一种基于Rasch模型的Angoff方法及其应用

水平厘定方法在近二十年来发展迅速,其中Angoff方法的应用最为广泛。该文在已有的方法上提出了一种基于Rasch模型的改进的Angoff方法,该方法使用Rasch模型来估计典型学生的位置,通过均方残差等方法剔除不一致的裁判。文章最后详细阐述了该方法在香港全港性系统评估中的应用。     

‘The essay method’ a qualitative method for studying therapeutic dialogues

ABSTRACT

This paper presents a qualitative research method for analysing therapeutic dialogues called ‘the essay method’. A central part of the method uses the format of the literary essay as a model. The method consists of a close monitoring of clinical material guided by an overall psychoanalytic/psychodynamic theoretical frame. It combines both clinical details and global patterns of clinical material and is especially fitted for studying relational qualities of psychotherapeutic dialogues. In this paper, the background for qualitative analysis in studying psychoanalytic material is discussed, and the procedures for using the method are demonstrated. A study of therapeutic competence in a group of student therapists is used as an example of the method in practice, demonstrating that the essay method may have potential for revising theory and establishing new concepts based on empirical findings. The essay method is compared with other qualitative methods, and it is argued that it is suitable for psychoanalytic research.     

用Delta法估计多维测验合成信度的置信区间

大量研究表明,一般情况下用合成信度可以较好地估计测验信度。对于合成信度及其置信区间的估计方法,在单维测验的情形已有不少研究。但罕有研究讨论多维测验合成信度的区间估计方法。本文用Delta法推导出计算多维测验合成信度的标准误公式,进而计算置信区间,并用一个例子说明如何编程估计多维测验合成信度及其置信区间。     

Multidimensional scaling of individual differences for an individual person with a hand-held calculator

An extension to the INDSCAL method of individual differences scaling was developed. The method derives measures of deviation from a “standard” observer, who is defined in terms of the INDSCAL subject weights. These weights are calculated for one individual person at a time. The method is implemented in a small hand-held calculator. The calculator is used for data acquisition, and it interacts with the user in a conversational manner through an alpha-numeric display. The present implementation of the method scales 84 judgments of pair similarity among seven objects and derives subject weights on two dimensions. The method can be extended to sets of up to 16 objects and up to six dimensions. The method is a feasible, cost-effective approach to classification of individuals on complex perceptual attributes within a single 1-h session.     

A non-graphical method for transforming an arbitrary factor matrix into a simple structure factor matrix

The most commonly used method of factoring a matrix of intercorrelations is the centroid method developed by L. L. Thurstone. It is, however, necessary to transform the centroid matrix of factor loadings into a simple structure matrix in order to facilitate the interpretation of the factor loadings. Current methods for effecting this transformation are chiefly graphical and require considerable experience and personal judgment. This paper presents a new method for transforming an arbitrary factor matrix into a simple structure matrix by methods almost completely objective. The theory underlying the method is developed and approximation procedures are derived. The method is applied to a matrix of factor loadings previously analyzed by Thurstone.     

Valued Relations Aggregation with the Borda Method

The Borda method is a well-known method used to aggregate a profile of crisp binary relations into one weak order. We present a generalization of the Borda method, allowing it to aggregate a profile of valued relations. We give a characterization of the Borda method in this context and we show some links with the PROMETHEE method.     

Estimation of the correlation coefficient in contingency tables with possibly nonmetrical characters

A method of estimating the product moment correlation from the polychoric series is developed. This method is shown to be a generalization of the method which uses the tetrachoric series to obtain the tetrachoric correlation. Although this new method involves more computational labor, it is shown to be superior to older methods for data grouped into a small number of classes.     

Some multiple correlation and predictor selection methods

The Doolittle, Wherry-Doolittle, and Summerfield-Lubin methods of multiple correlation are compared theoretically as well as by an application in which a set of predictors is selected. Wherry's method and the Summerfield-Lubin method are shown to be equivalent; the relationship of these methods to the Doolittle method is indicated. The Summerfield-Lubin method, because of its compactness and ease of computation, and because of the meaningfulness of the interim computational values, is recommended as a convenient least squares method of multiple correlation and predictor selection.     

A method of score conversion through item statistics

A method is presented for converting the scores on one form of a test to those on another form of the same test. The method is particularly applicable to the case where each form has been administered to a different group and the only link between the two forms is a subset of items common to both. The proposed method, called theitem method of conversion, has been applied to several tests for which other methods of conversion are available for comparison. The necessary data are limited to tests for which the total score is the criterion for item analyses. The method gives highly satisfactory results for all the tests to which it has been applied, particularly when the two groups are rather different, in which case the delta method (a different item method) is inappropriate.The authors are only two of a group, including W. H. Angoff, F. M. Lord, and M. K. Schultz, all of whom have made important contributions to this paper.     

A comparison of the thurstone method of multiple factors with the hotelling method of principal components

The factors found by the Hotelling method of principal components present the same necessity for rotation as those found by the Thurstone method of multiple factors.The Thurstone method of multiple factors is here found superior to the Hotelling method of principal components, and is suggested as the method of choice for all problems requiring the determination of elemental components from a group of related variables, since it was found on every point of comparison that, when applied to a correlation matrix of unknown communalities, the Thurstone method was more accurate than the Hotelling method, and also required much less time.     

Dual imputation model for incomplete longitudinal data

Missing values are a practical issue in the analysis of longitudinal data. Multiple imputation (MI) is a well‐known likelihood‐based method that has optimal properties in terms of efficiency and consistency if the imputation model is correctly specified. Doubly robust (DR) weighing‐based methods protect against misspecification bias if one of the models, but not necessarily both, for the data or the mechanism leading to missing data is correct. We propose a new imputation method that captures the simplicity of MI and protection from the DR method. This method integrates MI and DR to protect against misspecification of the imputation model under a missing at random assumption. Our method avoids analytical complications of missing data particularly in multivariate settings, and is easy to implement in standard statistical packages. Moreover, the proposed method works very well with an intermittent pattern of missingness when other DR methods can not be used. Simulation experiments show that the proposed approach achieves improved performance when one of the models is correct. The method is applied to data from the fireworks disaster study, a randomized clinical trial comparing therapies in disaster‐exposed children. We conclude that the new method increases the robustness of imputations.     

How to avoid the fallacies of cognitive subtraction in brain imaging

Functional neuroanatomy of cognitive processes is generally derived by subtraction of a control task from an experimental task. We show that this method is logically untenable and propose to replace it with an adaptation of the specific-effect method and of the additive-factor method. The basic flaw that undermines the subtractive method is the comparison between different tasks. We argue that the specific-effect method should be used when investigating activations produced by different levels of a qualitative variable, whereas the additive-factor method should be used for quantitative variables. The use of these methods allows one to distinguish between parallel and serial stages of processing and between local and distributed processing.     

Addressing the Person of the Therapist in Supervision: The Therapist's Inner Conversation Method

In this study a method of retrospective case supervision is presented aimed at helping the supervisee to become a better self‐supervisor. The method pays special attention to the therapist's self‐reflection and has the therapist's inner conversation as a central concept. The starting point of the method is an assignment in which the supervisee reflects on a case using a tape‐assisted recall procedure. The method helps trainees to develop experiential expertise to become more flexible and effective therapists. A case example of one training group of novice family therapists illustrates the use of the method.     

THE TIME-ORDER ERROR IN THE CONSTRUCTION OF A SUBJECTIVE TIME SCALE

(1) A method to obtain a scale of subjective time unbiased by the time-order error is suggested. (2) The method gives as by-product some preliminary results on the memory process which gives rise to the time-order error. (3) The method is applied in an experimental study. Two methods, the adjustment method and the reproduction method were used to construct a subjective time scale within the range 1–7 seconds. The former method gave an almost linear R-S curve while the latter resulted in a positively accelerated curve. (4) The time-order error seems to be a very rapid process. The memory curve reaches very near its asymptotic value within a few tenths of a second.     

A square root method of selecting a minimum set of variables in multiple regression: II. A worked example

Thesquare root method of selection has been explained in a previous article. In the present article a worked example is given which illustrates the compactness of the procedure. The square root method is compared with the Wherry-Doolittle method.     

A note on the sampling properties of the Vincentizing (quantile averaging) procedure

To assess the effect of a manipulation on a response time distribution, psychologists often use Vincentizing or quantile averaging to construct group or “average” distributions. We provide a theorem characterizing the large sample properties of the averaged quantiles when the individual RT distributions all belong to the same location-scale family. We then apply the theorem to estimating parameters for the quantile-averaged distributions. From the theorem, it is shown that parameters of the group distribution can be estimated by generalized least squares. This method provides accurate estimates of standard errors of parameters and can therefore be used in formal inference. The method is benchmarked in a small simulation study against both a maximum likelihood method and an ordinary least-squares method. Generalized least squares essentially is the only method based on the averaged quantiles that is both unbiased and provides accurate estimates of parameter standard errors. It is also proved that for location-scale families, performing generalized least squares on quantile averages is formally equivalent to averaging parameter estimates from generalized least squares performed on individuals. A limitation on the method is that individual RT distributions must be members of the same location-scale family.     

A method of estimating item characteristic functions using the maximum likelihood estimate of ability

A method of estimating item characteristic functions is proposed, in which a set of test items, whose operating characteristics are known and which give a constant test information function for a substantially wide range of ability, are used. The method is based on the maximum likelihood estimates of ability for a group of several hundred examinees. Throughout the present study the Monte Carlo method is used.     

CD-CAT初始阶段项目选取方法

CD-CAT是CDA同CAT的相结合的产物,适用于课堂教学,是教师补救教学、学生自我学习的重要工具。作为CD-CAT重要组成部分的初始阶段项目选取方法是影响测验判准率的重要因素。本文基于现有研究和CDA的项目区分度提出了四种新的初始阶段项目选取方法:CTTID法、CDI法、CTTIDR*法和CDIR*法。通过模拟研究发现,在定长的CD-CAT下,题库质量是HD-HV下,初始阶段结束时,CTTIDR*法的PCCR比现有的T阵法高了.2999,比PWKL高了.1707,其它题库下趋势相同。整个测验结束时CTTIDR*法的判准率仍然是最高的。在变长的CD-CAT下,最大后验概率大于.7、.8、.9下,CTTIDR*法的被试平均测验长度比T阵法分别缩短了2.6170、2.2347、1.7470道题。     

A note on interobserver reliability for sequential data

It is well accepted that reliability measures based on simple frequency counts of designated codes are inappropriate for sequential analysis. The stricter point-by-point method, however, is problematic, particularly because it is too strict. It is suggested that a less strict reliability measure based on the number of transitions be used. This method also avoids problems encountered with the point-by-point method and another less stringent method.     

A method for the analysis of complex avian song repertoires

A new method for the analysis of complex avian song repertoires is introduced. The method consists of sending a tape-recorded bird-song signal through five narrow band-pass filters and then to a five-channel high-speed pen recorder. A coding method that allows a researcher to convert the output from the pen recorder to numbers for later analysis by computer is described.