After 4 decades of severe criticism, the ritual of null hypothesis significance testing—mechanical dichotomous decisions around a sacred.05 criterion—still persists. This article reviews the problems with this practice, including its near-universal misinterpretation ofp as the probability that Ho is false, the misinterpretation that its complement is the probability of successful replication, and the mistaken assumption that if one rejects Ho one thereby affirms the theory that led to the test. Exploratory data analysis and the use of graphic methods, a steady improvement in and a movement toward standardization in measurement, an emphasis on estimating effect sizes using confidence intervals, and the informed use of available statistical methods is suggested. For generalization, psychologists must finally rely, as has been done in all the older sciences,

Abstract not found

A literature review was conducted to understand the limitations of well-known statistical analysis techniques, particularly analysis of variance. The review is structured around six major points: (1) averaging across participants can be misleading; (2) strong predictions are preferable to weak predictions; (3) constructs and measures should be distinguished conceptually and empirically; (4) statistical signi ® cance and practical signi ® cance should be distinguished conceptually and empirically; (5) the null hypothesis is virtually never true; and (6) one experiment is always inconclusive. Based on these insights, a number of lesser-known and less-frequently used statistical analysis techniques were identi ® ed to address the limitations of more traditional techniques. In addition, a number of methodological conclusions about the conduct of human factors research are presented. 1.

I examine the role of explaining variance in the construction of explanatory theory. Explaining variance can be an insufficient basis for evaluating a theory (Lieberson, 1985). Starting with this insight, I suggest that models that provide explanations of variance do not necessarily provide good explanations of causal mechanisms. I then explore the utility of process models and theories (Mohr, 1982) relative to variance theories. I clarify the role of stochastic processes in such model building and discuss the implications of such processes for evaluating explanatory `adequacy'. Under some conditions, explaining variance may be neither a necessary nor a sufficient condition for good explanatory theory. I then identify some implications of this argument for developing and analyzing explanatory theory. These arguments are applied to two examples: (1) meta-analysis and (2) the disposition versus situation debate (a variant on the nature vs. nurture argument) to illustrate the implications of ...

Employee portals are systems that provide employees with the timely and relevant information that they need to perform their duties and to make efficient business decisions. Although their use is widespread, the question on how benefits of these portals are materialized for their users has not been fully answered yet. Thus, the purpose of this paper is to gain a better understanding of the utilitarian value of employee portals for individual users. Therefore, we develop a second-order hierarchical conceptual model whose core structure is founded on the theoretical behavioral science concepts embedded in the diffusion of innovations theory, theory of planned behavior, and the research stream of engineering psychology. We empirically test the model by means of component-based structural equation modeling. For this, we collected 5,783 employees ' responses in a survey of 19 companies. Our results indicate that amongst the theorized factors, the quality of support provided to users is the most important factor that affects employee portal related performance gains. Furthermore, collaborative functionalities of an employee portal acts as a critical mediator that channels benefits arising as a result of efficient support and ergonomic employee portal design towards increasing goal oriented breadth of employee portal usage. Finally, we find that with increasing knowledge-intensity of employee tasks, ergonomicity of an employee portal and breadth of use has a stronger effect on performance gains.

Social virtual worlds (SVWs) have become increasingly important environments for social interaction, especially for the younger generations. For SVWs to be economically sustainable, attracting new users and retaining the existing ones existing users is a paramount issue. This calls for understanding of the reasons why people engage in social virtual worlds. This study investigates the motives for continuously engagement in SVWs and develops a research model grounded on the decomposed theory of planned behavior. The model is empirically tested with a data collected from Canadian active Habbo goers using PLS. Surprisingly, perceived behavioral control and subjective norm were found more important determinants of continuous use intention than attitude. The results indicated that hedonic motives were the main determinant of attitude. However, altogether only 21.9 % of attitude was explained by utilitarian, hedonic and social outcomes. As a result, the study revealed that rather relying on generic items in measuring attitude and the beliefs regarding the utilitarian and social outcomes, the characteristics of SVW context should be reflected in the operationalisations of the constructs.

Many public and private genome-wide association studies that we have analyzed include flaws in design, with avoidable confounding appearing as a norm rather than the exception. Rather than recognizing flawed research design and addressing that, a category of quality-control statistical methods has arisen to treat only the symptoms. Reflecting more deeply, we examine elements of current genomic research in light of the traditional scientific method and find that hypotheses are often detached from data collection, experimental design, and causal theories. Association studies independent of causal theories, along with multiple testing errors, too often drive health care and public policy decisions. In an era of large-scale biological research, we ask questions about the role of statistical analyses in advancing coherent theories of diseases and their mechanisms. We advocate for reinterpretation of the scientific method in the context of large-scale data analysis opportunities and for renewed appreciation of falsifiable hypotheses, so that we can learn more from our best mistakes.