摘要:

The paper discusses some practical implications of a new approach to factor analysis expounded in Bartholomew (1984). The problem with which it deals may be posed in the following way. Givenpvariables measured onnindividuals, to find one or more functions of these variables which, in some sense, effectively summarize what is common to the larger set. It was shown in the earlier paper that, under fairly general conditions, such a summarization can efficiently be achieved by linear functions of the manifest variables. Here, the consequences of this result are explored in relation to such matters as rotation of axes, factor scores, non‐linear and hierarchical factor analysis.The conclusion reached is that while almost all of the existing numerical methods relating to the estimation of factor models remain relevant, the presentation and interpretation of the results is substantially modified. In particular, the focus is shifted from unobservable factors to observable functions of the data and the distinction between orthogonal and oblique rotations has no significanc

ISSN:0007-1102    

DOI:10.1111/j.2044-8317.1985.tb00811.x

出版商:Blackwell Publishing Ltd    

年代:1985

数据来源: WILEY

摘要:

The use of the multiple runs distribution in the assessment of categorical clustering in free recall experiments is discussed. The results of comparing five large sample approximations to this distribution are presented and guidelines as to which approximation to use are given.

ISSN:0007-1102    

DOI:10.1111/j.2044-8317.1985.tb00812.x

出版商:Blackwell Publishing Ltd    

年代:1985

数据来源: WILEY

摘要:

A regression model for the order of recall of items in a free recall experiment which allows the effects of covariates to be measured is presented. The estimation of the parameters via Newton–Raphson iteration is discussed, as are methods of assessing the goodness of fit of the model. Diagnostic checks are illustrated on two sets of data and finally some extensions and drawbacks are considere

ISSN:0007-1102    

DOI:10.1111/j.2044-8317.1985.tb00813.x

出版商:Blackwell Publishing Ltd    

年代:1985

数据来源: WILEY

摘要:

Wetherill&Levitt (1965) suggested a class of sequential procedures for estimating points on a psychometric function. They investigated the properties of staircase estimators in yes‐no experiments. Such procedures have been used widely in two interval forced choice (21FC) experiments. In this paper large sample properties of estimators from yes‐no and 2IFC experiments are compared. These estimators are also compared by computer simulation. Results indicate that estimates from 2IFC experiments can have much greater biases and mean squared errors than those from yes‐no experi

ISSN:0007-1102    

DOI:10.1111/j.2044-8317.1985.tb00814.x

出版商:Blackwell Publishing Ltd    

年代:1985

数据来源: WILEY

摘要:

Given two partitions of the same set, an index of agreement between the two partitions is given by the number of pairs of elements appearing together in an equivalence class in both partitions. In this paper the exact permutational probability distribution of this index is derived, i.e. the probability distribution of the index conditional upon the cardinality of the equivalence classes in both partitions.

ISSN:0007-1102    

DOI:10.1111/j.2044-8317.1985.tb00815.x

出版商:Blackwell Publishing Ltd    

年代:1985

数据来源: WILEY

摘要:

Current methods of assessing nominal scale response agreement between two raters allow each rater to make only a single response per object. When raters are uncertain in their responses, such methods can underestimate the extent to which agreement exists in the range of possible classifications envisaged by raters for each object. Accordingly, an exact unweighted test procedure is developed for the analysis of nominal scale response agreement in the event of rater uncertainty. A rater who is undecided about the appropriate category label for an object is allowed to assign more than one category label to that object. The researcher can choose an operational definition of agreement under uncertainty which is appropriate to the goal of a study. The approach, which is based on the combinatorial technique of rook methodology, supplies the exact null distribution of the amount of agreement between two raters who may make multiple responses per object. The present findings also provide, as a special case, exact results for the unweighted single‐response approac

ISSN:0007-1102    

DOI:10.1111/j.2044-8317.1985.tb00816.x

出版商:Blackwell Publishing Ltd    

年代:1985

数据来源: WILEY

摘要:

The test scores of some examinees on a multiple choice test may not provide satisfactory measures of their abilities. The goal of appropriateness measurement is to identify such individuals. Earlier theoretical and experimental work considered examinees answering all, or almost all, test items. This article reports research that extends appropriateness measurement methods to examinees with moderately high non‐response rates. These methods treat non‐response as if it were a deliberate option choice and then attempt to measure the ‘appropriateness’ of the pattern of option choices. Earlier studies used only the dichotomous pattern of ‘right’ and ‘not right’ answers. A general polychotomous model is introduced along with a technique called ‘standardization’ designed to reduce the observed confounding between measured appropriateness and ability. A standardized appropriateness index based on a polychotomous model yielded higher rates of detection of simulated spuriously low examinees than the analogous index based on a dichotomous model. However, the converse was true for simulated spuriously high examinees. Standardization was found to reduce greatly the interaction between ability and measu

ISSN:0007-1102    

DOI:10.1111/j.2044-8317.1985.tb00817.x

出版商:Blackwell Publishing Ltd    

年代:1985

数据来源: WILEY

摘要:

Three types of restricted latent class models for binary data are discussed: models assuming equalities of certain latent parameters, linearly constrained latent class analysis, and linear logistic latent class analysis. The basic equations of the latter model state the decomposition of the log odds of the item latent probabilities and of the class sizes into weighted sums of basic parameters representing the effects of predictor variables which are hypothesized to be relevant indicators for the original parameters. The maximum likelihood equations for these effect parameters and a criterion for their local identifiability are given. Further, statistical tests for goodness of fit are sketched, and the practical application of linear logistic latent class analysis is demonstrated by several examples. They comprise some known scaling models as well as a simple model with located classes and items, a model which relates item difficulty to item structure, and a model for measuring latent changes. Finally, the generalization of linear logistic latent class analysis to polytomous manifest variables is outlined.

ISSN:0007-1102    

DOI:10.1111/j.2044-8317.1985.tb00818.x

出版商:Blackwell Publishing Ltd    

年代:1985

数据来源: WILEY

摘要:

Asymptotically distribution‐free (ADF) methods for the analyses of covariances or correlations require the calculation of an estimated variance–covariance matrix of those covariances or correlations.Formulae for this variance–covariance matrix are well known from the literature. However, it turns out that, in particular for the correlations, this matrix is quite elaborate because it seems to be necessary to compute and store fourth‐order standardized sample moments.In this paper we show that this can be avoided and that computations can be done in the data matrix directly. This may save a lot of core storage and computation time

ISSN:0007-1102    

DOI:10.1111/j.2044-8317.1985.tb00819.x

出版商:Blackwell Publishing Ltd    

年代:1985

数据来源: WILEY

摘要:

Users of the Cooley–Tukey Fast Fourier Transform algorithm may wish to test the hypothesis that their data contain genuine periodicities rather than a sequence of independent values. Tables are provided to enable them to do this at three different levels of significanc

ISSN:0007-1102    

DOI:10.1111/j.2044-8317.1985.tb00820.x

出版商:Blackwell Publishing Ltd    

年代:1985

数据来源: WILEY