A COMPARISON OF EXAMINEE SAMPLING AND MULTIPLE MATRIX SAMPLING IN TEST DEVELOPMENT

For the purpose of obtaining data to use in test development, multiple matrix sampling (MMS) plans were compared to examinee sampling plans. Data were simulated for examinees, sampled from a population with a normal distribution of ability, responding to items selected from an item universe. Three item universes were considered: one that would produce a normal distribution of test scores, one a moderately platykurtic distribution, and one a very platykurtic distribution. When comparing sampling plans, total numbers of observations were held constant. No differences were found among plans in estimating item difficulty. Examinee sampling produced better estimates of item discrimination, test reliability, and test validity. As total number of observations increased, estimates improved considerably, especially for those MMS plans with larger subtest sizes. Larger numbers of observations were needed for tests designed to produce a normal distribution of test scores. With an adequate number of observations, MMS is seen as an alternative to examinee sampling in test development.     

AN EXPERIMENTAL COMPARISON OF ITEM SAMPLING AND EXAMINEE SAMPLING FOR ESTIMATING TEST NORMS

An empirical comparison of the accuracy of item sampling and examinee sampling in estimating norm statistics. Item samples were composed of 3, 6, or 12 items selected from a total test of 50 multiple-choice vocabulary questions. Overall, the study findings provided empirical evidence that item sampling is approximately as effective as examinee sampling for estimating the population mean and standard deviation. Contradictory trends occurred for lower ability and higher ability student populations in accuracy of estimated means and standard deviations when the number of items administered increased from 3 to 6 to 12. The findings from this study indicate that the variation of sequences of items occurring in item sampling need not have a significant affect on test performance.     

A MONTE CARLO STUDY OF ITEM SAMPLING (VERSUS TRADITIONAL SAMPLING) FOR NORM CONSTRUCTION1

Using a computer-based model of an item trace line, a random sampling experiment concerned with comparing item sample estimates to traditional (examinee) sample estimates of the mean and variance of a distribution of test scores was conducted. The results indicated that the optimal method for estimating a test's parameters may depend on several conditions. As expected, item sampling proved superior to traditional sampling in estimating test means under all conditions. However, with certain test lengths, ranges of item difficulty, and discrimination, traditional sampling provided better estimates of test variance than did item sampling.     

香农采样理论的扩展及应用

本文基于香农采样理论,通过对离散序列重构信号的研究,给出一种在完备正交积分变换下的广义采样定理及其应用。     

AN APPLICATION OF ITEM-EXAMINEE SAMPLING TO SCALING ATTITUDES

The post mortem item-examinee sampling investigation described herein explored the feasibility of using item-examinee sampling to estimate scale values denoting degree of affect toward stimuli when measured by the method of paired-comparisons. Results indicate that such scale values can be approximated satisfactorily through item-examinee sampling. Defining one observation as the response made by one examinee to one item, the similarity between the estimated scale values and normative scale values increased generally with increases in the number of observations acquired by the sampling plan.     

A NOTE ON TERMINOLOGY AND NOTATION IN MATRIX SAMPLING

The proliferation of terminology in matrix sampling is beginning to cause some minor problems. Presented herein is one set of terms and notation which hopefully will facilitate communication among individuals engaged in all aspects of matrix sampling.     

AN EXAMINATION OF THE CONTEXT EFFECT IN ITEM SAMPLING

Item sampling and/or multiple matrix sampling techniques have been recommended for a variety of purposes. For some of these purposes, it must be assumed that examinee performance on a set of items is unaffected by the conditions under which the items are taken (i.e., no context effect exists). In this paper factors that may lead to a context effect among high school students are discussed. The net effect of such factors on examinee scores for an English test and a mathematics test is investigated empirically. For the English test there was little support for the existence of a context effect, However, a definite context effect was found for the mathematics test.     

AN INVESTIGATION OF THE CONTEXT EFFECT IN MATRIX SAMPLING1

Practical use of the matrix sampling (i.e. item sampling) technique requires the assumption that an examinee's response to an item is independent of the context in which the item occurs. This assumption was tested experimentally by comparing the responses of examinees to a population of items with the responses of examinees to item samples. Matrix sampling mean and variance estimates for verbal, quantitative, and attitude tests were used as dependent variables to test for differences between the “context” and “out-of-context” groups. The estimates obtained from both treatment groups were also compared with actual population values. No significant differences were found between treatments on matrix sample parameter estimates for any of the three types of tests.     

ITEM-EXAMINEE SAMPLING PROCEDURES AND ASSOCIATED STANDARD ERRORS IN ESTIMATING TEST PARAMETERS

Selected parameters for a negatively skewed and a normally distributed normative distribution were estimated in a post mortem item-examinee sampling investigation. Manipulated systematically were number of subtests, number of items per subtest, and number of examinees responding to each sub-test. Each item-examinee sampling procedure was replicated five times. Defining one observation as the score received by one examinee on one item, the results of this investigation support the conclusion that, in estimating parameters by item-examinee sampling, the variable of importance is not the item-examinee sampling procedure but is instead the number of observations obtained by that procedure. Degree of skewness in the normative distribution and failure to distribute all items among subtests were found to be relatively unimportant variables.     

AN INVESTIGATION OF AN EXTENSION OF ITEM SAMPLING WHICH YIELDS INDIVIDUAL SCORES1

The sampling procedures were designed so that the full matrix of item variances and covariances could be estimated. Three subtest sizes were investigated- subtests of size five, nine and sixteen items. In each of these implementations a double cross validation was used yielding two predicted scores for each individual. Discrepancy measures were also computed showing the difference between the observed and the predicted scores. The prediction of individual scores was accomplished within various ranges of error. The correlations between predicted scores and observed scores ranged from the .70′s to the .90′s, depending on the number of predictor variables used. The procedure is applicable in situations in which large numbers of individuals are tested or in situations where multiple measures are taken.     

FURTHER RESULTS ON THE STANDARD ERRORS OF ESTIMATE ASSOCIATED WITH ITEM-EXAMINEE SAMPLING PROCEDURES

Defining one observation as the score received by one examinee on one item, the results of this investigation suggest that, for a given test length, item-examinee sampling procedures having the same number of observation have, for all practical purposes, the same standard error in estimating μ but different standard errors in estimating σ. Additionally, the variance of the item difficulty indices (proportion answering the item correctly) was found to be a significant factor in accounting for differences in standard errors of estimating μ between normative distributions differing primarily in degree of skewness.     

SCRAMBLING CONTENT IN ACHIEVEMENT TESTING: AN APPLICATION OF MULTIPLE MATRIX SAMPLING IN EXPERIMENTAL DESIGN

This study was designed to research the question of scrambling item content in the construction of achievement tests, so that very general implications could be drawn for both examinee and item populations. To achieve this generality, the methodology of multiple matrix sampling was combined with a simple two group experimental design: a random group of 8th graders responded to mathematics, science, social studies, reading, and language arts achievement items organized in a scrambled (random) test format, while another random group responded to the same items organized in a fixed (segregated by subject matter) test format. The results indicated that scrambling cognitive test items has minimal or no effect on mean examinee test performance or on any of the other parameters included in the analysis.     

“两双”、“口嘴”的意象图式

文章从隐喻的意象图式角度分析"两"与"双"、"口"与"嘴"在语言表达中的同现限制。"两"与"双"的字形不同,它们的意象图式也不同:"两"是平分图式,"双"是配偶图式;"两"、"双"词语在言语交际中的意义和用法也就不同:选择"两"的语境强调"分而为二、两相对立",在选择"双"的语境强调合二为一、配成一对。"口"与"嘴"隐喻投射到对事物域的认识当中,但是各有侧重:"N+口"强调的是"容器通往外界的地方","N+嘴"强调的是"容器中形状突出的部分"。     

《孟子》中“正”“长”“贤”“轻”“良”的形容词用法

通过对《孟子》中“正”、“长”、“贤”、“良”、“轻”5个形容词的语义和语法功能作较为细致的描写,试图从中寻找出《孟子》中不同的形容词在语法功能上的共性与个性,并进一步提出问题,即这些共性与个性的历时和共时的差异,以及造成这些差异的原因。     

从"思辩"经"实证"到"理解"——试比较三种类型的教育研究范式

教育研究大致经历了从哲学--思辩研究范式经科学--实证研究范式到人文--理解研究范式的历程.范式的转换不只是方法的更替,更是哲学理论背景、思维方式、价值取向、研究过程等的变更和创新.本文以众多学者的研究为基础尝试作多方面的比较.