Putting More Science into Political Science
In 1909, the president of the American Political Science Association exhorted his colleagues to eschew experimental designs. Yet by the 1980s the tide had started to turn: Political scientists began folding theoretical approaches from psychology and economics into their research.
“In many scientific disciplines, most breakthroughs and advances come through experiments,” said James Druckman, an IPR political scientist. “While this is not yet the case in political science, experiments are on the rise because they provide very transparent and quantifiable ways to examine causal inferences and empirically grounded theories.”
“Today, many political scientists prefer to use some form of experiments—either field, survey, or laboratory—to delve deeper into a variety of topics, including political behavior, beliefs, opinions, and policy implementations,” Druckman said.
He pointed to how such experiments have shed light on critical questions. For example, how survey wording, i.e., “global warming” vs. “climate change,” affects how voters view climate change. Or how they might upend long-standing views, such as how opinion surveys might not accurately reflect how respondents think about a particular topic.
Political scientists are conducting more experiments than ever before, but their use is still relatively recent compared with other disciplines, and challenges remain, Druckman said. These were some of the driving considerations that led to the Cambridge Handbook of Experimental Political Science, which he co-edited.
For the first time, more than 50 leading political scientists undertook a comprehensive survey of experiments in the discipline. The 2011 volume’s contributors covered a host of topics from the nuts and bolts of designing experiments to the use of experiments in decision making, elite bargaining, and race and identity in politics. These topics were refined through robust paper critiques at an IPR-sponsored conference in 2009 (PDF) held at Northwestern University before taking their place in the book.
“We hope the book will help further develop experimentation in the field,” Druckman continued. “Political scientists might not always agree on the methods, but well-designed studies and the ‘stubborn facts’—or reliable information—that they produce can provide transformative findings that everyone can agree on.”
Druckman, J., and A. Lupia. 2012. Experimenting with politics. Science 335(6073): 1177-79.
Cambridge Handbook of Experimental Political Science. 2011. Cambridge University Press (576 pp.), ed. J. Druckman, D. Green, J. Kusklinski, and A. Lupia.