ABSTRACT
Validity studies of polygraph examinations are basically divided into two categories. Either they are based on real life situations and examinations or on some sort of laboratory experimental setting. The 84 records refer to 42 cases with a pair of examinations per case. The correlation between the two scores over the 84 records which indicates the reliability was found to be 0.85. Of the 84 records there were seven records from seven different pairs for which the two scorers gave opposite final scores. The internal validity of the accuracy rate depends heavily on the assumption that the two examinations in each other were conducted independently. Thus, one can raise the question that knowledge of the result of the first examination in each pair might influence the manner in which the second examination is conducted, leading to an articial increase in the probability of obtaining an opposite outcome.
