I started life as an experimental social psychologist but migrated to applied social psychology research. Each time I have been involved in a major applied project I have learned things that have helped in subsequent projects. Most of the time the lesson has been about research design (e.g., you must know how the study should be done before you deal with the reality of how it has to be done). Examples of other lessons include using appropriate research technology, and the importance of program planner awareness of psychological research. In this paper I describe some of the major studies in which I was the investigator and the lesson (s) I learned from each. I also touch on the relevance of the scientist/practitioner model for the applied researcher.
Phil Zimbardo, in his introduction to Pious (1993) wrote:
In recent years, the field of social psychology has emerged as central in psychology's quest to understand human thought, feeling, and behavior. Thus, we see the inclusion-by-hyphenation of social psychology across diverse fields of psychology, such as social-cognitive, social-developmental, social-learning, and social-personality, to name but a few recent amalgamations.
In their role as the last generalists in psychology, nothing of individual and societal concern is alien to social psychological investigation - from psychophysiology to peace psychology, from students' attributions for failure to preventive education for AIDS. . . . Indeed since the days when George Miller, former president of the APA called upon psychologists to "give psychology back to the people," social psychologist have been at the forefront, (p. xi)
I would add to Phil's comments that social psychologists are also the best general methodologists in psychology. One of the things we have to "give hack to the people" is a method of learning about the world, studying problems, and suggesting solutions. The core of what we do is to try to ascertain causal relationships beyond the attributions made as part of the normal human attribution process.
Aria Day (2003), in her column in The Canadian Industrial and Organizational Psychologist, wrote about the scientist/practitioner model in I/O. Until I had read her column I had thought that the s/p model was relevant to clinical practitioners, not to social psychologists. After reading Aria's column, I reali/cd that the model was relevant and that I am a s/p. The difference between my work and that of a clinical practitioner is that in the work I do there is usually a "proper" or ideal way to conduct the studies, whereas in clinical work there is not always a "proper" treatment. However, in applied social research the ideal study is not often possible so we must work as best we can and be aware of the threats to our conclusions.
I am not an artist but I have always understood that in order to be able to draw like Picasso, you first have to know the ways to accurately represent the physical world in your pictures. Only then can you deviate to indicate the way the world "really looks" to you.
Research methods are like art in that you have to know how it should be done to determine causality before you can deviate from the straight and narrow. For instance, you must know what threats you are introducing to your ability to ascertain causality (in other words, the validity of your results and conclusions) when you do not randomly assign people to experimental conditions.
Luckily for me, I started as an experimental social psychologist, working with judson Mills and Elliot Aronson, both of whom trained me to do proper experimental research. I do not remember when I first came across Campbell and Stanley's little book, Experimental and Quasi-experimental Designs for Research (1963), but it was truly enlightening. It lists all of the threats to internal and external validity - and theirs was probably the first (and only) explanation of "regression toward the mean" that I have ever understood. …