Graduate School at Ohio State


In the fall of 1952, I started my graduate study in clinical psychology at Ohio State University. I believe that I had a great advantage intellectually, socially, and personally by having had the opportunity to continue at Ohio State. I was completely comfortable there, I had many connections to faculty and graduate students that were important to me, and I already “knew the ropes.” I was able to continue with my assistantship with John Hemphill and already had my second publication in the works, something quite unusual in those days.

Moreover, without any question, Ohio State was one of the premier psychology departments in the country at that time. Kelly and Rotter were superstars in clinical psychology, and Paul Mussen was a star in the making. Delos Wickens was widely known in experimental psychology, and Donald R. Meyer, although junior in rank, was a rapidly developing star in physiological psychology. John Hemphill was one of the best known researchers in leadership and a bit later became a leader in educational organization and leadership. Caroll Shartle and Ralph Stogdill were eminent in personnel and organizational psychology (although I am not sure the latter term was in common use at that time). Counseling psychology was led by Francis Robinson, nationally known, and Harold and Pauline Pepinsky came at about that time and added luster to that program.

The graduate program in clinical psychology was not remarkably competitive, or it did not seem all that competitive to me, but it did help enormously to be in the midst of a splendid group of graduate students who were mutually supportive and characterized by high morale. My classmates included Walter Mischel, David Newman, and Walter Katkovsky, all of whom went on to academic careers. Ahead of me, besides Bill Lyle, were a great many graduate students whom I much admired and who served as both models and sources of support. Brendan Maher was a truly remarkable student and was twice important in my career. David Hunt, Harry Schroeder, June Chance, and Al Mahrer were other graduate students influential in one way or another (but positively) in my career as a graduate student. To complete the picture, there was also a group of outstanding graduate students coming along after me who contributed to the exciting intellectual atmosphere, e.g., Douglas Crowne, David Marlow, and Milton Lipetz. It did not surprise me to learn some years later that thirteen graduate programs in clinical psychology were, at one time, headed by graduates of Ohio State.

It did not take me long to learn that I did not much like clinical work. In the first year program at Ohio State, we were put in various settings in which we engaged in various clinical activities, mainly assessment but also involving some interviewing. Those experiences were undoubtedly very good for learning, but the lessons were various. One involved a field trip to a nearby community in which we, first-year graduate students, engaged in “intelligence” testing in a school. The work was not simply an exercise as I learned when I became aware that a teacher or supervisor in the school was greatly interested in the “IQ” score of a kid I had tested because if the score were 69 or less, the school would be permitted to refer the boy for special education; a score of 70 or higher meant that they had to keep him. I cannot remember which way the score fell, but it was very close, and I was bothered a lot by the idea that so much could hinge on a single point on an obviously fallible test administered by a nervous first-year graduate student.

A later experience in that first year entailed a trip to the local state mental hospital, where we were to interview and “test” (Wechsler? Probably some personality instrument?) a patient in the receiving unit, i.e., newly admitted. I will confess that I was never completely comfortable around “mentally disturbed” patients, and my first contact was with a large (probably 6’2, 200 lb), dark-haired man who, I knew, was being confined because of “family violence.” Ummh! So I interviewed him, and his story was that he worked hard to provide for his family (wife and, I think, two kids). And, anticipating “Rabbit, run,” he regularly came home to find his wife and kids watching awful shows on TV and with nothing more facing him than a TV dinner (just then invented) and an evening of stolid indifference. So, on a recent evening, he had lost his head and taken a large hammer and smashed the TV to smithereens while yelling at his family. I was greatly relieved to realize that he was actually pretty harmless but completely at a loss to know what ought to be done about his predicament. Neither my heart nor my head was ever in the clinical setting. Maybe I had no stomach for it either.

I was not a particularly good psychological intern either. I did my internship on the OSU campus at what was then the Ohio Receiving Hospital. Patients were send there for “short term” stays (up to 120 days, and in some instances a bit more!) for diagnostic study and treatment in lieu of direct admission to the state hospital. The experience was outstanding because of the great variety of patients and the generally smart and alert staff. Among other cases I encountered were a case of porphykria, one of silver toxosis, a stark, raving maniac, several genuine psychopaths, some severe OCDs, and any number of depressives, schizophrenics, and other typical mental patients. During my time there—only nine months, that being all that was required at the time—I was the “therapist” for one patient during the course of whose treatment I made a serious error. The patient immediately got better and went home. I was just after that therapist for another patient and did what I thought was a very good, sensitive job. That patient got worse and worse and finally was sent to the state hospital. Those cases ended any illusions I might have had about being a psychotherapist. I did almost no clinical work of any kind after the completion of my internship. I was not cut out for it..

From an academic standpoint t, my doctoral training at Ohio State was excellent. Mostly. I took a very few classes that tried my soul, but I liked graduate school—a lot. I decided to minor in “statistics,” a fairly natural outgrowth of my experience and interests. Robert Wherry was the main statistics instructor, and he was a good teacher and a considerate man. I learned something about factor analysis, two-way ANOVA, and a fair amount about multiple regression, Wherry’s area of special expertise. At least at Ohio State, however, we learned a good bit about a fairly wide array of special, even odd statistics. I am the only person I know who has even heard of “triserial r.” We also learned a good bit about test statistics, including the derivation of Kuder-Richardson reliability formulae from “the grandfather formula.” Not that I remembered that formula, let alone the derivations, but Wherry did start pretty much from first principles. I only wish I had had the math and quantitative background to benefit fully from his instruction.

Statistical calculations were done on hand-cranked (Monroe) or electric (Friden) calculators.

George Kelly contributed in important ways to my intellectual development. First, he was a genuine cognitive psychologist before there were any, at least any of the kind who abound today. At the time, psychology was dominated by Hullian learning theory, and clinical psychology at Ohio State by social learning theory. (Skinnerian thinking was, at that time, largely a curiosity at Ohio State.) Kelly guided me toward the cognition that the ways in which people construe (cognize) events may be more important than the reality of those events. It was about that time, I think, that I encountered the famous statement by W.I. Thomas that “If things are believed to be real, then they are real in their consequences.” That statement antedated Kelly the 1950s by more than two decades, although it was surely not from Kelly that I heard of it.

Kelly had, by the time I came under his influence, fairly well worked out his comprehensive Personal Constructs Theory, and a good bit of his 1955 two-volume book was in draft form. Unfortunately, because of limited capabilities for reproducing print material, copies of Kelly’s drafts were not widely available, and I do not believe that I ever came into possession of a copy of even one chapter. Kelly had a formal system for thinking about personal constructs, the nature of those constructs, how they made a difference in people’s lives, and so on. Strangely, at least to me, he seemed to have rather little interest in how personal constructs came into being, and he was vague about how they might actually affect behavior. It seemed that behavior was pretty much taken for granted.

I was interested in behavior, and there was another influence in the clinical psychology program at Ohio State that was greatly influential in the development of my thinking: social learning theory as expounded by Julian Rotter. Kelly and Rotter could scarcely have been more different as men. Kelly was protestant, a churchman, definitely a Mid-westerner, staid, and rather remote. None of his students ever called him “George,” and he regularly addressed students by their surnames, with an appropriate titular prefix. Even when invited by him at the end of the final doctoral dissertation defense to “Call me George,” it was difficult to do so. Julian Rotter was New York Jewish, a graduate of Brooklyn college, outgoing, informal, and engaging. Most of his students called him “Jules.” I am not sure I ever did. I was too much a small town Midwestern boy, outwardly (and usually inwardly) deferent to authority.

Rotter was a fine classroom teacher, and he had a great deal, also systematized, to say about behavior, about the development of behavior, and about changing behavior. He and Kelly were clearly in competition, but the competition was managed well and never became much of an issue in the clinical program. Mostly they stayed apart, they did not debate, they did not contend for resources or students. Kelly had been Director of Clinical Training until about 1950, when he passed the job along to Rotter. From that time on, although Kelly had been a very strong clinical director, I do not think he ever interfered in any way in the running of the clinical program. Social learning theory was exciting, especially at that time, and Rotter’s students were vigorous, active in their research, much of which was of a laboratory nature and taking place all around me.

Why did I end up as a Kelly student rather than as one of Rotter’s? As I saw it, Rotter’s students formed a sort of cult, an in-group, and I was never comfortable in such situations. I always felt myself a bit of an outsider, and I found it difficult to try to insinuate myself into groups that were, as I saw it, inclined to be judgmental and “elitist.” Several of Rotter’s male students (most graduate students were male in those days) smoked pipes, as he did, and some also wore tweed jackets with leather elbow patches and soft leather shoes with crepe rubber soles, also Rotter’s customs. I saw these adoptions as affectations and further signs of the in-group status of Rotter’s students. Perhaps I was wrong, but that is how it seemed to me. Another reason I ended up with Kelly, however, was that William Lyle, my graduate student mentor/model was one of Kelly’s students, and I was able easily to gravitate into that group.

I never felt very competent during my graduate school days in trying to think in systematic theoretical ways. I am not sure why that was so difficult for me, but social learning theory seemed particularly challenging. Discussions of, let us say, level of aspiration studies would end up with my fellow graduate students spinning out with seeming great ease all sorts of ideas for further studies while I was stuck at trying to figure out what had been done in the first place and why. Kelly’s theory was no easier for me to handle, but it was easier in his research group simply to sit quietly and not reveal one’s mental incapacity. I ended up doing both my Masters thesis and my dissertation on completely atheoretical, i.e., non-deductive problems. I never had any problem spinning out theory after the data were in.

It was not until I came under Don Campbell’s influence at Northwestern University that I began to catch on to the idea of overarching, more comprehensive theoretical constructs and how they might be used in order to achieve conceptual integration. I remember, sort of, talking one day to Don about my dissertation, which was on the perceived similarity between psychotherapists and other figures in patients’ lives, especially other professional persons. Don congratulated me on my insight in discerning that such perceived similarity was a manifestation of stimulus generalization and urged that I publish an article to that effect as soon as possible. After arriving at that insight, previously unrecognized by me, I did write an article and succeeded in getting it published. Don was a great cognitive therapist, or maybe therapist of cognition, able to penetrate murky areas to uncover all sorts of ideas about which one had previously been unaware.

At any rate, I do believe that from that time on, I began to think more in terms of higher level theoretical constructs that could integrate observations (“findings”) of diverse sorts and provide a framework for undertaking further research. I cannot in truth, however, say that I ever became facile at hypothetico-deductive thinking. My own further observation, however, was that a lot of what passed for systematic theoretical thinking in psychology seems to be not much more than noting in the report of any given study some loose end and then devising a further study aimed at tucking in that end. That sort of “theorizing” does not guarantee any vector-like, directional property for a field but only the piling up of not-necessarily related findings, what Gordon Allport, whom I later came to know, called “itty-bitty facts.” I have called that style of research “a random walk through a field of science.”