Statistics in and for the Social Sciences
By Katherine Sinyavin
Now in its second year, the Division’s Summer Institute in Social Research Methods (SISRM) offers undergraduate and graduate students opportunities to expand their social science research skills. This year, one of the five courses offered is Introductory Statistical Methods and Applications for the Social Sciences, taught by Yanyan Sheng, Senior Lecturer in the Committee on Quantitative Methods in Social, Behavioral, and Health Sciences.
“This class prepares students to answer research questions that require the use of quantitative methods. If they have questions in sociology or in political science or in psychology or in any social science, they will be able to identify variables and determine what kind of analysis to use. In this class I also discuss how to carry out analysis and report the results. By the end of the third week, students will be able to set up a research question, specify hypotheses, and carry out the analysis by themselves,” Sheng says.
The overarching objective of this class, Sheng emphasizes, is for students to gain “a whole picture of statistical methods and know which to use for specific situations.”
Sheng gives an example to demonstrate the role of statistics in the social sciences. If one wishes to test the effect of a certain change in methods of education one has read theories about, this class teaches the skills to “try to answer the question of whether there is a difference between the two groups” and whether that difference is attributable to the experimental design or random fluctuations by performing analysis on, for instance, the participants’ test scores.
Chloe Roske, a rising third year in the College, took this class in order to complete the Statistics requirement for her Psychology major. Following the course, and as part of her SISRM experience, she will do research with Micere Keels, Associate Professor in the Department of Comparative Human Development, on trauma responsive educational practices. Keels has constructed a course that “focuses on helping educators navigate the pandemic and the effects it has on their students” Roske explains. Roske’s role in the project involves qualitatively coding the responses educators give to survey questions asked throughout the course.
For Roske, the most difficult part of the class was applying the correct statistical method to a given scenario. She appreciated that the problems involved real-world statistics that she also recognized from her own experiences in psychology. Many of the problems were based on experiments such as the fairness of a dice roll or a study of verbal and reading scores and how these differed between various groups of test takers. Roske found it rewarding to understand how to make these applications.
In addition to developing research skills in the social sciences, this class also offers a foundation in statistics that will allow students to take more advanced statistics classes. While this class teaches single and bivariate statistics, they can later pursue classes concerned with, for instance, multivariable statistics or causal inference analysis.
Roske plans on pursuing a PhD in Psychology, and she sees that this class will be useful in helping her understand the statistics in psychology papers she will read in future classes.
Labs figure prominently in the class, during which students learn how to use SPSS, a statistical software that carries out various aspects of analysis using quantitative methods. Students learn how to input data, how to run different analyses, how to interpret the output tables, and how to write up the results.
“It is challenging to demonstrate SPSS over Zoom and is not the same as teaching face-to-face, but it is still doable. And a nice thing about it is that I can record all my sessions and make the recording available for all the students. So those students who may be missing some of the steps can always go back and review all the procedures,” says Sheng.
Given the remote nature of the Summer Quarter, Sheng has found that one of the biggest challenges is the lack of student interaction and collaboration outside of class. During the spring quarter, Sheng encouraged students to form virtual study groups, and this quarter she decided to assign students into groups to meet after class to share ideas and work on homework together, with one student from each group volunteering to schedule meetings and host Zoom sessions.
Roske explained that while her class was small, there were students from many different backgrounds, including a graduate student, a student from UChicago’s Lab school, and a rising first year. The study groups outside of class allowed her to interact with these students, which she really enjoyed.
Sheng, too, emphasized that one of her favorite aspects about teaching this class is that she teaches a “wider range of students who have different interests and backgrounds” but who are all learning statistical methods that are applicable across many different fields of study.