Reading: Using Psychological Type Tests
from Tracking the Elusive Human, Vol. II


Can a psychological type test solve the diagnostic problem? There are two ways to look at the question. First of all, a great deal of effort has gone into the construction of these tests and their validation. They condense into usable form extensive experience in typology and make available to us a reservoir of good questions. Therefore, they can be an important aid in making an evaluation of type. The test can be a check on our subjective impressions that can be misled by preconceived ideas and projections. If the test and our personal evaluations radically disagree, it is time to reconsider the matter. The test can help us focus our typological knowledge and experience and bring it to bear on the case at hand. We can go beyond the simple administration and scoring of the test and take up the questions one by one with the client and use them as a point of departure for developing a full-fledged typological interview that will combine objective and subjective elements. The test then becomes an impetus to correlate our subjective evaluations with the test scores and the self-understanding of the test-taker.

But there is a second way of using the test, which is less commendable. In it the test results become automatically the final results. The test becomes the sole way we diagnose type. It is the single tool at our disposal. We don't see. The test sees for us. We simply administer the test and assume that solves the diagnostic problem. This is too much to ask from any type test. The test was originally dependent on personal experience. In the initial construction of the MBTI, for example, observation necessarily had to precede any numerical results. "The initial questions were tested first on a small criterion group of about twenty relatives and friends whose type preferences seemed to the authors to be clearly evident from long acquaintance, and from a twenty-year period of "type watching"." (MBTI Manual, 1985, p. 142) Its questions and answers had to be evaluated against the clinical judgment of what the test-taker's type actually was. The questions are forced to break up into pieces what is actually connected in reality. Extraversion and thinking, for example, in the extraverted thinker, are not two separate factors, but one and the same thing. Then the test has to confront the enormous complexity that we saw in types themselves, and do all this by means of the self-report of the test-takers. The validation of the test from a statistical point of view is not identical with their authentication individual by individual, which is precisely where the results are most important.

The test creators themselves are often keenly aware of how things could be different: questions rephrased, and answers scored in a different manner. Isabel Briggs-Myers, when discussing the potential uses of the MBTI, felt the test is applicable in many areas, and can be put to good use by many kinds of decision- makers, but only, she admonishes: "if they will remember they are dealing with a theory and that the hypothesis the Indicator provides about a given person must always be submitted to their own informed and critical judgment. The Indicator is no substitute for good judgment. Being a self-report instrument, in any given case it could be wrong, no matter how high the scores."

"The safe and proper way to use the Indicator is as a stimulus to the user's insight." (MBTI Manual, 1962, p. 5) Chapter 5 of the new MBTI manual presents a balanced approach to the problem entitled, "Initial Interpretation and Verification "No questions, however accurate, can explain all human complexity. The MBTI results are a first step toward understanding the respondent's true preferences."

We have seen the quandary that A. Platt landed in. Katherine Broadway gave the Gray-Wheelwright and the MBTI to 17 Jungian analysts who had typed themselves before the test. She compared the results without reference to the superior or auxiliary function (which, no doubt, would have made the results more dissimilar). The self-typing and the two tests agreed almost 100% on the evaluation of introversion and extraversion. This agreement falls to around 75% for sensation-intuition and a bit less for thinking and feeling. These results, which might be respectable from a statistical point of view, certainly can give us pause from a practical or clinical point of view. There is no reason to suppose that the analyst's judgments of themselves were completely correct, nor to imagine it was the tests or one of the tests which was completely accurate. Thirteen years later Bradway and Detloff looked at the issue again with 92 participants, this time just using the Gray-Wheelwright test. The results were similar. The classification of the whole type by self agreed with the Gray-Wheelwright results 58% of the time, and this is, again, combining the superior and auxiliary functions. Therefore, it becomes a question of how many times will the self-typing of the analyst and the results of the MBTI and Gray-Wheelwright and Singer-Loomis agree in typing a person?

A comparison between the MBTI and the GrayWheelwright highlights these difficulties in type testing, even when we make the admission that both tests are tapping into the same basic reality. In a study of 159 university students who took both tests, only 21% came out the same type on both instruments. In another study of 98 students the correlation between the different scores of the two tests were "E .68 (p<.01), I .66 (p<.01), S .54 (p<.01), N .47 (p<.01), T .33 (p<.01), and F .23 (p<.05)." (MBTI Manual, 1985, p. 209)

It would be unfair to appear to chastise the tests and leave the impression that diagnosis by Jungian analysts is better, even when it is a question of their own type. Introversion and intuition are outstanding in analysts' self-diagnosis and test-taking, but whether this is an actual fact or reflects their image of what they ought to be remains undecided. Probably it is some of both. The fundamental reason for not relying on the estimate of Jungian analysts is that their education in this area has often been sporatic at best. To return to our analogy between typology and nutrition, when we read Plaut's survey, we see that the Jungian analysts are like medical doctors when it comes to nutrition: some are profoundly immersed in it, others are distressingly ignorant, and in the profession as a whole it has not penetrated that nutrition is an integral and indispensible part of medicine. It may be that an analyst has learned typology through his own analysis, but if half the analysts find no great use for it in practice, there is no guarantee this will happen.

We have come to another dilemma. We can't rely implicitly either on self-reports or test results. There is no magic solution. We may decree that a person's type is whatever the tests say it is, but this decree need not coincide with reality. We have to hold on to the two horns of this dilemma: the clinical viewpoint and attempts to be more objective. Testing and observation by an experienced diagnostician should go together in order to accurately determine type. There are several principles that will aid successful diagnosis.

1. An intimate working knowledge of one's own type. This means not simply an accurate knowledge of what our type is, but the inner implications of this knowledge in terms of how it expresses itself in the myriad of details of our daily behavior, and especially how our inferior function, as well as the undeveloped aspects of the other functions, influence our conduct towards other people. Without a knowledge of our personal mechanisms of projection, how can we avoid them?

2. An extensive knowledge of the practical peculiarities of different types. This is an appreciation of the nuances and qualities that affect the function, not only when we consider whether it is introverted or extraverted, but the qualities it manifests in each of the four positions.

3. The use of a psychological type test in such a way that it becomes a key element in a typological interview that helps objectify our observations. This includes a dialogue with the client about what he considers his type to be. Naturally, the weight given to this self-report will have to vary with the degree that the person is informed about types. Someone who has just been introduced to typology can hardly be expected to render a definitive judgment that will bind himself and everyone else. Many people, however, do show a remarkable degree of insight after they have grasped the basic principles involved.

Type diagnosis becomes a three-way discussion between the person being diagnosed, the typologist and the objective testing. In this way we can limit the deficiencies that exist in each area. Though such a process is more time-consuming, in most situations where knowledge of type is to be the starting point for a process of self-development, even if an initial quick diagnosis is correct, it lacks the practical efficacious certitude that comes from a slower and more thorough process of typological discovery. If I am told I am an extraverted intuitive type, it does me little good as long as this remains a purely verbal definition, and I have generated no self-insight in how these typological principles can help me focus better on the actual course of my life.

Now it is your turn to contribute to this discussion. Send us your questions and comments:

How to contribute to this discussion

Up to What Are Psychological Type Tests Worth?
Back to Tracking the Elusive Human, Vol. 2