redesign.

Patients who were moved from a traditional medical ward to a new state-of-the-art medical ward were surveyed regarding their perceptions of quality during their hospitalization. Respondents rated the environment of the state-of-the-art facility, as well as the overall quality of their hospital stay, more positively. However, fewer differences in perceptions of the quality of the broader hospital environment and little difference in the perceived quality of staff-patient interactions were found.