Premium
Factor analysis of a picture‐preference test of personality
Author(s) -
Auld Frank,
Kline Paul
Publication year - 1984
Publication title -
british journal of psychology
Language(s) - English
Resource type - Journals
SCImago Journal Rank - 1.536
H-Index - 92
eISSN - 2044-8295
pISSN - 0007-1269
DOI - 10.1111/j.2044-8295.1984.tb01916.x
Subject(s) - psychology , psychoticism , personality , eysenck personality questionnaire , neuroticism , developmental psychology , social psychology , cronbach's alpha , preference , personality test , psychometrics , test (biology) , test validity , big five personality traits , extraversion and introversion , statistics , paleontology , mathematics , biology
A picture‐preference test of personality, developed by Cowan (1967), has shown promise of usefulness but poor reliability of its subscales. In order to create more reliable subscales from Cowan's 106‐item test, the authors chose two small groups of items as seeds for scales and then tried adding items to the original scales to increase the K‐R 20 coefficient, adding items to each of these groups so long as the internal consistency index increased. After these two scales had been created, the authors factor analysed 17 variables – including the two scales created from the seed items – that had been derived from the 106 items, using a sample of 1180 subjects. Three factors emerged: ( a ) a factor measuring deviance, related to Eysenck's ‘psychoticism’, ( b ) a factor defined by a scale measuring ‘Avoidance of sexual intimacy’, and ( c ) a factor measuring overcontrol of impulse. Demonstration of the first factor suggests that the picture‐preference method offers an alternate way to measure Eysenck's P; demonstration of the other two factors suggests that there are two hitherto unstudied dimensions of personality that can be measured by Cowan's test.