一种基于科学计算器的方差分析计算方法

利用科学计算器直接计算出总标准差和总方差以及处理标准差和处理方差,然后用方差乘以自由度得到二者的平方和,再根据平方和公式计算出误差的平方和及方差。该方法在计算程序上比传统方法更简便,可用于统计计算和教学。     

Examining the Precision of Cut Scores Within a Generalizability Theory Framework: A Closer Look at the Item Effect

An Angoff standard setting study generally yields judgments on a number of items by a number of judges (who may or may not be nested in panels). Variability associated with judges (and possibly panels) contributes error to the resulting cut score. The variability associated with items plays a more complicated role. To the extent that the mean item judgments directly reflect empirical item difficulties, the variability in Angoff judgments over items would not add error to the cut score, but to the extent that the mean item judgments do not correspond to the empirical item difficulties, variability in mean judgments over items would add error to the cut score. In this article, we present two generalizability-theory–based analyses of the proportion of the item variance that contributes to error in the cut score. For one approach, variance components are estimated on the probability (or proportion-correct) scale of the Angoff judgments, and for the other, the judgments are transferred to the theta scale of an item response theory model before estimating the variance components. The two analyses yield somewhat different results but both indicate that it is not appropriate to simply ignore the item variance component in estimating the error variance.     

A SECOND LOOK AT ANALYSIS OF VARIANCE ON DICHOTOMOUS DATA

The work of Lunney (1970) concerning the appropriateness of analysis of variance (ANOVA) techniques on dichotomous data is discussed and extended. Relations between standard statistical techniques for analyzing dichotomous data and ANOVA procedures are indicated. The need for usefulness of analyzing transformed data as opposed to direct analysis of dichotomous data are discussed. Required statistical procedures employing transformed data are outlined.     

Forward Position Analysis of 5-DOF Parallel Mechanism

Forward position analysis is one of the importantand difficult issues in the kinematic analysis of paral-lel manipulators and has been extensively investiga-ted[1—3]. Most of the approaches for solving forwardposition problems of parallel manipulators with six de-grees of freedom (6-DOF), the Stewart platform forinstance, can be classified into two categories, i.e.closed-form analytical method using polynomial ap-proach and numerical method using iterative algo-rithm. Although the polynomial…     

Simulation study of item bias detection with restricted factor analysis

Restricted factor analysis (RFA) can be used to detect item bias (also called differential item functioning). In the RFA method of item bias detection, the common factor model serves as an item response model, but group membership is also included in the model. Two simulation studies are reported, both showing that the RFA method detects bias in 7‐point scale items very well, especially when the sample size is large, the mean trait difference between groups is small, the group sizes are equal, and the amount of bias is large. The first study further shows that the RFA method detects bias in dichotomous items at least as well as an established method based on the one‐parameter logistic item response model. The second study concerns various procedures to evaluate the significance of two‐item bias indices provided by the RFA method. The results indicate that the RFA method performs best when it is used in an iterative procedure.     

Understanding the degrees of freedom of sample variance by using Microsoft Excel

In this article, the degrees of freedom of the sample variance are simulated by using the Visual Basic for Applications of Microsoft Excel 2010. The simulation file dynamically displays why the sample variance should be calculated by dividing the sum of squared deviations by n‐1 rather than n, which is helpful for students to grasp the meaning of degrees of freedom.     

A Generalized DIF Effect Variance Estimator for Measuring Unsigned Differential Test Functioning in Mixed Format Tests

One approach to measuring unsigned differential test functioning is to estimate the variance of the differential item functioning (DIF) effect across the items of the test. This article proposes two estimators of the DIF effect variance for tests containing dichotomous and polytomous items. The proposed estimators are direct extensions of the noniterative estimators developed by Camilli and Penfield (1997) for tests composed of dichotomous items. A small simulation study is reported in which the statistical properties of the generalized variance estimators are assessed, and guidelines are proposed for interpreting values of DIF effect variance estimators.     

ADU LT COGNITIVE STYLES AND TEST PERFORMANCE

Ninety adult males ages 25 to 58 were individually tested with three cognitive style measures: the Estimation Scale (ES), a measure of category width; the Matching Familiar Figures (MFF) Test; and the Hidden Figures Test (HFT). Multiple correlations were performed between the three cognitive style measures and the standardized scores from the Chartered Life Underwriter (CLU) area achievement examinations. The results of this analysis indicated that only MFF error scores produce a significant correlation with CLU exam performance. Further analyses conducted on the MFF data indicated that there was a significant negative correlation between response latencies and errors, there was a significant positive correltation between response latency and age, and reflective subjects had significantly higher mean CLU scores than did impulsive subjects. It was concluded that error scores on the MFF are a better predictor of test performance than are latencies. This conclusion was consistent with other studies that have found error scores to be more strongly related to personality variables and performance on other cognitive style than latencies prove to be.     

Scale Considerations in Using Vague Quantifiers

The responses to many survey questions are often made in terms of vague quantifiers such as very often or occasionally. However, we know little about the scale characteristics of such quantifiying terms. This paper reviews the literature on the use of frequency expressions as questionnaire anchor points. Data are presented which suggest that the popularly used five-point response category set of always, often, occasionally, seldom, and never is not intervally scaled. Utilizing data obtained through ratio estimation, an approximately equal interval measurement scale is developed for a five-point response category set.     

Seismic response analysis of damper-connected adjacent structures with stochastic parameters

Dynamic response analysis of damper connected adjacent multi-story structures with uncertain parameters is carried out. A formula of the multi degree of freedom （MDOF） for the structure-damper system with stochastic parameters is derived. The uncertainties of mass and stiffness are taken into consideration firstly. The ground acceleration is represented by Kanai-Tajimi filtered non-stationary process. The mean square random responses of structural displacement and story drift are chosen as the optimization objective. The variations of mean square responses of top floor displacements and bottom story drifts in neighboring structures with the damper stiffness and damping coefficient are analyzed in detail. Through the parametric study, the acquiring optimum parameters of damper are regarded as numerical results. Then, a reducing order model of the MDOF system for adjacent structures with mean parameters is presented. The explicit expressions for determining optimal parameters of Kelvin model-defined damper which is used to connect adjacent single degree of freedom （SDOF） structures subjected to a white-noise excitation are employed to achieve the appropriate damper parameters, which are called theory results. Through a comparative study, it can be found that the theory values of damper parameters are consistent with the results based on extensive parametric studies. The analytical results can be obtained by using the first natural frequencies and the total mass of the adjacent deterministic structures with mean parameters. The analytical formulas can be used to find appropriate parameters of damper between adjacent structures for engineering applications. The performance of damper is investigated on the basis of mitigations of mean square random responses of inter-story drifts, displacements and accelerations in adjacent structures. The numerical results demonstrate the robustness of coupled building control strategies.     

Effects of stimulant medication on learning in children with ADHD

It is well established that stimulant medication improves classroom manageability and attention in terms of time on task, but does stimulant medication improve learning or long-term academic achievement in children with ADHD? There is no clear evidence that it does, but there are at least two reasons why beneficial effects may be obscured in research studies and clinical practice: (1) Higher-than-optimal doses may be prescribed if behavioral response (rather than cognitive response) is used to titrate the dose, and (2) treatment may be overinclusive if diagnostic groups are targeted in which a significant proportion of cases do not have favorable cognitive responses to medication. This article addresses these two issues and describes a large clinical series of patients who were evaluated using a double-blind medication assessment protocol designed to overcome some of the theoretical deficiencies suggested by these issues.     

Path-Selection Behavior of Young Children in Body-Mazes

A sample of 120 subjects, 11 to 14 years of age, was used to examine the effects of gender and community socioeconomic setting (CSES) on intransitivity of paired-comparison responses. A two-way analysis of variance was used to analyze data. Small gender differences were attributable to chance alone. CSES differences were large, however, with the lower CSES group having significantly (p = .014) more intrasitive responses than the higher CSES group. Interaction effects were inconsequential. It was not possible to isolate the specific aspect(s) of community socioeconomic setting (i.e., family income, parents’ education, etc.) that accounted for the CSES effect. It was concluded that CSES is too complex to treat as a simple dichotomous variable (higher or lower) and that additional research is needed to examine its multidimensional nature. Suggestions for future research are given.     

Assessing the Sensitivity of Weighted Least Squares Model Fit Indexes to Local Dependence in Item Response Theory Models

Given the relationships of item response theory (IRT) models to confirmatory factor analysis (CFA) models, IRT model misspecifications might be detectable through model fit indexes commonly used in categorical CFA. The purpose of this study is to investigate the sensitivity of weighted least squares with adjusted means and variance (WLSMV)-based root mean square error of approximation, comparative fit index, and Tucker–Lewis Index model fit indexes to IRT models that are misspecified due to local dependence (LD). It was found that WLSMV-based fit indexes have some functional relationships to parameter estimate bias in 2-parameter logistic models caused by violations of LD. Continued exploration into these functional relationships and development of LD-detection methods based on such relationships could hold much promise for providing IRT practitioners with global information on violations of local independence.     

二级计分数据DIF模拟研究的数据产生原理及其软件实现

和基于实测数据的DIF研究相比,基于模拟数据的DIF研究不仅可以自由操纵实验条件,而且可以给出检验力和I型错误指标。本文详细阐述了二级计分DIF模拟数据的产生原理,其产生过程包括四个阶段：选择DIF产生思路,选择项目反应理论模型,确定考生特征、题目特征和复本数,计算考生在题目上的正确作答概率并转化为二级计分数据。并且分别利用常用软件Excel和专业软件WinGen3展示了二级计分DIF模拟数据的产生过程。     

Probabilistic Cues to Grammatical Category in English Orthography and Their Influence During Reading

We investigated probabilistic cues to grammatical category (noun vs. verb) in English orthography. These cues are located in both the beginnings and endings of words—as identified in our large-scale corpus analysis. Experiment 1 tested participants' sensitivity to beginning and ending cues while making speeded grammatical classifications. Experiment 2 tested sensitivity to these cues during lexical decisions. For both tasks, words with consistent ending cues (with respect to grammatical category) were processed more quickly and with lower error rates than words with inconsistent ending cues. However, for beginnings, consistent cues resulted in lower errors but no differences in response times. The data reported here point to the multifaceted nature of grammatical category representation and indicate that probabilistic orthographic cues relating to grammatical category have a clear influence on lexical processing particularly when these cues are located at the end of the word.     

Conditioned cardiac and behavioral response topography to an olfactory CS dissociates with age

Conditioned cardiac and behavioral responses to an olfactory stimulus (CS) paired with electric shock (US) were studied in 16-, 23-, and 75-day-old rats. Heart rate and behavior were recorded during a single 15-trial training session in which a 10-sec olfactory stimulus (amyl acetate) was either paired or explicitly unpaired with the US. In the preweanling rats, the conditioned cardiac response consisted of anincrease in heart rate (tachycardia) and a slight increase in investigatory behaviors. In contrast, the conditioned cardiac response in the adult rats consisted of adecrease in heart rate (bradycardia) and a striking increase in behavioral immobility (freezing). The 23-day-old rats showed a dichotomous pattern of responding; some showed preweanling-like conditioned responses and some showed adultlike responses. Developmental changes in neural organization and degree of learning are discussed as possible determinants of the transition from infant to adult modes of responding.     

Using Total Sample Size Weights in Meta-Analysis of Log-Odds Ratios

The log-odds ratio (ln[OR]) is commonly used to quantify treatments' effects on dichotomous outcomes and then pooled across studies using inverse-variance (1/v) weights. Calculation of the ln[OR]'s variance requires four cell frequencies for two groups crossed with values for dichotomous outcomes. While primary studies report the total sample size (n..), many do not report all four frequencies. Using real data, we demonstrated pooling of ln[OR]s using n.. versus 1/v weights. In a simulation study we compared two weighting approaches under several conditions. Efficiency and Type I error rates for 1/v versus n.. weights used to pool ln[OR] estimates depended on sample size and the percent of studies missing cell frequencies. Results are discussed and guidelines for applied meta-analysts are provided.     

Environmental Education and Environmental Advocacy: Revisited

Abstract

Environmental education and environmental advocacy have a contentious relationship. In this article, the author argues that there will always be uncertainty about educationally appropriate responses to controversial issues. Although uncertainty is inherent in this task, the choices are not dichotomous. The author also argues that education suggests a fluidity of meaning that shifts across a range of contexts, and what needs to be done will be found on a case-by-case and context-by-context basis in a mediated and negotiated third space. Some tentative guide-posts are offered.     

A Comparison of Diagonal Weighted Least Squares Robust Estimation Techniques for Ordinal Data

This study compared diagonal weighted least squares robust estimation techniques available in 2 popular statistical programs: diagonal weighted least squares (DWLS; LISREL version 8.80) and weighted least squares–mean (WLSM) and weighted least squares—mean and variance adjusted (WLSMV; Mplus version 6.11). A 20-item confirmatory factor analysis was estimated using item-level ordered categorical data. Three different nonnormality conditions were applied to 2- to 7-category data with sample sizes of 200, 400, and 800. Convergence problems were seen with nonnormal data when DWLS was used with few categories. Both DWLS and WLSMV produced accurate parameter estimates; however, bias in standard errors of parameter estimates was extreme for select conditions when nonnormal data were present. The robust estimators generally reported acceptable model–data fit, unless few categories were used with nonnormal data at smaller sample sizes; WLSMV yielded better fit than WLSM for most indices.