OBJECTIVE EVALUATION OF PERSONALITY TESTS

The final judgment that the scientist must perforce arrive at is that no objective evaluation can now be made of personality tests because they have not yet attained the status of tests yielding specifically defined scores. As techniques for aiding in clinical judgment, they have proved their worth. As independent tests they are found wanting.

The reason for their failing is that no two responses are ever sufficiently identical to be classified as equivalent. In order to evaluate responses some abstract dimensions must be provided. But insufficient theoretical frameworks have thus far been provided for abstracting the concrete response. The present scoring systems are too concrete, too close to the original response, and not sufficiently abstracted to yield a measurable dimension of behavior. When the personality tests yield abstract scorable dimensions of the variety that the physicist finds when he abstracts from a given concrete object its weight, temperature, volume, etc., we shall be able to make more headway. To those who say that no such measures will ever be possible, let me point out this allegory: The ancient Babylonians and Egyptians in the dawn of history had no thermometers but they did evaluate temperature concretely in terms of the heat of fire, and the heat of summer and cold of winter. They had a scale ranging from the heat of fire, through the hottest day of summer to the coldest day of winter. This scale obtained until the thermometer was born. History does not record the reaction of the populace to the first thermometer, but I can conjure for you the following complaints about it:"You can't measure such an imponderable characteristic as temperature—it is too global, too diffuse, too all-encompassing to yield its secret to that mercury stick. Besides, yesterday I perspired freely, today is relatively cool, but that thermometer registers the same reading." Had the thermometer makers discarded their instrument at that point, science would have suffered a severe loss. Only by persevering with the thermometer did we discover the other factors that go into the global concept of subjective warmth.

For this reason, scientists are providing rating scales for catching the essence of the concrete responses and for classifying these essences along proper dimensions. The virtue of these scales is that they reveal not only what types of responsiveness the patient exhibited, but also which type he failed to exhibit. Another solution that scientists are seeking is to reduce the personality techniques to simpler structure, and to utilize more specific direction for obtaining measurable performance.

Whether such approaches will eventually elevate projective techniques to the status of personality tests is still debatable. Some evidence, however, has been provided to show that scaling of variables and simplifications of test material lead to more precisely definable relationships with clinically observable variables. Whether these tests, even in their higher level of development, can replace clinical judgment, interviews, or examination is highly doubtful. But that they can be of greater helpfulness and of greater dependability is much to be hoped.