Item bias detection using loglinear irt

A method is proposed for the detection of item bias with respect to observed or unobserved subgroups. The method uses quasi-loglinear models for the incomplete subgroup × test score × Item 1 × ... × itemk contingency table. If subgroup membership is unknown the models are Haberman's incomplete-latent-class models.The (conditional) Rasch model is formulated as a quasi-loglinear model. The parameters in this loglinear model, that correspond to the main effects of the item responses, are the conditional estimates of the parameters in the Rasch model. Item bias can then be tested by comparing the quasi-loglinear-Rasch model with models that contain parameters for the interaction of item responses and the subgroups.The author thanks Wim J. van der Linden and Gideon J. Mellenbergh for comments and suggestions and Frank Kok for empirical data.     

Expected performance at the human/computer interface as a function of user proficiency and system power

How do managers expect the proficiency of users and the power of the computers to determine overall performance? Five different models are proposed: (a) a matching model in which optimal performance is achieved when the power of the system is judged to be compatible with the proficiency of the user, (b) an averaging model in which expected performance is the average of the values of user proficiency and system power, (c) a multiplying model in which performance is the product of the values of user proficiency and system power, (d) a human/computer ratio model in which performance is determined by the ratio of system power over total effort, and (e) a computer/human ratio model in which performance is determined by the ratio of user proficiency over total effort. The applicability of these models was assessed by having managers and students of management predict performance in human/computer systems from information about the user's proficiency with computers and the power of the system. Participants rated 16 combinations of user proficiency and system power from a 4 × 4 factorial design. The pattern of ratings indicated that 51 per cent used a multiplying model and 25 percent used an averaging model; whereas, only 6 percent used the matching model and 4 percent used a ratio model. The remaining 14 percent did not follow any model clearly. Implications of these results were discussed for the design of the human/ computer interface, training and selection of users, and the cost-benefit tradeoffs for investment in user training versus equipment acquisition.     

Increasing item complexity: A possible cause of scale shrinkage for unidimensional item response theory

When the three-parameter logistic model is applied to tests covering a broad range of difficulty, there frequently is an increase in mean item discrimination and a decrease in variance of item difficulties and traits as the tests become more difficult. To examine the hypothesis that this unexpected scale shrinkage effect occurs because the items increase in complexity as they increase in difficulty, an approximate relationship is derived between the unidimensional model used in data analysis and a multidimensional model hypothesized to be generating the item responses. Scale shrinkage is successfully predicted for several sets of simulated data.The author is grateful to Robert Mislevy for kindly providing a copy of his computer program, RESOLVE.     

Factor analysis for non-normal variables

Factor analysis for nonnormally distributed variables is discussed in this paper. The main difference between our approach and more traditional approaches is that not only second order cross-products (like covariances) are utilized, but also higher order cross-products. It turns out that under some conditions the parameters (factor loadings) can be uniquely determined. Two estimation procedures will be discussed. One method gives Best Generalized Least Squares (BGLS) estimates, but is computationally very heavy, in particular for large data sets. The other method is a least squares method which is computationally less heavy. In one example the two methods will be compared by using the bootstrap method. In another example real life data are analyzed.This paper has partly been written while the author was a visiting scholar at the Department of Psychology, University of California, Los Angeles. He wants to thank Peter Bentler who made this stay at UCLA possible and for his valuable contributions to this paper. This research was supported by the Netherlands Organization for the Advancement of Pure Research (Z.W.O) under number R56-150 and by USPHS Grant DA01070.     

Type development in childhood and beyond

This article defines and explains type development across the lifespan and summarizes what other researchers have presented as a reference. A timeline for development is identified, and ways to differentiate development of a function from the development of other personality characteristics are described. Type development is about the dispersion of energy from an unconscious level to a managed conscious level. This development occurs in a spiral pattern, so development of each function occurs repeatedly but at higher levels of use and understanding each time. Although the dominant function develops most because of its innate drive to be the lead, each function can develop throughout the lifespan. The relative energy associated with each function always exists. When the energy is in the unconscious, it drives behavioural responses, but these are not within the individual’s control. The ideal is to develop the conscious awareness of each of the functions so each can be consciously used as needed to match a situation. Because development of the functions is essential, this article examines potential interferences with development and then offers some suggestions for ways to promote developmental opportunities.     

Motivation for orienteering An exploratory analysis using confirmatory factor analytic techniques

The aim of the present study was to investigate and identify different dimensions of motivation for the sport of orienteering, as well as to compare the strength of these dimensions by sex and age. Data for 512 subjects was analyzed with confirmatory factor analytic techniques in an extensive process where model generating was the primary matter of interest. There appear to be three broad main dimensions of motivation for orienteering. There is a main distinction between a social-recreational dimension and one reflecting a desire to achieve. Within the social-recreational dimension it is possible to identify more specific factors. In addition, there seems to be a relatively broad and distinct dimension, which is partly related to the other two, reflecting a search for internal stimulation. Comparisons of means of latent variables by sex and age indicate a number of clear and interesting differences.     

智力开发的基本理念与实践

以高度分化的g因素研究为理论基础,智力开发活动形成了神经潜能开发、心理管理与反省经验开发、专家技能开发、多元智力开发、社会分布式认知开发、知识表征重组开发、环境重组开发等智力要素开发理念以及一些代表性的实践。随着对智力结构的整合研究的出现和不断深入,智力开发理念和模式将由分化的要素开发向整合开发转变,以智力与其他心理结构、智力与内部世界和外部世界系统关系为基础的智力开发将是理论和实践研究的重点内容     

文本阅读研究的技术模型和新观点

文本阅读心理研究一直是心理学界广泛关注的课题。从20世纪30年代第一个严格的文本阅读实验至今,这一领域的研究方法和技术模型不断进步,具体理论和实际应用不断成熟。到目前为止,文本阅读领域已经形成了庞大的体系。作者首先对文本阅读研究中常用的技术模型进行了梳理,从目标、空间、时间、因果和主角五个方面分别对其进行了介绍。在此基础上,文章评述了文本阅读研究的两种新观点,即文本阅读的双加工理论和知觉符号理论。最后作者对文本阅读研究的未来趋势进行了展望