As chair of the social psychology section of the American Sociological Association, I am planning the section’s portion of ASA’s annual meeting. The general title I’m using for the section’s sessions is “Social Structure in Sociological Social Psychology: A Distinctive Concern.” The implied distinction is between sociological social psychology and psychological social psychology. I hope this distinction will become clearer as you read on.
Although I was trained as a psychological social psychologist, and my degree is from a department of psychology, through a combination of accident and intellectual proclivity a substantial proportion of my professional career has taken place in the area of sociological social psychology.
During my graduate studies at New York University, my mentor, Marie Jahoda, suggested that I attend the graduate courses in sociological theory given by Robert Merton. These courses resonated with my own not-quite-articulated thoughts about the importance of social structure, and they had a deep effect on me. And partially – but not wholly – by accident, my entire post-doctoral career has been spent in a cross-disciplinary setting: the unit of the National Institute of Mental Health intramural program dedicated to socioenvironmental studies. Although the unit is interdisciplinary, the majority of researchers have been sociologists, all of whom have had a notable interest in psychological phenomena.
At least 40 percent of my papers are in sociological journals, much of this research having been carried out in collaboration with sociologists. (As it turns out, most of my publications in psychological journals are concerned with the determinants of cognitive functioning in older or schizophrenic individuals, rather than with issues of central concern to psychological social psychology.) Nevertheless, even my sociological social psychology research is heavily influenced by my “inherent” psychological perspective. 
In this sociological research, the dependent variable is almost invariably a psychological concept, such as intellectual functioning, self-directed values, coping, or self-esteem. But, in contrast to the general approach in psychological social psychology, the independent variable (or, as is often the case in my research, one of the variables in a pair whose reciprocal effects are estimated through structural equation modeling) is some environmental variable strongly affected by the individual’s position in the social structure of his or her society. Hence the title of the program I am arranging for the ASA meeting.
The benefits of a cross-disciplinary career are primarily intellectual. For many research questions facing behavioral scientists, interdisciplinary and cross-field approaches can decrease the likelihood of overlooking relevant empirical and theoretical considerations that do not seem important from the vantage point of some particular field. These approaches also increase the likelihood that innovative methodological techniques developed in one field will be applied to the other. This is particularly the case when research questions center on causal connections between different levels of phenomena generally investigated by different disciplines.
Consider as an example Kohn & Schooler (1983). This work brought a psychological perspective to my interdisciplinary collaborative research on the effects of occupational conditions, which meant that measures that historically have been of particular concern to psychologists, such as intellectual functioning and anxiety, were included in the study. On the other hand, a purely psychological approach might have overlooked questions about the linkage between individuals’ occupational conditions and their positions in the social structures of their societies.
Cross-disciplinary differences can affect both methodological approach and publication site. An interesting example of this is seen in our unit’s ongoing long-term research project on the psychological effects of occupational conditions (cf. Kohn & Schooler, 1969, 1978; Schooler, Mulatu, & Oates, 1999). When this research began, circa 1962, any reasonably trained psychological social psychologist was well acquainted with such parametric, experimentally-oriented statistics as analyses of variance and covariance, and had at least an introductory knowledge of multiple regression analyses. Sociologists tended to be much more focused on categorical, non-parametric statistics.
This difference reflected, and affected, not only the modal analyses performed by those in the two disciplines, but also the types of data collected and, perhaps most importantly, the very nature of the questions asked. In personal terms, it meant that I was cockily convinced that I could, at least on statistical issues, run rings around most sociologists.
Starting in the mid 1960s, however, a near-revolution took place in sociological statistics. With considerable influence from Herbert Simon, as he passed through sociology on his way from economics to psychology, sociology progressively became more open to applying various regression-based techniques to the problem of estimating causal effects from non-experimental data.
An early example of this openness was the acceptance of path analysis (originally developed by the geneticist Sewall Wright in the 1920s) as a way of elucidating causal connections. This acceptance set the stage for an even more dramatic change – the discipline’s particular openness to Karl Joreskog’s development of Structural Equation Modeling in the late 1960s.
Sociology’s openness to SEM was in part due to concerns about the effects of measurement error and the estimation of possible reciprocal effects in regression analyses, both of which are addressed by SEM. Consequently, substantive papers using SEM began appearing in the sociological literature in the mid-1970s and became unexceptionable by the mid-1980s.
Although SEM analyses have become generally accepted in many areas of psychology (particularly developmental), and now occasionally even appear in the Journal of Experimental Psychology: General, psychology was at least a decade and a half behind sociology in the acceptance of SEM. The chance that a substantive paper based on SEM would be published in a psychological journal before the late 1980s was almost nil.
The issue of where one publishes when doing cross-disciplinary research is not a trivial one. This is closely tied to the issue of what research literature one tends to read and cite. Given the limits of time, it is quite hard to keep up with the research literature of even adjacent fields. (Nevertheless, I find myself constantly bemused by the neglect or hoariness of the citations, more so in psychological than sociological social psychology, of relevant research in the other discipline.)
Reading and citing the literature from the multiple disciplines involved in one’s cross-disciplinary research is actually just one aspect of the broader problem of how those doing such research spread their allegiance among the disciplines involved: What literature do you cite? What meetings do you attend? In what arena do you compete professionally? What reference group do you use to judge your success?
Equally difficult are the “guild control” issues: For example, where do you stand when your disciplines come into conflict over whether a sociological social psychologist can teach a course with psychology in the title? Furthermore, given the disciplinary lines along which most academic departments are structured, those crossing the lines are often evaluated as dilettantes rather than as intellectually broad. Also, as famous examples at Michigan, Harvard and Columbia illustrate, attempts to set up cross-disciplinary departments including sociological and psychological social psychology often founder, at least in part, because of such guild and intellectual problems.
Despite all of these difficulties, it is intellectually rewarding to lead a professional research life that bridges disciplinary boundaries. Doing so, however, without actually becoming a dilettante or ending up in a professional wasteland between established disciplinary borders, requires the academic equivalent of a balancing act. Such balancing is necessary to remain intellectually identified with the relevant disciplines and to ensure that one’s research is appropriately grounded in their respective theories and methodologies.
Kohn, M.L. & Schooler, C. (1969). Class, occupation, and orientation. American Sociological Review, 34, 659-678.
Kohn, M.L. & Schooler, C. (1978). The reciprocal effects of the substantive complexity of work and intellectual flexibility: A longitudinal assessment. American Journal of Sociology, 84, 24-52.
Kohn, M.L. & Schooler, C. (1983). Work and personality: An inquiry into the impact of social stratification. Norwood, NJ: Ablex Publishing Co.
Schooler, C., Mulatu, M.S. & Oates, G. (1999). The continuing effects of substantively complex work on the intellectual functioning of older workers. Psychology and Aging, 14, 483-506.
1. Although it would be impossible to isolate the effects of heredity, environment, and their interaction, as well as of selective mating, I should note that my uncle Isidore Chein was a psychologist. My wife Nina Schooler is a psychologist, as are both of my sons, Jonathan and Lael, and their wives, Tonya Engstler Schooler and Julia Kushner Schooler. My only sib, Miriam Schooler Bendiksen, is a psychologist, as is her husband, Ivan Bendiksen, and one of their two sons, Daniel Bendiksen. I also have about a half dozen cousins of different degrees of consanguinity who are psychologists.
Whether or not it is “inherent”, my interest in understanding how society affects psychological functioning is certainly long-standing. I have a very distinct recollection of having a discussion with my best friend in fifth grade about what we were going to do when we grew up. I was going to be something like a “cultural psychologist/anthropologist”, who would study how peoples’ societies’ affect how they think; he was going to be a nuclear physicist. How much psychology may have gained by my being right is a moot point. How much it gained by Saul Sternberg’s being wrong is not.
Leave a comment below and continue the conversation.