Rotter Internal-External Scale in an African Context: Problems of Construct and Equivalence

Abstract

It is argued, in relation to the cross-cultural equivalence of personality measures, that of the three phases of test development namely construction, establishment of reliability and of validity, the second tends to receive too little attention. Accordingly, Rotter I-E Scale data from black and white students in Zambia and Rhodesia were compared with each other and with data collected in other countries. It is concluded that the Rotter I-E Scale is not a reliable instrument even for American groups; that responses may reflect manifest item content more than underlying cognitions about control; and that while locus-of-control seems a valuable and perhaps valid construct for cross-cultural personality and motivation research, we need better scales to measure reliability.