Comments on “Construct Validity”
I have no idea if anyone will find this interesting. When I read it today, 50 years after it was written, I find it terribly academic. Whether it says anything about the nature of construct validity that is of interest to the present work of psychologists, I don’t know.
I spent two years, working mostly on my own, trying to figure out how to measure attitudes in a way that was scientifically respectable, yet would pay attention to the way in which people use reason to adopt some of their policy stands. My B.A. was in psychology, but I was so dismayed by behaviorism that I decided to go into graduate school in political science. I discovered that political science, at least at the University of Minnesota, was trying to become behaviorist, without any sophistication of why this was important. It appeared that the methodological discussions in psychology about why and how to measure human behavior were much more rewarding than anything I could find in political science.
By 1965, I felt that most of my political science professor had give up on any hope that I was going anywhere in graduate school. The interest in my work shown by Paul Meehl was essential in giving me the confidence that I should spend time examining fundamental concepts. From this paper, I turned to a study of meta-ethics, the results of which can be seen in Section 1, #2, below. I stayed in contact with Meehl for several years and he encouraged me in my attempts to explore fundamental academic questions.