ABSTRACT
The method of scale construction has been called the scale-discrimination method because it makes use of Thurstone's scaling procedure and retains R. Likert's procedure for evaluating the discriminatory power of the individual items. The scale-discrimination method is essentially a synthesis of the methods of item evaluation of L. L. Thurstone, Likert, and L. Guttman. It can easily be demonstrated that items with comparable Thurstone scale and Q values may differ tremendously in their power to differentiate between those with favorable and those with unfavorable attitudes. The method of equal appearing intervals and the method of summated ratings are similar in that both provide techniques for selecting from an initial large number of items, a set of items which constitutes the measuring instrument. Numerical weights are assigned to the categories of response using the successive integers from 0 to 4, the highest weight being consistently assigned to the category which would indicate the most favorable attitude.
