Some implications of a social learning theory for the prediction of goal directed behavior from testing procedures.

Many sophisticated observers are aware that a wide gap exists between personality theory and the techniques or procedures used to measure personality variables. The low level of prediction of such testing procedures may well be a function of the failure to apply the theories themselves to the methods of measurement. Particularly, it is a failure to apply an analysis of the determinants of behavior in general to the specific test taking behavior of the subjects (Ss). The gap itself may be described as having three aspects. The first of these relates to the question of the constructs used in the theory and the constructs which the tests were developed to measure. In many instances rather than devising tests which measure specific theoretical constructs which are carefully denned and for which the test behavior can be understood as a logical referent, the descriptive constructs used to classify test response do not logically relate to the new theoretical constructs but are bent or twisted to measure the new variables. That is, test constructs which were used to classify test responses developed earlier are translated to be measures of the new variables. Examples of this are use of Rorschach variables such as color, movement, and shading which arose from imagery-type theory, to measure such constructs as ego strength, rigidity, and tolerance for ambiguity.