The Research Process: Finding the Best Data to Test Your Hypothesis

The Research Process: Finding the Best Data to Test Your Hypothesis

Once researchers develop a hypothesis, they face the sometimes daunting task of finding appropriate data and statistical methods to test this hypothesis.  In some cases, this might involve conducting field work and/or developing appropriate survey instruments, and some faculty and Ph.D. students in our department engage in such field work.  However, other research, including my own, typically involves collecting and analyzing data from publically available datasets or documents.  As an economist interested in international trade issues, for example, I rely on such datasets as the World Bank’s World Development Indicators, the International Monetary Fund’s International Financial Statistics Database, and the United Nation’s Comtrade (an international trade statistics) database.  Researchers more interested in US domestic issues may rely on consumer or industry level surveys published by such agencies as the US Census Bureau or Bureau of Labor Statistics.

Students in the MA in Economics program also rely on publically available data sources to complete their capstone projects.  American University’s library provides easy access to the databases listed above among other data repositories such as the Inter-University Consortium for Political and Social Science Research (ICPSR).  Our faculty and librarians can help you find and download the best data to study your question of interest.  To give you some ideas of what you can accomplish, here are two recent examples of capstone projects completed by MA students in the program.

Impact of Trade Liberalization on Labor Market Outcomes
One student was interested in studying the impact of trade liberalization on labor market outcomes, particularly unemployment rates.  The student used the Ricardian Trade Model to motivate his hypothesis that higher levels of trade liberalization should result in lower rates of unemployment.  He tested his hypothesis using two different econometric models.  The first was a cross-country analysis, for which he collected data primarily from the World Bank’s World Development Indicators.  The second was an analysis of US industries using data from the National Bureau of Economic Research’s Manufacturing Productivity Database merged with trade data from the International Trade Administration’s (ITA) National Trade Database.  Interestingly, the student found only weak evidence in support of his hypothesis, suggesting that trade liberalization may not consistently result in positive employment effects in a country or individual industry.

Dutch Disease
Another student was interested in a phenomena known as the Resource Curse or Dutch Disease in which a geographic area blessed with natural resources such as crude oil actually grow more slowly than other, less endowed economies  In particular, the student hypothesized that areas in the United States that have seen production of natural gas skyrocket in recent years due to new “fracking” technologies will experience lower growth rates in per capita income than similar neighboring counties.  To test this hypothesis, the student gathered county-level data for five states between 2000 and 2013.  Natural gas production data is available at the county level from the U.S. Department of Agriculture, while county-level demographic and economic data is available from federal agencies such as the Bureau of Economic Analysis and U.S. Census Bureau.  Using an instrumental variable approach to control for the potential endogeneity of his explanatory variables, the student concluded that there is no statistical evidence that increased natural gas production has actually increases per capita GDP in the counties in Eastern United States.

Although you may not always be able to find the perfect data to answer your question of choice (very little in life is perfect after all), there is almost always data out there that will at least allow you to explore your area of interest, and the online MA in Economics program at American University will train you to find this data.

About the Author
Kara Reynolds’ research interests include the political economy of trade protection, particularly antidumping protection, and the impact of trade liberalization on workers and firms. She will be teaching ECON-600 Microeconomic Theory. 

To learn more about American University’s online Master of Arts Economics, Applied Economics specialization, request more information or call us toll free at 855-725-7614.