Reflections on incorporating quantitative training in undergraduate social science programs

On July 1, 2021, I will conclude my term as Program Director for the Quantitative Social Analysis (QSA) undergraduate program that I helped establish here at HKUST after I arrived in 2013. Building on a plan initiated by my colleagues Raymond Wong and Xiaogang Wu, and working with colleagues in Social Science, Math, and other campus units, after I arrived at HKUST we developed and launched an undergraduate major that combined a comprehensive training in social science theory and evidence, including topical courses in sociology, political science, and economics, with rigorous methodological training in the quantitative analysis of social data, with the goal of preparing students for work or for postgraduate studies. It concludes with a Capstone Project where they showcase what they have learned. The requirements are probably more intense than any undergraduate social science program anywhere. Notably, they include three semesters of calculus and a statistics course with a more formal emphasis offered by the mathematics department, along with more hands-on courses in analysis of social data in R here in the Division of Social Science. The training goals are generally similar to those of the Q-Step programs launched at a number of institutions in the UK, and which we only found out about after we were well into the preparations for our own program.

Quantitative Social Analysis had its first intake in fall 2017, and those students are now graduating. Watching the students grow over the last few years has been very gratifying. They had a challenging time, especially in their first two years, but it is paying off.Even though they came in as secondary graduates without prior experience with analyzing data, they have just completed Capstone Projects in the conducted analyses in R using a variety of methods, and addressing a variety of topics. For example, one student used ACS data to study the sources of income differences according to country of origin for recent immigrants from East Asia. Other students used survey data from ICPSR to examine topics including relationships between elderly and adult children in rural China, gender differences in consumer spending in the US from the 1970s to the present. Still others used Hong Kong Census data to examine associations between education and income in HK.

I am very happy with the outcomes so far. While they were still in the program, many students had internships where they applied their skills, or joined faculty research projects. Four graduates are currently enrolled or accepted for this fall at postgraduate programs at Chicago, Columbia, and Washington University St. Louis, and HKU. Other graduates have accepted offers from employers including a major investment bank, two major e-commerce firms, a major market research firm, two local universities (as research assistants), and some local startups. Other graduates are still interviewing.

It has been an interesting six years, including the four years since we had our first intake, and the two years before that when we were designing the program. Different to undergraduate programs in the US, programs here accept applicants directly, so I was involved in all aspects, including recruitment, admissions, and so forth. I learned a lot about the JUPAS and DSE here in HK, the different application profiles for non-local applicants coming through the gaokao, IB and other systems. We tweaked the program based on feedback from students, especially ones in the first intake. We also had an external review last fall that was very positive, but also offered many concrete suggestions for improvement. Most gratifying to me was that when we conducted an anonymous online survey of students in preparation for the review, and when the review committee met with students without any of us present, the feedback was positive, and students provided concrete, constructive suggestions for improvement.

There is a lot that I am proud of with the program, but for me one of the most important aspects, which we have in common with all undergraduate programs at the UGC universities in Hong Kong, is that we are an engine of upward social mobility. I don’t have the most recent numbers, but as of only a few years ago, close to half of undergraduates at institutions in HK were first-generation college students. At all the universities in Hong Kong, many are from families of limited means, and the stories that students have told me about their circumstances are often harrowing. For these students, we have provided an education that is not only broadening, exposing them to a wide range of ideas and evidence in social science, but practical, in the sense of providing skills that will make them attractive to employers or postgraduate programs.

I think QSA fulfills what I believe is the original mission of a liberal arts education: a broad education that not only teaches students to think and widens their intellectual horizons, but also prepares them for further studies or work. I hope that our program could be a model for other social science undergraduate programs that are looking for ways to remain relevant to contemporary needs and attractive to students. I also think this points a way forward for bridging a polarized divide between advocates of a liberal arts education who seem to define it in opposition to relevance, practicality, or market demand, and are contemptuous of the notion that students might be required to learn any maths or science or indeed any marketable skills, and advocates of an entirely practical education focused on STEM and other skills demanded by employers, who sometimes sound like they want to turn universities into high-end vocational training schools.

There are problems with both of these extremes. A liberal arts education that consists exclusively of arts, humanities, and selected topics in social science is producing graduates who not only may find that their skills are not in as much demand as those of students in other programs, but are also not equipped to participate in ongoing debates about the most important issues of our day, because they don’t have any foundational training in mathematics, science, or statistics that would equip them to assess the evidence presented in those debates. Conversely, a university education that is purely STEM or business end up being high-end vocational training, in which students take almost no courses in any subject outside their major, leaves them unequipped to think deeply and critically about complex issues and open-ended problems, and also doesn’t give them the soft skills they need to advance past their first job or succeed in graduate school.

The reality is that we can broaden students, developing their ability to think independently and critically about complex and open-ended problems through coursework in the arts, humanities and social sciences, and at the same time give them foundational training in maths, computer science, and other topics that not only make them appealing to employers and graduate programs, but also equip them to participate in debates about contemporary issues by evaluating quantitative evidence or producing their own analyses. The design of Quantitative Social Analysis reflected my desire to restore what I thought was the original intent of a liberal arts education, which was to produce citizens with a range of skills that allowed them to think deeply about the world around them and participate in ongoing debates while at the same time making them productive.

The other topic that this experience has led me to reflect on is the need for training in ‘data science’ or ‘big data’ or ‘machine learning’ to be linked to training in the substance of the areas where it is to be employed. I am wary of the idea of presenting ‘data science’ and related topics as an amorphous free-floating set of methods of general applicability, which once learned can be used at will to study any topic from light bulbs to frogs to cycles to people, without a deeper understanding of their context. With QSA, as much as possible we have embedded methodological training in consideration of specific issues that arise in considering topics related to society, politics, and the economy.

Below I have copied a letter that I sent to all the QSA students in the program a few days ago.