Appendix CThe following multivariate analyses of a small nexus of five interrelated variables are intended only as a didactic example of different ways of looking at a
nexus. They merely illustrate what the results of these analyses look like when
applied to a real (but small) nexus. The example is not intended to make any
particular argument. Detailed computational algorithms for the various types of
analysis can be found in textbooks on multivariate statistical methods. The correlation matrix and computational procedure for the path analysis were taken
from Li ( 1975, pp. 324-328).The five variables in the nexus, listed in temporal order, are: father's education (FED), father's occupation (FOC), his child's IQ in childhood (CIQ), the child's education as total years of schooling (CED), and the child's adult IQ
Multivariate Analyses of a
|A. ||The correlations among these five variables, based on a large sample of
white males, aged twenty-five to sixty-four, are shown in Table C.1.|
|B. ||A principal components PC analysis yields two meaningful components,
labeled I and II (see Table C.2). The remaining three PCs have been dropped
based on the criterion that their eigenvalues (or latent roots) are less than one.|
|C. ||Also shown are the multiple correlations (Rs) of each variable with every
other variable in the matrix. They indicate the degree to which any given variable in the nexus can be predicted by all of the other variables in the nexus.
The proportion of variance that any given variable has in common with all of
the other variables is indicated by R2. The fact that all of the loadings on PC I
(i.e., the general factor in this matrix) are all fairly large indicates that this is a
quite close-knit nexus; the main division among the variables is clear from the
opposite signs for the father and child variables in PC II. The multiple R and
the R2 for Father's Education are the smallest in the whole set, showing that it
is the least well predicted by all of the other variables. The Child's Education|
Questia, a part of Gale, Cengage Learning. www.questia.com
Book title: The G Factor:The Science of Mental Ability.
Contributors: Arthur R. Jensen - Author.
Place of publication: Westport, CT.
Publication year: 1998.
Page number: 593.
This material is protected by copyright and, with the exception of fair use, may
not be further copied, distributed or transmitted in any form or by any means.