PROPERLY RESCALED COMPONENTS OF SMOOTH TESTS OF FIT ARE DIAGNOSTIC

The paper illustrates a typical pitfall associated with a conventional interpretation of components of smooth tests of fit, for example the first nonzero components of the smooth tests for Poissonity, exponentiality, normality and the geometric distribution. In order to achieve a directed diagnosis concerning the kind of departure from a hypothesised model, an appropriate rescaling of components is necessary.     

A comparison of subgroup identification methods in clinical drug development: Simulation study and regulatory considerations

With advancement of technologies such as genomic sequencing, predictive biomarkers have become a useful tool for the development of personalized medicine. Predictive biomarkers can be used to select subsets of patients, which are most likely to benefit from a treatment. A number of approaches for subgroup identification were proposed over the last years. Although overviews of subgroup identification methods are available, systematic comparisons of their performance in simulation studies are rare. Interaction trees (IT), model‐based recursive partitioning, subgroup identification based on differential effect, simultaneous threshold interaction modeling algorithm (STIMA), and adaptive refinement by directed peeling were proposed for subgroup identification. We compared these methods in a simulation study using a structured approach. In order to identify a target population for subsequent trials, a selection of the identified subgroups is needed. Therefore, we propose a subgroup criterion leading to a target subgroup consisting of the identified subgroups with an estimated treatment difference no less than a pre‐specified threshold. In our simulation study, we evaluated these methods by considering measures for binary classification, like sensitivity and specificity. In settings with large effects or huge sample sizes, most methods perform well. For more realistic settings in drug development involving data from a single trial only, however, none of the methods seems suitable for selecting a target population. Using the subgroup criterion as alternative to the proposed pruning procedures, STIMA and IT can improve their performance in some settings. The methods and the subgroup criterion are illustrated by an application in amyotrophic lateral sclerosis.     

Illegitimate tasks as a source of work stress

Illegitimate tasks represent a task-level stressor derived from role and justice theories within the framework of “Stress-as–Offense-to-Self” (SOS; Semmer, Jacobshagen, Meier, & Elfering, 2007). Tasks are illegitimate if they violate norms about what an employee can properly be expected to do, because they are perceived as unnecessary or unreasonable; they imply a threat to one's professional identity. We report three studies testing associations between illegitimate tasks and well-being/strain. In two cross-sectional studies, illegitimate tasks predicted low self-esteem, feelings of resentment towards one's organization and burnout, controlling for role conflict, distributive injustice and social stressors in Study 1, and for distributive and procedural/interactional justice in Study 2. In Study 3, illegitimate tasks predicted two strain variables (feelings of resentment towards one's organization and irritability) over a period of two months, controlling for initial values of strain. Results confirm the unique contribution of illegitimate tasks to well-being and strain, beyond the effects of other predictors. Moreover, Study 3 demonstrated that illegitimate tasks predicted strain, rather than being predicted by it. We therefore conclude that illegitimate tasks represent an aspect of job design that deserves more attention, both in research and in decisions about task assignments.     

Investigation about a screening step in model selection

In many studies a large number of variables is measured and the identification of relevant variables influencing an outcome is an important task. For variable selection several procedures are available. However, focusing on one model only neglects that there usually exist other equally appropriate models. Bayesian or frequentist model averaging approaches have been proposed to improve the development of a predictor. With a larger number of variables (say more than ten variables) the resulting class of models can be very large. For Bayesian model averaging Occam’s window is a popular approach to reduce the model space. As this approach may not eliminate any variables, a variable screening step was proposed for a frequentist model averaging procedure. Based on the results of selected models in bootstrap samples, variables are eliminated before deriving a model averaging predictor. As a simple alternative screening procedure backward elimination can be used. Through two examples and by means of simulation we investigate some properties of the screening step. In the simulation study we consider situations with fifteen and 25 variables, respectively, of which seven have an influence on the outcome. With the screening step most of the uninfluential variables will be eliminated, but also some variables with a weak effect. Variable screening leads to more applicable models without eliminating models, which are more strongly supported by the data. Furthermore, we give recommendations for important parameters of the screening step.     

Permutation tests — A revival?!

It is shown that permutation tests have optimum properties for interesting classes of continuous distributions as well as for discrete ones. General conditions sufficient for uniformly maximal power on subclasses are given. Moreover, a variety of examples is presented. Research supported by Deutsche Forschungsgemeinschaft (DFG) grant No. Schm 677/5-1     

On the bias of factor score determinacy coefficients based on different estimation methods of the exploratory factor model

The factor score determinacy coefficient represents the common variance of the factor score predictor with the corresponding factor. The aim of the present simulation study was to compare the bias of determinacy coefficients based on different estimation methods of the exploratory factor model. Overall, determinacy coefficients computed from parameters based on maximum likelihood estimation, unweighted least squares estimation, and principal axis factoring were more precise than determinacy coefficients based on generalized least squares estimation and alpha factoring.     

Work Related and ``Private' Social Interactions at Work

Fifty-four young professionals in their first job after apprenticeship described their task-related and private interactions at work during five days, using a variant of the Rochester Interaction Records self-observation method (Reis and Wheeler, 1991). Results showed that more task-related interactions were reported than private interactions at work. The latter are described as more personally meaningful and more often initiated by the person or mutually initiated than task-related interactions. The number and duration of task-related interactions depended on the profession and the working conditions. The individual's own activity level during task-related interactions was predicted by social skills, measured 6 months before the diary study. Frequency of private interactions at work depended on familiarity level with colleagues. After controlling for role ambiguity and social stressors, more interactions and higher satisfaction with interactions at work predicted affective commitment, and more interactions at work also predicted job satisfaction. The study highlights the importance of conducting micro-analyses of social interactions at work.