不同铆测验设计下多维IRT等值方法的比较

实际应用中测验往往具有多维结构, 如果仍采用单维IRT方法进行等值, 会得到不准确的结果。因此对于多维结构的测验, 需要使用多维IRT等值方法来实现参数的转换。基于共同题设计, 文章通过模拟研究的方法, 考察了不同铆测验设计下几种多维IRT等值方法的表现, 同时考虑了测验长度、两个维度题目数量的比例、铆测验长度、铆测验的选择策略、两个维度之间的相关和等值群体的能力水平差异六个因素的影响。所比较的多维IRT等值方法有：均值/均值(MM)方法, 均值/标准差(MS)方法, Stoking-Lord (SL)方法, Haebara (HB)方法, 最小平方(LS)方法。结果显示：(1) SL, HB和LS方法得到的等值误差均方根最小, 且在各条件下表现较为稳定。(2) MM和MS方法在非等组条件下呈现出很大的误差均方根。(3)铆测验设计对SL, HB和LS方法的等值结果没有显著影响。(4)在两个维度之间的相关较高, 测验长度和铆测验长度较长, 等值群体的能力水平没有差异的条件下, SL, HB和LS方法得到的等值误差均方根最小。     

Bicriteria cost versus service analysis of a distribution network—a case study

In order to improve the distribution system for the Nordic countries the BASF AG considered 13 alternative scenarios to the existing system. These involved the construction of warehouses at various locations. For every scenario the transportation, storage, and handling cost incurred was to be as low as possible, where restrictions on the delivery time were given. The scenarios were evaluated according to (minimal) total cost and weighted average delivery time. The results led to a restriction to only three cases involving only one new warehouse each. For these, a more accurate model for the cost was developed and evaluated, yielding results similar to a simple linear model. Since there were no clear preferences between cost and delivery time, the final decision was chosen to represent a compromise between the two criteria. Copyright © 1999 John Wiley & Sons, Ltd.     

消费行为研究中的联合分析法

联合分析(Conjoint Analysis)是一种定量的研究消费者选择偏好的方法。虽然从它诞生到现在只有40年的历史,但它对消费者行为学的影响很大,它的发展更是日新月异。文章首先介绍了联合分析的基本概念,然后重点阐述了联合分析的原理、步骤和发展历程,最后对联合分析的发展方向作了一个简要概括。     

Analysing multisource feedback with multilevel structural equation models: Pitfalls and recommendations from a simulation study

When multisource feedback instruments, for example, 360-degree feedback tools, are validated, multilevel structural equation models are the method of choice to quantify the amount of reliability as well as convergent and discriminant validity. A non-standard multilevel structural equation model that incorporates self-ratings (level-2 variables) and others’ ratings from different additional perspectives (level-1 variables), for example, peers and subordinates, has recently been presented. In a Monte Carlo simulation study, we determine the minimal required sample sizes for this model. Model parameters are accurately estimated even with the smallest simulated sample size of 100 self-ratings and two ratings of peers and of subordinates. The precise estimation of standard errors necessitates sample sizes of 400 self-ratings or at least four ratings of peers and subordinates. However, if sample sizes are smaller, mainly standard errors concerning common method factors are biased. Interestingly, there are trade-off effects between the sample sizes of self-ratings and others’ ratings in their effect on estimation bias. The degree of convergent and discriminant validity has no effect on the accuracy of model estimates. The χ² test statistic does not follow the expected distribution. Therefore, we suggest using a corrected level-specific standardized root mean square residual to analyse model fit and conclude with further recommendations for applied organizational research.     

Parent–child attachment as a mediator of the relationship between parenting style and gelotophobia among adolescents

Fear of being laughed at and family interaction are highly related. Parental over‐control and over‐protection influence children's excessive anxiety over being laughed at. Conversely, parental attachment is an important index of the parent–child relationship and is closely correlated to children's gelotophobia. However, is it the style of parenting or the outcome of parenting (i.e. attachment) that influences a child's gelotophobia? To answer this question, the present study analysed the relationships between gelotophobia, perceived parenting of children and parent–child attachment, as well as the mediating role of attachment between parenting and children's gelotophobia, using a sample of 373 high‐school students. The results show that being highly communicative and close attachment completely weakened the negative correlation between warm, caring parenting and the child's gelotophobia; moreover, being highly communicative and close attachment, together with over‐protective and over‐controlling parenting, influence children's gelotophobia. In sum, this study indicates that parent–child attachment has a direct and indirect influence on perceived parental care and protection and children's fear of being laughed at.     

Prototypes of Faith: Findings with the Faith Q‐Sort

Defining religion and finding ways to assess it in individual lives has long challenged psychologists of religion. At first, open‐ended questionnaires were used, but with the advent of modern statistical methods, a succession of religiosity scales was developed. But these usually brief scales were typically based on the preconceptions of their authors, who were overwhelmingly Protestant Christian and often conservative, much as were those who completed them. To provide a more adequate way of assessing “faith,” a term here encompassing both religious and nonreligious attitudes, a new assessment device was developed that incorporates the singular advantages of Q methodology. The Faith Q‐Sort consists of 101 statements that respondents sort on a nine‐category continuum, indicating the degree to which each statement describes himself or herself. Factor analysis based on correlations of the sorts rather than individual items yielded, for the initial group of participants, three major prototypes and five minor ones, accounting for 67 percent of the variance. Of the 42 participants, 31 proved to be exemplars of one or another of the eight prototypes. Subsequent explorations illustrate the wealth of possibilities the FQS offers, both as a research instrument and a counseling tool.     

Randomization procedures in single‐case intervention research contexts: (Some of) “the rest of the story”

Following up on articles recently published in this journal, the present contribution tells (some of) “the rest of the story” about the value of randomization in single‐case intervention research investigations. Invoking principles of internal, statistical‐conclusion, and external validity, we begin by emphasizing the critical distinction between design randomization and analysis randomization, along with the necessary correspondence between the two. Four different types of single‐case design‐and‐analysis randomization are then discussed. The persistent negative influence of serially dependent single‐case outcome observations is highlighted, accompanied by examples of inappropriate applications of parametric and nonparametric tests that have appeared in the literature. We conclude by presenting valid applications of single‐case randomization procedures in various single‐case intervention contexts, with specific reference to a freely available Excel‐based software package that can be accessed to incorporate the present randomization schemes into a wide variety of single‐case intervention designs and analyses.     

Reaching Wuthering Heights with Brave New Words: The Influence of Originality of Words on the Success of Outstanding Best‐Sellers

The judgment and ultimately the success of creative products should be determined by their properties. However, it has not been considered so far whether the same applies to books. Earlier research has found an inverted‐U relationship between originality of stimuli and their success. Linguistic originality as a text feature could influence the success of books as creative products in a natural experiment. The present historiometric study investigated whether originality predicts the popularity in a significant sample, the most best‐selling English books from 200 years. Originality was calculated based on word frequency using a commercial service. Popularity was assessed with a composite measure including indicators like number of sold copies. Regression analysis indicated originality had a direct linear effect on popularity, partially moderated by the time of first publishing. That is, while originality was generally beneficial for the success of a best‐seller, the same originality was more appreciated at a later point in time. This is the first quantitative study supporting the long‐held assumption that the evaluation of creative narratives is influenced by temporal context. The fact that some works like The Great Gatsby got more appreciation when they were rediscovered may reflect a general principle rather than mere exceptions.     

Simple discrimination in stingless bees (Melipona quadrifasciata): Probing for select‐ and reject‐stimulus control

Simple and conditional discrimination training may produce various types of controlling relations. Responses may be controlled primarily by the positive stimulus (select–control relation) or by the negative stimulus (reject–control relation; the subject excludes the negative stimulus and chooses the positive). Bees learn to respond in simple and conditional discriminations. However, no study has searched for reject–control responding in Melipona bees. We trained Melipona quadrifasciata on a simple discrimination task (S+ vs. S‐; e.g., blue vs. yellow) and then probed for stimulus control with two types of probe trials, S+ versus a new stimulus (Select–control probes) and S‐ versus a new stimulus (Reject–control probes). For Group Different, a new‐stimulus color (e.g., white) was used in one type of probe and another color (e.g., black) was used in the other type. For Group Same, a single new‐stimulus color was used in both types of probes. On Select probes, the bees always preferred S+ to the new stimulus. On Reject probes, results were mixed. Depending on the colors used in training and probing, bees responded to both stimuli, and even preferred the S‐. The data suggest no control by the negative function of the S‐ and support the select‐stimulus control hypothesis of responding.