This article demonstrates the congruence of the inductive and deductive models in logic and studies this congruence in the context of new test items designed to exemplify general-to-particular induction, a field virtually untapped in psychology to this date. Results obtained from the study corroborate the logical foundations of deduction and induction as convergent models. Factor analysis of the items demonstrated factorial convergence, which lends additional support to the postulate of convergence.Requests for reprints should be sent to Magda Colberg,

The purpose of this paper is to define logic‐based measurement as a method of measurement in which all measures (test items) are constructed according to the applicable inferential formulae of logic. Such a method of test construction brings into verbal reasoning tests the inferential objectivity of numerical reasoning tests. The availability of such a technology has powerful implications in personnel selection regarding real‐life prediction of inferential performance, the legal defensibility of personnel selection tests, the design of economical, inferentially nonredundant tests, and the design of test taxonomies in which basic inferential processes can be thoroughly sampled. These implications are discussed in this paper. Also the paper includes a discussion of the linkage of logic‐based measurement with other recently developed technologies such as validity generalization and computer‐tailored testing.

This paper presents a discussion of the problems inherent in verbal test taxonomies which are not based on the schematic domain of logic. This discussion is made in the context of syntactically complex tests such as reading comprehension and inference. A suggested structure for a logically-constructed taxonomy, which has been developed for use in a Federal testing program, is delineated.

No abstract

No abstract