摘要:

It is argued that current scaling models when they are applied to perceptual and mnemonic processes are out of touch with psychological theory. As an alternative, a psychological model is proposed in which stimuli are analysed by a series of feature tests, each feature test being dependent on the results of earlier feature tests in the series and on the other features present in the stimulus. Computer programs are described which fit such models to data, the output of such programs being the best‐fitting hierarchy of feature tests. Detailed analyses are produced for data drawn from studies of speech perception and memory for speech; differences between normal, deaf and aphasic subjects are also discussed. It is claimed that, in comparison with normal scaling methods or with descriptions based simply on raw confusions, this approach offers a more insightful analysis of such dat

ISSN:0007-1102    

DOI:10.1111/j.2044-8317.1975.tb00544.x

出版商:Blackwell Publishing Ltd    

年代:1975

数据来源: WILEY

摘要:

Extends the theory presented in Smithet al.(1975) to larger confusion matrices. Stimuli and responses are described by a series of features which take a discrete set of values. Processing models are proposed where each feature is tested independently (Independence model), where the processing of a given feature depends on the responses that have been made in testing for other features (Response Conditional model) or where the processing of a given feature depends on whether other features have been correctly identified or not (Correct Conditional model). Computer programs are described which fit these models to arbitrarily large confusion matrices. Difficulties in interpreting the behaviour of these models with real data are reviewed and solutions are offered. These points are illustrated by detailed analyses of confusion matrices obtained in studies of speech perception, memory for speech and semantic memory.

ISSN:0007-1102    

DOI:10.1111/j.2044-8317.1975.tb00545.x

出版商:Blackwell Publishing Ltd    

年代:1975

数据来源: WILEY

摘要:

Under a stated assumption, a formula is derived for estimating at any specified ability level the relative efficiency of number‐right score and formula score. Analysis of one set of data is summarize

ISSN:0007-1102    

DOI:10.1111/j.2044-8317.1975.tb00546.x

出版商:Blackwell Publishing Ltd    

年代:1975

数据来源: WILEY

摘要:

This paper discusses briefly the classic multiple linear regression model and indicates the principal ways in which its assumptions are inadequately met when it is used as a model for prediction purposes in the social sciences. It also considers the damping effects of errors of measurement and of selective sampling on estimates of partial regression and multiple correlation coefficients and describes techniques whereby these effects may in part be overcome.

ISSN:0007-1102    

DOI:10.1111/j.2044-8317.1975.tb00547.x

出版商:Blackwell Publishing Ltd    

年代:1975

数据来源: WILEY

摘要:

Introduces a new criterion for analytic rotation based upon the concept of simple structure. The rationale for the method stems from a consideration of the way in which the variance of squared factor loadings is partitioned as in an analysis of variance. It is argued that a maximization of the relative amount of squared factor loading variation which is due to the ‘interaction’ of factors and tests should produce results consistent with the simple structure hypothesis. Other methods of orthogonal rotation, such as quartimax and varimax, may be defined as linear combinations of the component sums of squares. The criterion of this paper differs from previous ones in that it is defined as a ratio of sums of squares. The use of a gradient method to maximize the criterion is discussed. A method for employing the Fletcher‐Powell algorithm is presented. The solutions obtained with three well‐known illustrative examples are similar to those obtained with varimax, although two of them are somewhat closer than varimax to Harman's subjective so

ISSN:0007-1102    

DOI:10.1111/j.2044-8317.1975.tb00548.x

出版商:Blackwell Publishing Ltd    

年代:1975

数据来源: WILEY

摘要:

WhenFminanovais significant, Scheffé's procedure provides no limit to the number of contrasts which can be declared significant. The new procedure described here provides a limit (r) to the number of significant contrasts which can be asserted. If assertions are made forv1contrasts, which includerdeclarations of significance, the expectation ofr/v1will be a linearly decreasing function ofv1if traditional criticalFvalues are used. A table is provided of new criticalFvalues which set this expectation at a constant (0.05 or 0.01). The new criticalFvalues are recommended for use whenever contrasts are sought,post hoc, for significance

ISSN:0007-1102    

DOI:10.1111/j.2044-8317.1975.tb00549.x

出版商:Blackwell Publishing Ltd    

年代:1975

数据来源: WILEY

摘要:

The squared correlation coefficient,w2, between an empirically chosen linear function of predictors, B̂0+ B̂′x, and a criterion,y, is employed as a measure of predictive precision. This coefficient is defined over the entire population but is conditional on B̂. Assuming a multinormal distribution for x and y, approximations for the expected value and variance ofw2are derived. If too many predictors are employed, precision of prediction can decrease. This is illustrated by means of an example of a sequence of values of ℰ (w2). A function of the sample squared multiple correlation coefficient,r2, is proposed as an estimator ofw2. Results of Monte Carlo experiments are employed to give an impression of the precision of the estimates ofw2, and the accuracy of the approximations for ℰ (w2) and

ISSN:0007-1102    

DOI:10.1111/j.2044-8317.1975.tb00550.x

出版商:Blackwell Publishing Ltd    

年代:1975

数据来源: WILEY

摘要:

For the situation in which the contrasts of interest are limited to the (K/2) comparisons among the means ofKsamples, Spjøtvoll&Stoline's (1973) extension of Tukey's multiple‐comparison procedure is compared with Scheffé's method (1959) and Dunn's procedure (1961) for significance levels not exceeding 0.05. Rules are given for determining if any method is uniformly preferable (best for all contrasts). Non‐uniform preference rules are also given and applied to some examples. Auxiliary tables are provided for significance levels 0.01 and 0.05 for several values ofKandv, the number of degrees of freedom of an independent variance estimate. It is shown that the extended Tukey procedure is uniformly preferable when the sample sizes are equal or ‘nearly’ equal, while Dunn's and, in some cases, Scheffé's method is uniformly preferable when all sample sizes are ‘sufficientl

ISSN:0007-1102    

DOI:10.1111/j.2044-8317.1975.tb00551.x

出版商:Blackwell Publishing Ltd    

年代:1975

数据来源: WILEY

摘要:

A technique for evaluating the non‐randomness of a proximity matrix prior to a cluster analysis has been suggested by Ogilvie which incorporates an asymptotic result developed by Erdös&Rényi for the size of the largest connected component in a random graph. On the basis of Ogilvie's Monte Carlo comparisons, the use of Erdös&Rényi's asymptotic formula appears justified even when the number of objects in the set to be clustered is small. Several additional empirical comparisons are given in this paper that would indicate a more cautious attitude toward the use of the Erdös‐Rényi asymptotic formula in a non‐asympto

ISSN:0007-1102    

DOI:10.1111/j.2044-8317.1975.tb00552.x

出版商:Blackwell Publishing Ltd    

年代:1975

数据来源: WILEY

摘要:

Two procedures for estimating the mean squared error of prediction of an empirically determined linear prediction equation are examined. The method usually employed makes use of a second validation sample; another method makes use of the calibration sample alone. The mean squared error of estimation is derived for each of the two estimation procedures and a comparison made. A test is provided also for the hypothesis that use of a prespecified subset of predictors results in no increase in the expected mean squared error of prediction.

ISSN:0007-1102    

DOI:10.1111/j.2044-8317.1975.tb00553.x

出版商:Blackwell Publishing Ltd    

年代:1975

数据来源: WILEY