What Social Science Does-and Doesn't-Know

Outline:

In this article, the author primarily examined the development of the methods economists, or social scientists, experiment their theories.

1.       The innate difficulty of social science: why we need controlled experimentation:

i.                     We wouldn’t know which economists were right even after the fact.

ii.                   We have no reliable way to measure counterfactuals-that is, to know what would have happened had we not executed some policy-because so many other factors influence the outcome.

2.       The beginning of scientific experimentation:

i.                     The distinction between empirical observation and experiment: control. Firstly widespread in physics and chemistry.

ii.                   Controlled group experimentation then went into the field of medicine: the famous Scurvy test.

iii.                  Since the number and complexity of potential causes of the outcome of interest, it became difficult even to identify, never mind actually hold constant, all these causes. Thus, the method of randomized field trials (used by social scientists.) appears.

3.       The problem of generalization, the Achilles’ heel of any experiment.

i.                     The even more complexity within social science.

ii.                   The failure of randomized field trials in social science.

iii.                  The success of randomized field trials in business: the concept of replication of field experiments can be pushed much further than most social scientists had imagined.

4.       What do we know form the social-science experiments that we have already conducted?

i.                     Few programs can be shown to work in properly randomized and replicated trials.

ii.                   Within this universe of programs that are far more likely to fail than succeed, programs that try to change people are even more likely to fail than those that try to change incentives.

iii.                  There is no magic: those are programs that do work usually lead to improvements that are quite modest, compared with the size of the problems they are mean to address or the dreams of advocates.

5.       Conclusion:

At the moment, it is certain that we do not have anything remotely approaching a scientific understanding of human society. And the methods of experimental social science are not close to providing one within the foreseeable future. Science may someday allow us to predict human behavior comprehensively and reliably. Until then, we need to keep stumbling forward with trial-and-error learning as best we can.

What’s interesting: The most interesting part I found in this article is about the innate difficulties within social science, which, as the author argued, come from the impossibility of examining whether certain theory work. However, I do realize that there was an era during which economists, or in other words, pre-Adam Smith “economists”, post their own economic and political theories on an extant social entity, hence making significant contribution to human civilization: The ancient Greek. With no doubts, the best way to examine an economic theory is to place it in a real society, and compare the outcome of that theory with that of another country. However, given the impact of this behavior in modern world, this approach is definitely impossible. Hence, the only ways economists organize their own theories, in my opinion, are limited into two categories: Math and History.

Question: Why there’s so many economists arguing about problems that have already been settled in a mathematical graph? Like, the sunk cost, the influence of a tax on a society, and the problem of monopoly?

Annotation: I believe the function of this article will gradually emerge in my future study of economic. It teaches me not only how to think as an economist, but also how to experiment as an economist.