scholarly journals An Algorithm to Generate Variates with Desired Intercorrelation Matrix

2004 ◽  
Author(s):  
S. K. Mishra
1964 ◽  
Vol 15 (2) ◽  
pp. 543-548 ◽  
Author(s):  
R. Travis Osborne

In order to estimate the dimensionality of the Wechsler Intelligence Scale for Children, factor analysis was applied to a 30 by 30 intercorrelation matrix of the WISC and four reference tests. The 10 standard WISC subtests, except Coding, were split into two, three, or four parts to yield as many variables as possible. Ss, were 111 Negro pre-school children; mean age, 6 yr. 1 mo.; the mean full scale IQ was 84. Evidence is presented supporting 10 statistically significant orthogonal dimensions. Not all factors are perfectly congruent with the subtest structure of the WISC or concordant with the results of prior factorizations of the WISC at the pre-school level. There is no factor for Block Design apart from Picture Arrangement. The Digit Span subtest splits involve three different WISC factors, two are from the performance section of the test and one is from the verbal section. Coding is involved in only one factor; Manipulation of Areas, one of the non-verbal reference tests. At least 7 of the 10 significant pre-school factors are readily identified by WISC subtests or combinations of WISC subtests.


1969 ◽  
Vol 29 (1) ◽  
pp. 283-288 ◽  
Author(s):  
M. H. Tillman ◽  
W. L. Bashaw ◽  
Michael Bradley

Juurmaa (1967) carried out the first comprehensive study to assess the ability domain of totally blind males. Using Thurstone's centroid method and graphic orthogonal rotation, 9 factors were extracted from a 23 × 23 intercorrelation matrix. Abilities measured included verbal comprehension, arithmetic reasoning, spatial ability, memory, dexterity, kinesthetic memory, tactual discrimination, and auditory sensitivity. One factor, “sensory discrimination,” was obtained from the blind group which had no corresponding factor in a sighted male comparison group. The purpose of the present study was to explore further the interpretation of this unique factor and to raise major methodological issues about Juurmaa's analyses, namely, the make-up of the blind sample, the choice of factor analytic tools, and the arbitrariness of the decision concerning dimensionality. Consideration of these issues in conjunction with a reanalysis of Juurmaa's data, led the authors to conclude that the ability structure of the blind needs considerably more research before any conclusions can be drawn without extreme caution.


1982 ◽  
Vol 54 (2) ◽  
pp. 487-491 ◽  
Author(s):  
Ruth Guttman ◽  
Ilana Shoham

800 individuals were given a battery of 8 spatial tests which had been assembled with the aid of a mapping sentence of four content facets: rule type, dimensionality, presence or absence of rotation, and test format. An intercorrelation matrix of 49 items from these tests was analyzed by Smallest Space Analysis, SSA-I. All three facets formed distinct regions in a two-dimensional projection of a three-dimensional space. It is suggested that further facets be hypothesized to elaborate on the structure of spatial abilities.


1978 ◽  
Vol 43 (2) ◽  
pp. 619-623 ◽  
Author(s):  
Joseph Levin ◽  
Eliezer Karni ◽  
Yaacov Frankel

The Tennessee Self-concept Scale is a two-faceted instrument, one of which is the internal modality of the self-concept and the other one is the self as an external referent. The smallest space analysis of the subscales intercorrelation matrix yielded a definite pattern replicated in 8 studies. The order of the subscales of the internal selves was Am, Act, Accept, and of the external selves was Physical, Social, Personal, Family, Moral/ethical. Psychological interpretations of the order are suggested, and possible effects of some particular features of the instrument are discussed.


1996 ◽  
Vol 83 (1) ◽  
pp. 35-40 ◽  
Author(s):  
John R. Sumerlin ◽  
Iola Thompson ◽  
Nolan Thaxton ◽  
S. A. Berretta

Principal components analysis of an intercorrelation matrix for the Physical Self-efficacy Scale in an all black sample of 320 mostly confirmed the original validation study of Ryckman, Robbins, Thornton, and Cantrell who used a wholly white sample; however, the analysis identified items with factor loadings at criterion on more than one factor, one item that loaded on a different subscale, and additional factors. It is unknown whether differences in this sample are attributable to race or other influences. Further investigation is suggested.


1980 ◽  
Vol 46 (3) ◽  
pp. 849-850 ◽  
Author(s):  
Bernard Lubin ◽  
Marc E. Caplan ◽  
James F. Collins

Forms E, F, and G of the Depression Adjective Check Lists (DACL) were administered to 240 college undergraduates. Half completed this “Actual” condition before random assignment either to a “Fake Good,” “Fake Bad,” or “Fake Average” condition, 20 males and 20 females per experimental condition, and half were randomly assigned first to one of the experimental conditions and then to the “Actual” condition. All 36 intercorrelations were significant and ranged from .65 to .92. Mean rs for the experimental conditions when they occurred first were .90, .87, and .77 and .82, .66, and .86 when they occurred second (“Fake Bad,” “Fake Good,” and “Fake Average,” respectively). An intercorrelation matrix of all six half-lists of Forms E, F, and G showed all rs significant. These findings increase confidence in the consideration of Lists E, F, and G of the check list as equivalent and in the potential utility of the half-lists.


1983 ◽  
Vol 11 (2) ◽  
pp. 83-89 ◽  
Author(s):  
W. Gary Howard ◽  
David Redfering

The legal profession is replete with assertions about the relationships of social, economic, and psychological characteristics to jury verdicts. The purpose of this research was to identify and test some of these concepts. A survey of jurors serving on criminal cases was conducted which incorporated pertinent socio-economic, demographic, psychological and experiential factors. An intercorrelation matrix and a linear step-wise regression model were employed to identify those variables which showed a statistically significant relationship to a trial vote of guilty. None of the notions of attorneys regarding socio-economic characteristics had a significant relationship to jury verdicts; however, items concerning authoritarianism, religious ideology, and prior experience with courts did correlate with jury behavior.


Psychometrika ◽  
1987 ◽  
Vol 52 (4) ◽  
pp. 555-564 ◽  
Author(s):  
Haruo Yanai ◽  
Bishwa Nath Mukherjee

1967 ◽  
Vol 25 (2) ◽  
pp. 517-521 ◽  
Author(s):  
James C. Reed ◽  
Kathleen B. Fitzhugh

A factor analysis was performed on the intercorrelation matrix of the subtests of WB-I and WAIS. 2, 3, and 11 factors were orthogonally rotated. The two-factor and three-factor solution yielded factors that had previously been found for WB-I and WAIS separately. In the eleven-factor solution, the corresponding subtest from each scale did not load in pairs. Implications for generalizing conclusions from one form of the test to the other are discussed.


1996 ◽  
Vol 78 (1) ◽  
pp. 79-82 ◽  
Author(s):  
Félix Neto

This study investigated the relationship between shyness and sociability among 194 Portuguese college students. A factor analysis of the intercorrelation matrix of shyness and sociability items gave two distinct factors indicating that shyness and sociability are distinct personality dispositions. Shyness scores showed only a moderate negative correlation with sociability. Further, the measures of shyness and sociability had different patterns of correlations with other personality scales. The Portuguese translation of shyness and sociability scales seems sufficiently reliable and valid to warrant further use with Portuguese individuals.


Sign in / Sign up

Export Citation Format

Share Document