Download full issue
Vantage Point: From the Executive Director
Sally T. Hillsman,
The Importance of Statistics within the Discipline
The following is a version of a letter I recently sent to an Associate Dean for Undergraduate Studies at a research university on why sociology departments should and often do teach the required statistics course.
The Sociology Department at your university has invited me to comment on the reasons why sociology departments at both the undergraduate and graduate levels across the country almost always teach BA majors, as well as MA and PhD candidates, the required statistics courses for their degrees. I offer my observations about this as Executive Officer of the American Sociological Association (ASA), having conferred with colleagues including the three other PhD sociologists at ASA who have examined this issue from their various perspectives as Directors of Academic and Professional Affairs, Minority Affairs, and Research on the Discipline and Profession.
Sociologists teach statistics, not some special variant of statistics, but statistics. Sociologists are trained in statistics and many have extremely rigorous mathematical training. (The National Science Foundation [NSF] views statistics as a part of Mathematics, not as an independent discipline, a view with which I know many statisticians disagree but that is important to understanding how the field is taught.) Sociologists, particularly those with rigorous mathematics backgrounds, also develop statistical methods.
In the Beginning
While in the early post-WWII period, sociologists tended to borrow statistical methods (largely from the biomedical sciences), sociologists have since moved on to develop a number of statistical approaches that are now the cornerstone of the biomedical sciences and many other scientific disciplines, including, for example, multi-level modeling and network analysis. Sociologists, however, tend to teach statistics contextually; they do not teach a different statistics but a different context for statistics. Frequently statistics taught in a statistics department is heavily mathematical and abstract, with limited application and hypothesis testing; but even when application and hypothesis testing do occur, they often involve non-social science examples or social science examples that are not meaningfully contextualized within a body of social science research.
This may or may not be the case at your university. But it is the case in many Research One universities and outstanding private colleges and universities, and it is a major reason why sociology departments across the country have tended to teach statistics courses for disciplinary degrees. (I am fairly sure this is also the case with other social science disciplines and for similar reasons.) It not a matter of good or bad statistical methods; it is about intellectual and pedagogical context. Sociology as a scientific discipline teaches us that context is extremely consequential in all aspects of social life and certainly in teaching and learning. In fact, one of our journals is Teaching Sociology, which is the major research journal on teaching and learning in sociology. I myself was taught doctoral-level statistics in the Columbia Sociology Department by the sociologist Paul Lazarsfeld, a founder of modern applied statistics.
Obtaining Professional Jobs
Undergraduate sociology statistics classes, in contrast to most statistics classes taught in statistics or mathematics departments, are often heavily based in applied examples, social science data sets, and hypothesis testing. NSF-funded research on the BA and Beyond—a national panel study of sociology majors after graduation conducted by the ASA Department of Research—has shown that mastery of (and even reasonable familiarity with) applied statistics is highly predictive of students obtaining more professional jobs related to their major after graduation. It appears that learning core statistics in the context of sociological theory, sociological methods, and substantive research findings about the social world is what makes students sufficiently confident of their mastery of these skills that they can present them to others as part of their marketable skillset.
It is empowering to students to fully grasp and to be able to explain the meaningfulness of statistics as they have understood them within an intellectual context—one that they have found stimulating and challenging. When taught statistics within their discipline, students feel comfortable that they can properly apply those statistical principles to problems they face in the workplace, read in the news media, or discuss at the dinner table. (Other data from that same NSF-funded research indicate that students select sociology as a major after their first introductory course in college because the concepts they learn—that are fundamental to our discipline—are experienced as exciting and illuminating, including the concept of sociology as a scientific discipline.)
ASA has been deeply involved in bringing quantitative literacy to undergraduate sociology departments across the country for a number of years, first through a demonstration grant from the Ford Foundation that focused departments on developing curricular changes that explicitly included statistical techniques for data analysis into all sociology classes, not just statistics and methods courses. We also developed, implemented, and evaluated a prototype Advanced Placement course that was designed to be quantitative, and we tested it in four urban (including central city) high schools spread across the country. It was highly successful in both engaging students in disciplinary concepts and improving quantitative learning.
The reason I bring these ASA activities to your attention is that they are central to the issue of teaching statistics to undergraduates. As a discipline we have learned through innovations such as these that sociology as a discipline attracts students, including the subset of students who have anxieties about mathematics and “science.” Within the context of the subject matter they learn to understand and employ statistical concepts and tools as a means of stretching their understanding about content they find interesting and intellectually challenging. I would add that many of the PhD candidates who have been part of the ASA Minority Fellowship Program (MFP) for the last 40 years often reflect these findings when they share with us and their MFP cohort members their personal experiences of becoming professional sociologists. These MFPs are now part of the professoriate at Yale and Johns Hopkins, University of North Carolina and Duke, as well as community colleges and HBCUs across the country. They are research scientists and policymakers in foundations, research organizations, government and industry.
I recognize that there are many parameters to a university’s decision about curriculum. I only hope the above observations add to the context in which you and your colleagues will make the decision regarding the teaching of statistics to undergraduate sociology students. They are what our discipline as a science is learning about itself.
Sally T. Hillsman is the Executive Officer of ASA. She can be reached by email at email@example.com.
Back to Top of Page
Back to Front Page of Footnotes | Table of Contents