Lijphart, Arend. "Comparative Politics and the Comparative Method". The American Political Science Review, Vol.65, No.3 (1971): 682-693.

Lijphart, Arend. "Comparative Politics and the Comparative Method". The American Political Science Review, Vol.65, No.3 (1971): 682-693.

• The main problems facing the comparative method of political science, and making it less accurate that the experimental or statistical methods of science, is that it has only a small number of cases to look at a large number of variables. This limits its usefulness in discovering causation (685-686).
• The solution to this downside is either to attempt an replace it with statistical analysis where possible, to decrease the number of relevant variables studied, or to expand the number of cases by broadening geographic scope or looking at historical precedents (686-687).
• The concept of statistics as determining a method of difference, as developed by John Stuart Mill, was meant to summarize work in the physical sciences, and both John Mill and Emile Durkheim strongly disagreed with its application to the social sciences, arguing that no two societies were similar enough that a single factor of difference could be identified (688).
• Social scientists have, obviously, disagreed with this opinion, esp. Dr. Giovanni Sartori, who dismissed John Mill and Emile Durkheim as perfectionist (688).
• Area studies seemed to provide a solution to this problem, with scholars like Drs. Gunnar Heckscher, Roy Macridis, and Richard Cox positing that regions share enough characteristics to allow for useful comparative studies of geographic areas. However, Dr. Dankwart Rustow asserts again that regional proximity does not imply cultural similarity and that regional comparative politics is just as flawed as other areas of comparative politics (688-689).
• A number of political scientists have pointed out the importance and relevance of looking at sub-national divisions as cases in comparative politics, as these smaller areas will often have more in common that nations as a whole. A classic example being a comparison of Manitoba and North Dakota instead of Canada and the USA (689-690).