Premium
THE BI‐POLAR AND UNIDIRECTIONAL MEASUREMENT OF INTELLIGENCE
Author(s) -
THOMPSON J. W.
Publication year - 1961
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.2044-8295.1961.tb00763.x
Subject(s) - polarity (international relations) , psychology , social psychology , cognitive psychology , polar , test (biology) , resolution (logic) , artificial intelligence , computer science , chemistry , paleontology , biochemistry , physics , astronomy , biology , cell
Insufficient distinction has been made in psychology between bi‐polarity which can be converted to unidirectional measurement, and bi‐polarity which is not thus convertible. As one or other kind of polarity is entailed by many kinds of psychological data, methods of scoring and measuring techniques ought to take into consideration the kind of polarity which the data entail. This is most easily illustrated in the scaling of attitudes. In the use of factor analysis, unless attention is drawn to the difference between what the writer calls ‘convertible’ or tautologous and ‘inconvertible’ bi‐polarity, the assumption is too readily made that it is not of importance whether factors having all positive loadings emerge from an analysis. This laxity is due solely to convention and is not justifiable on logical grounds. In the field of intelligence, although it is taken for granted that intelligence is a unidirectional concept, this may also be merely due to an habitual mode of thinking. It would therefore appear desirable to test the hypothesis that intelligence is inconvertibly bi‐polar, perhaps by using a scoring procedure similar to that adopted in multi‐dimensional tests for the measurement of values. If the hypothesis were proved, ‘brightness’ tests and tests for ‘backwardness’ scored in this way might lead to the resolution of existing difficulties connected with the concept of intelligence.