Methodology Matters

Today, the Faculty will begin to take on the central and difficult question of what students should know to graduate from Harvard. The Task Force on General Education has produced a serious and thoughtful answer to this question. It has proposed that the College train students for citizenship in a global society and, to that end, require students to take courses in ten diverse areas from reason and faith to analytical reasoning. I fear, however, that the proposal goes too far in rejecting the Core Curriculum’s “approaches to knowledge” in favor of teaching knowledge itself. Methodology, particularly the scientific approach to human society, should play a prominent role in general education.

Like any self-involved faculty member, I could argue that the proposed program gives too little attention to my own field of economics while spending too much time on other less important disciplines. In extreme fits of economo-centrism, I can certainly convince myself that reading, writing, and breathing are pretty irrelevant relative to understanding the laws of supply and demand. I am not, however, writing this column to argue that my discipline deserves more recognition in general education. Indeed, I do not profess to know how much space in the general education curriculum should be allocated to any field, and I do not know what subjects should be focused on in other fields. I do, however, know that with regards to economics, the report focuses too much on social science topics but too little on social science methodology.

Over the past 230 years, the definition of economics has changed from topic to method. While economics was originally defined by the task of running a government and then the understanding of formal markets, the field has come to define itself by a scientific approach to human society. This method starts with formal models of decision-making agents and exposes these models to statistical tests. In that sense, the methodological heart of economics does not distinguish it from any other field that believes in the value of applying the scientific method to mankind.

Since the scientific approach to humanity is practiced by many disciplines, it is particularly appropriate for general education. Many sociologists, psychologists, political scientists, and anthropologists believe as firmly as any economist in an approach that combines rigorous hypotheses and empirical tests. Increasingly, social scientists are learning from physical scientists who also use formal theory and empirical testing. Economics does not need a special place in the curriculum, but the scientific approach to human society, as practiced by many disciplines, should be central to general education.

Harvard’s system of general education should emphasize methodology over topic because methods are harder to teach and learn than facts. Facts become easier to absorb by one’s self once one has a handle on methods. Harvard students can learn facts about the United Nations Security Council or the Federal Reserve Board from the New York Times or Wikipedia, but they cannot learn the tools to make sense of these institutions so readily. As students learn to think rigorously about society and how to use data to test their thoughts, they acquire a set of tools that can then be used to acquire knowledge in any setting.

The Task Force’s turn against methodology may reflect a dissatisfaction with the Core, which, as I am urging, vaunted methods over facts. The Core’s approach to methodology, however, teaches methods via intensive exploration of a single, sometimes narrow topic. The scientific method’s general applicability, however, can be best taught by applying it in a wide range of settings. The Task Force’s emphasis on broader courses that teach a wide range of facts is a step in the right direction, especially if those facts serve to teach the scientific method.

A thorough general education requirement on the scientific approach to society would require two courses. First, students should take a course that teaches the crafting of rigorous hypotheses. This could be a class on evolutionary theory and human nature, psychology, political theory, or even economics. The key requirement should be a focus on rigorous theory about mankind. I tend to think rigor improves with mathematics, but I am perfectly willing to accept that there are verbal substitutes.

Second, students should take a class on evidence and statistical inference. This could either be pure statistics or empirical tools taught through the lens of a particular topic. Decent citizenship of the world is incompatible with statistical ignorance. A Harvard education must train people to separate compelling evidence from froth. Statisticians do have a comparative advantage in this, but I can readily imagine great core courses taught by Florence Professor of Government Gary King or Ford Professor of the Social Sciences Robert J. Sampson teaching students empirical methods with a focus on politics or sociology. The analytical reasoning component of the proposed system includes such courses but comes up short of mandating them. While other methods of analytical reasoning like logic are important, a statistics-oriented course should be required.

The scientific method should not be an afterthought at Harvard and it should not be confined to the physical sciences. Whether Harvard students are going to be running non-governmental organizations in Africa, hedge funds in Greenwich, or even academic institutions in Cambridge, they will need to analyze situations and process data. In considering a new system of general education, the Faculty should embrace the scientific methodology that will enable students to do this effectively.



Edward L. Glaeser is the Glimp Professor of Economics.