摘要:

The statistical theory is developed for a simple model, originally presented by Batchelder&Riefer (1980), that is capable of separately measuring storage and retrieval processes in human memory. The model is intended for use by researchers who have hypotheses about the effects of independent variables on storage and retrieval that are general enough not to demand any particular human memory paradigm for their test. In Section 1, the statistical model is presented in detail, along with an experimental paradigm for collecting data to be analysed by the model. In Section 2, parameter estimation for the model is provided assuming independent observations under constant conditions. Confidence regions for the parameters are also provided. In Section 3, hypothesis‐testing methods based on the likelihood ratio method are developed for a variety of hypotheses in the case where observations are available from several experimental conditions. These methods are also illustrated with a data set. In the conclusion, the relative advantages of the statistical model over traditional methods for studying storage and retrieval are discussed. Specifically, an example is presented where the traditional ANOVA approach yields very misleading conclusion

ISSN:0007-1102    

DOI:10.1111/j.2044-8317.1986.tb00852.x

出版商:Blackwell Publishing Ltd    

年代:1986

数据来源: WILEY

摘要:

In a set of examinations it is often the case that not every candidate takes every paper and yet a single overall mark must be assigned to each of the candidates. Some methods for tackling the problem of producing such marks, and difficulties they may lead to, are discussed. A fairly general framework into which many particular proposals can be fitted is described.

ISSN:0007-1102    

DOI:10.1111/j.2044-8317.1986.tb00853.x

出版商:Blackwell Publishing Ltd    

年代:1986

数据来源: WILEY

摘要:

A new algorithm for computing the exact distribution of the number of runs in a sequence is derived. A number of large sample approximations is discussed. It is shown that an easily calculated test which takes the number of runs per category into account is more sensitive than the usual test, based on the global number of runs. The test statistic is asymptotically chi‐squared distributed. A general index of agreement between categorical and sequential data is propose

ISSN:0007-1102    

DOI:10.1111/j.2044-8317.1986.tb00854.x

出版商:Blackwell Publishing Ltd    

年代:1986

数据来源: WILEY

摘要:

A randomization procedure is proposed for comparing the means of two or more groups with respect to several variables. The procedure allows the determination of the significance level to apply with testing individual variables such that the probability of declaring any significant differences by chance alone is α. The Bonferroni inequality is sometimes used to obtain the same end. However the Bonferroni inequality gives significance levels that are too stringent with correlated variables

ISSN:0007-1102    

DOI:10.1111/j.2044-8317.1986.tb00855.x

出版商:Blackwell Publishing Ltd    

年代:1986

数据来源: WILEY

摘要:

In the light of evidence suggesting that replication attempts are often markedly deficient in power, a case can be made for a more rigorous approach to sample size determination in such studies. A method is proposed which uses information provided by the original experiment to arrive at a distribution of probable effect sizes. The sample size to be employed in a replication attempt is that which supplies an expected power of the desired amount over the distribution of probable effect sizes. To facilitate the application of the method, formulae arc developed which provide a rapid, straightforward means of establishing the appropriate sample size when the significance test employed is the standard normal z test. Examples of sample size determination using the formulae are presented for replication studies investigating the difference between two means, the difference between two correlations, and the difference between two proportions.

ISSN:0007-1102    

DOI:10.1111/j.2044-8317.1986.tb00856.x

出版商:Blackwell Publishing Ltd    

年代:1986

数据来源: WILEY

摘要:

It is shown that in the two‐parameter Birnbaum latent trait model, all the parameters are identifiable under wide conditions, provided the number of items exceeds two, and all the slopes are strictly positiv

ISSN:0007-1102    

DOI:10.1111/j.2044-8317.1986.tb00857.x

出版商:Blackwell Publishing Ltd    

年代:1986

数据来源: WILEY

摘要:

A number of studies have questioned the robustness of the parametricFratio to assumption violation and have recommended alternative procedures. These procedures differ in terms of whether they are parametric or non‐parametric, and whether they involve data modification or not. The present study compares Type I error rates and power of these approaches when they are calculated on data from a common mixed normal distribution. The tests investigated areF(parametric test, without data modification),Fm(parametric test, with data modification), Approximate randomization test (non‐parametric test, without data modification), and Mann–WhitneyU(non‐parametric test, with data modification). Contrary to a previous study, the approximate randomization approach did not show good power characteristics. TheFmtest and the Mann–WhitneyUdid show good power, with the former outperforming t

ISSN:0007-1102    

DOI:10.1111/j.2044-8317.1986.tb00858.x

出版商:Blackwell Publishing Ltd    

年代:1986

数据来源: WILEY

摘要:

This paper investigates the prediction of a common factor in a single common factor model with infinite items when the structural parameter vector (factor loadings and unique variances) is unknown. A condition in terms of the sample sizenand the number of itemspis established under which the regression predictor for a unique common factor, in which the parameter vector is replaced by the least squares estimator, converges to it in quadratic mean. The condition is thatp→ ∞ andp2/n→ 0 under some mild assump

ISSN:0007-1102    

DOI:10.1111/j.2044-8317.1986.tb00859.x

出版商:Blackwell Publishing Ltd    

年代:1986

数据来源: WILEY

ISSN:0007-1102    

DOI:10.1111/j.2044-8317.1986.tb00860.x

出版商:Blackwell Publishing Ltd    

年代:1986

数据来源: WILEY

摘要:

Book reviewed in this article:New Developments in Statistics for Psychology and the Social Sciences.Edited byA. D. Lovie.Factor Analysis and Related Methods.ByRoderick P. McDonald.Exploring Data Tables, Trends and Shapes.Edited byDavid C. Hoaglin, Frederick Mosteller&John W. Tukey.The Intelligence Men: Makers of the IQ Controversy.ByR. E. Fancher.

ISSN:0007-1102    

DOI:10.1111/j.2044-8317.1986.tb00861.x

出版商:Blackwell Publishing Ltd    

年代:1986

数据来源: WILEY