Department of Political Science, Queens College

How to Research a Political Science Paper
Peter Liberman, Queens College Dept. of Political Science, September 2001

2. What kind of information is most useful for the various kinds of research papers?
All kinds of political science requires research, unless you are a genius spinning out original theoretical deductions from the armchair! The kind of research you need to do depends on the type of question and research project you are pursuing, as well as what kinds of relevant information, evidence, data, etc. are available for you to study.

Theory-proposing projects research requires the least research of all. But it is very difficult to come up with new theoretical insights without first mastering prior theoretical and empirical research on a subject (e.g., through stock-taking). Sometimes theoretical insights develop in the course of empirical research, other times by finding theories used to explain related or analogous phenomena, but not yet to the one at question.

Theory-testing research uses evidence intensively. Political scientists use all sorts of evidence: history, interviews of officials or elites, data on public opinion and any other variable bearing on political behavior, and even laboratory simulations. Researchers can often rely on evidence collected by others (e.g., published histories), contributing by scrutinizing it for different purposes or in different ways than has been done previously. Otherwise researchers have to collect the raw data themselves (e.g., going to the archives and other primary sources, in the case of historical evidence).

Stock-taking projects primarily require finding the best and most complete published research on a particular question.
Historical explanation is like theory testing research, except restricted to a single historical episode. The research required is primarily primary and secondary historical sources.

Policy analysis requires finding the best arguments for a policy, as well as existing theoretical, empirical, and theory-testing research to verify the arguments' factual and theoretical assumptions. A researcher might also engage in his or her own original deductive and empirical work to this end.

Predictive projects requires evidence on current events, and finding the best theoretical and theory-testing research bearing on the phenomena you wish to predict.