Open AccessA FACTOR ANALYSIS OF THE PICTURE COMPLETION ITEMS OF THE WAIS 1 2
Author(s) -
Saunders D. R.
Publication year - 1959
Publication title -
ets research bulletin series
Language(s) - English
Resource type - Journals
eISSN - 2333-8504
pISSN - 0424-6144
DOI - 10.1002/j.2333-8504.1959.tb00938.x
Subject(s) - wechsler adult intelligence scale , psychology , factor (programming language) , dimension (graph theory) , short forms , multidimensional scaling , test (biology) , comprehension , cognitive psychology , psychometrics , principal component analysis , perspective (graphical) , developmental psychology , statistics , clinical psychology , cognition , artificial intelligence , mathematics , computer science , paleontology , neuroscience , pure mathematics , biology , programming language
Factor analysis is applied to a matrix of item intercorrelations for items of the WAIS Picture Completion subtest, using a sample of 228 college‐oriented males. Three factors are found, and identified in terms of the three principal causes of failure on PC described by Rapaport. The factors are examined in relation to the overall framework of the WAIS reported by Saunders, and found not to require any new dimension. The multidimensional character of PC explains why the items of this subtest have been particularly resistant to scaling for difficulty. Factor I is most appropriately labelled “Maintenance of Contact” versus “Loss of Contact” or withdrawal, is best measured by items 16 and 18, and is not linearly related to any other subtest. This factor, which apparently embodies the unique contribution of PC to the Wechsler battery, might also be called “Recognition of Incongruity.” Factor II is most appropriately labelled “Maintenance of Perspective” versus “Loss of Distance,” is best measured by items 8 and 9, and is related to certain items in the Comprehension subtest. Factor III is related to the effect of “uncertainty” on the subject, is best measured by items 2 and 14, and is related to total scores on the Object Assembly subtest.