Optimum sample size allocation to minimize cost or maximize power for the two‐sample trimmed mean test

When planning a study, sample size determination is one of the most important tasks facing the researcher. The size will depend on the purpose of the study, the cost limitations, and the nature of the data. By specifying the standard deviation ratio and/or the sample size ratio, the present study considers the problem of heterogeneous variances and non‐normality for Yuen's two‐group test and develops sample size formulas to minimize the total cost or maximize the power of the test. For a given power, the sample size allocation ratio can be manipulated so that the proposed formulas can minimize the total cost, the total sample size, or the sum of total sample size and total cost. On the other hand, for a given total cost, the optimum sample size allocation ratio can maximize the statistical power of the test. After the sample size is determined, the present simulation applies Yuen's test to the sample generated, and then the procedure is validated in terms of Type I errors and power. Simulation results show that the proposed formulas can control Type I errors and achieve the desired power under the various conditions specified. Finally, the implications for determining sample sizes in experimental studies and future research are discussed.     

Fixed‐links models for investigating experimental effects combined with processing strategies in repeated measures designs: A cognitive task as example

Fixed‐links models enable the investigation of experimental effects in combination with processing strategies in the framework of repeated measures designs. The consideration of different processing strategies is reasonable whenever the task requirements can be met in different ways. Effects due to specific processing strategies become obvious in comparing the models representing these processing strategies and also by investigating the variances of the corresponding latent variables. The usefulness of such models is demonstrated by the reaction time data of a cognitive task including three treatment levels. Two different processing strategies were considered: analytic and holistic. The results indicated that the participants applied both processing strategies. The analytic strategy was primarily applied for performing with respect to the second and third treatment levels and the holistic strategy with respect to the first treatment level.     

Improved Fisher z estimators for univariate random‐effects meta‐analysis of correlations

Several authors have studied or used the following estimation strategy for meta‐analysing correlations: obtain a point estimate or confidence interval for the mean Fisher z correlation, and transform this estimate to the Pearson r metric. Using the relationship between Fisher z and Pearson r random variables, I demonstrate the potential discrepancy induced by directly z‐to‐r transforming a mean correlation parameter. Point and interval estimators based on an alternative integral z‐to‐r transformation are proposed. Analytic expressions for the expectation and variance of certain meta‐analytic point estimators are also provided, as are selected moments of correlation parameters; numerical examples are included. In an application of these analytic results, the proposed point estimator outperformed its usual direct z‐to‐r counterpart and compared favourably with an estimator based on Pearson r correlations. Practical implications, extensions of the proposed estimators, and uses for the analytic results are discussed.     

Cross validation issues in multiobjective clustering

The implementation of multiobjective programming methods in combinatorial data analysis is an emergent area of study with a variety of pragmatic applications in the behavioural sciences. Most notably, multiobjective programming provides a tool for analysts to model trade offs among competing criteria in clustering, seriation, and unidimensional scaling tasks. Although multiobjective programming has considerable promise, the technique can produce numerically appealing results that lack empirical validity. With this issue in mind, the purpose of this paper is to briefly review viable areas of application for multiobjective programming and, more importantly, to outline the importance of cross‐validation when using this method in cluster analysis.     

Multilevel modelling: Beyond the basic applications

Over the last 30 years statistical algorithms have been developed to analyse datasets that have a hierarchical/multilevel structure. Particularly within developmental and educational psychology these techniques have become common where the sample has an obvious hierarchical structure, like pupils nested within a classroom. We describe two areas beyond the basic applications of multilevel modelling that are important to psychology: modelling the covariance structure in longitudinal designs and using generalized linear multilevel modelling as an alternative to methods from signal detection theory (SDT). Detailed code for all analyses is described using packages for the freeware R.     

Childhood maltreatment and adult personality disorder symptoms in a non‐clinical sample

It can be argued that the well‐substantiated relationship between childhood maltreatment and adult personality disorder (PD) symptoms may be confounded by comorbid symptoms of depression, anxiety or dysfunctional childhood family environments. Therefore, the current study was designed to test the hypothesis that retrospective reports of childhood maltreatment would still be significantly related to reports of more PD symptoms when statistically controlling for these factors. One hundred and seventy‐eight non‐clinical participants were divided into groups reporting childhood maltreatment (n = 54) or not (n = 124) according to scores on the Childhood Trauma Questionnaire. Participants also completed questionnaires measuring current depression, anxiety, and PD symptoms as well as retrospective reports of their childhood environment. Results showed that individuals reporting childhood maltreatment reported more symptoms of PD than those not reporting childhood maltreatment, even when statistically controlling for depression, anxiety and retrospective reports of dysfunctional family environment. These findings underscore the relevance and independent contribution of childhood maltreatment to the development of PDs, with important implications for further research and clinical practice.     

Measuring organisational climate and employee engagement: Evidence for a 7 Ps model of work practices and outcomes

This study presents evidence supporting the psychometric properties of the Voice Climate Survey: an employee opinion survey that measures work practices and outcomes. The tool is tested across 13,729 employees from 1,279 business units representing approximately 1,000 organisations. Exploratory factor analyses, confirmatory factor analyses and internal reliability analyses support 31 lower‐order work practices and outcomes that aggregate into seven higher‐order work systems broadly covering practices and outcomes such as organisational direction, ethics, resources, involvement, recognition, development, teamwork, wellness, work/life balance, change management, customer satisfaction, job satisfaction, organisational commitment and employees' intention to stay. External validation of the tool is demonstrated by linking scores from the employee survey with independent manager reports of turnover, absenteeism, productivity, health and safety, goal attainment, financial performance, change management, innovation and customer satisfaction.     

Preparing for Year 12 examinations: Predictors of psychological distress and sleep

This study examined the influence of protective factors on psychological distress and sleep among Year 12 students 1 month prior to their externally assessed examinations. The sample included 195 Year 12 students who were to sit for at least one examination. Students completed the Adolescent Resilience Questionnaire, a measure of subjective stress, the Depression Anxiety and Stress Scales – 21, and measures of sleep. A sizeable minority of students reported high levels of subjective stress, severe symptoms of depression, anxiety and stress, and very inadequate sleep. Freedom from negative cognitions was the factor most strongly associated with protection from psychological distress. None of the factors protected against inadequate sleep. Results showed few gender differences. The findings validate popular concern about the wellbeing of Year 12 students. They also confirm that protective factors do not provide equal protection for all individuals against all adverse outcomes.     

USE OF CHAINING TO INCREASE COMPLEXITY OF ECHOICS IN CHILDREN WITH AUTISM

Echoics are a critical target of language intervention for children with autism, because a well‐established echoic repertoire on the part of the child allows the clinician to use vocal modeling as a flexible, low‐effort prompting procedure during teaching. In this study, we implemented a chaining procedure to increase the complexity of echoics in 2 children with autism and 1 child with developmental delay. The procedure was effective for all 3 participants, and gains were maintained after treatment was withdrawn in most cases.     

Dyslexia and dyscalculia: Two learning disorders with different cognitive profiles

This study tests the hypothesis that dyslexia and dyscalculia are associated with two largely independent cognitive deficits, namely a phonological deficit in the case of dyslexia and a deficit in the number module in the case of dyscalculia. In four groups of 8- to 10-year-olds (42 control, 21 dyslexic, 20 dyscalculic, and 26 dyslexic/dyscalculic), phonological awareness, phonological and visual-spatial short-term and working memory, naming speed, and basic number processing skills were assessed. A phonological deficit was found for both dyslexic groups, irrespective of additional arithmetic deficits, but not for the dyscalculia-only group. In contrast, deficits in processing of symbolic and nonsymbolic magnitudes were observed in both groups of dyscalculic children, irrespective of additional reading difficulties, but not in the dyslexia-only group. Cognitive deficits in the comorbid dyslexia/dyscalculia group were additive; that is, they resulted from the combination of two learning disorders. These findings suggest that dyslexia and dyscalculia have separable cognitive profiles, namely a phonological deficit in the case of dyslexia and a deficient number module in the case of dyscalculia.