A field experiment applies the scientific method to experimentally examine an intervention in the real world rather than in the laboratory. Field experiments generally randomize subjects (or other sampling units) into treatment and control groups and compare outcomes between these groups. Clinical trials of pharmaceuticals are one example of field experiments. Economists have used field experiments to analyze discrimination, health care programs, and education programs. Social psychologists generally avoid field experiments because of the context dependence of experimental outcomes.

Contents

1 History and Philosophy 2 Applications 3 Methodology 3.1 Compared with Laboratory Experiments 3.2 Compared with Natural Experiments 3.3 Compared with Non-Experimental Field Data 4 Caveats 5 Prospects 6 Sources

History and Philosophy

Michael Kremer, a development economist and major practitioner of field experiments, argues that an analogy should be drawn between government's role in overseeing pharmaceuticals and overseeing social programs. More specifically, randomized trials of new drugs are required before the FDA approves them for sale. Similarly, perhaps randomized trials of candidate social programs should be required before such programs are implemented. In both cases, the randomized trials allow careful measurement of the innovation's actual effect before the innovation is rolled-out to the whole population.

Applications

Methodology

Compared with Laboratory Experiments

Compared with Natural Experiments

Compared with Non-Experimental Field Data

Caveats

• Fairness of randomization
• Contamination of the randomization
• General equilibrium and "scaling-up"

Prospects

Sources

• Michael Kremer's homepage (http://post.economics.harvard.edu/faculty/kremer/papers.html)