Donald T. Campbell, Master of Many Disciplines, Dies at 79

Photograph of Donald T. Campbell

By ROBERT McG. THOMAS Jr.

May 12, 1996, Sunday, Late Edition - Final

Donald T. Campbell, a nimble-minded social scientist who left his mark on half a dozen disciplines and helped revolutionize the fundamental principles of scientific inquiry common to them all, died on Monday in a hospital near his home in Bethlehem, Pa. He was 79. His wife, Barbara Frankel, said the cause was complications from surgery for colon cancer.

Dr. Campbell, who did his major work at Northwestern University, was by training and his Berkeley doctorate a social psychologist, but it was a tribute to his bewildering range as a master methodologist that when he took up his last academic post, at Lehigh University, in 1982 university officials threw up their hands and simply designated him "university professor," with faculty listings in the departments of psychology, sociology and anthropology and the department of education.

They could easily have thrown in biology, the philosophy of science and market research. For a generation, virtually no respectable researcher this side of the chemistry lab has designed or carried out a reputable scientific study without a thorough grounding in what Dr. Campbell called quasi-experimentation, the highly sophisticated statistics-based approach he invented to replicate the effects of the truly randomized scientific studies that are all but impossible in the slippery and unruly world of human interactions.

Whether he was doing what he called veranda research among the tribal peoples of East Africa or studying Head Start programs in the United States, his abiding interest was the very study of knowledge: how it is acquired, recognized, evaluated, refined and passed on -- and sometimes lost.

Although he worked on the most abstruse levels of the most arcane fields, Dr. Campbell, who turned out more than 200 scholarly papers, had an impish streak. One of his most compelling "crackpot papers," as he called them, was titled "The Fish Scale Model of Omniscience," his whimsical term for his dead-serious theory that knowledge is analogous to the overlapping scales on a fish, each representing a distinct field of study.

Curiously, Dr. Campbell, who proposed what he called evolutionary epistemology as a unifying theory of knowledge, had as a major focus throughout his career the study of false knowledge -- the biases and prejudices that poison everything from race relations to academic disciplines where erroneous theories are perpetuated by those with vested interests in them.

Indeed, he made his first mark as a young psychologist by discovering that the very conflicting misconceptions that underlie ethnic, racial, national and other group biases tended to be reflections of those that inform, or rather misinform, the rival prejudices of city dwellers and country people.

Dr. Campbell knew whereof he spoke. A native of Grass Lake, Mich., he was the son of a farmer who became an agricultural extension agent in California. After high school, Dr. Campbell worked for a year on a turkey farm before going off to San Bernadino Junior College, a delay that later helped underscore the family brilliance. When he completed his undergraduate education at Berkeley, he and his younger sister, Fayette, graduated first and second in the class of 1939.

After serving in the Naval Reserve in World War II, he returned to Berkeley to get his doctorate in 1947, then taught at Ohio State and the University of Chicago before moving to Northwestern in 1953.

Within a decade, Dr. Campbell had identified fundamental flaws in the way social scientists were approaching research and was arguing that the sophisticated use of many approaches, each with its own distinct but measurable flaws, was required to design reliable research projects.

It is a reflection of the complexity of the subject that the paper he wrote with Donald W. Fiske to present his thesis was titled "Convergent and Discriminant Validation by the Multitrait-Multimethod Matrix." It is a measure of its influence that archivists have called the thesis the most frequently cited paper in social science.

Yet Dr. Campbell was just getting warmed up. In a 1973 collaboration with a statistician, Julian C. Stanley, he published "Experimentation and Quasi Experimental Designs for Research," his 79-page prescription for replicating randomized experiments. Dr. Campbell and his colleague at Northwestern, Thomas D. Cook, deepened and expanded the thesis in 1979 with "'Quasi-Experimentation: Design and Analysis Issues for Field Settings." It has become the research bible, particularly in evaluating social intervention programs like Head Start, but the original refuses to be displaced: More than 300,000 copies of the Campbell-Stanley book have been sold, a formidable figure for such a highly technical work.

When Dr. Campbell came up with the notion of evolutionary epistemology, his view that the development of human knowledge closely parallels that of the species, it seemed he might be on the verge of creating an all-embracing academic discipline of his very own.

But Dr. Campbell would have none of it. He had, after all, studied how academic departments often stifle the advance of knowledge. Disciples were not for him. The man who helped revolutionize social science preferred to trust the future to the academic rebels of tomorrow.

In addition to his wife, a retired professor of anthropology at Lehigh, he is survived by two sons from a previous marriage, Martin, of St. Croix, V.I., and Thomas, of Mudelein, Mich.; a sister, Louise Silver of Salmon Arm, Canada, and two grandchildren.

NAME: Donald T. Campbell

SECTION: Section 1; Page 37; Column 1; National Desk