Statistical Analysis of Experimental Data
While an increasing number of observational studies in modern political science use quite sophisticated statistical methods, experimental studies often continue to apply rather simple statistical instruments like t-tests or analysis of variance (ANOVA). At first sight this is surprising if one considers that many modern statistical methods had been developed and invented for the analysis of data generated in random experiments (see, for example, Fisher, 1935). It is, however, less surprising if one considers that more sophisticated statistical methods were developed much later in order to cope with specific data problems in observational studies. Looking from the perspective of the random experimentalist, the most serious statistical challenges in observational data arise from treatment imbalance and from the violation of the assumption that the independent variables are distributed independently and identically at random.
KeywordsCommittee Member Ideal Point Test Person Condorcet Winner Subject Effect
Unable to display preview. Download preview PDF.
- Bräuninger, Thomas, Susumu Shikano and Michael Stoffel (2008) ‘An Experimental Study of Decision Costs in Majority Rule Voting’, prepared for the Workshop ‘Experimental Political Science’ at Hanse-Wissenschaftskolleg, Delmenhorst.Google Scholar
- Fisher, Ronald A. (1935) The Design of Experiments (Edinburgh: Oliver & Boyd).Google Scholar
- Greiner, Ben (2003) ‘An Online Recruitment System for Economic Experiments’ in Kurt Kremer, Volker Macho (eds.) Forschung und wissenschaftliches Rechnen, GWDG Bericht 63, Göttingen: Ges. für Wiss. Datenverarbeitung, 79–93.Google Scholar
- Herzberg, Roberta and Rick Wilson (1991) ‘Costly Agendas and Spatial Voting Games: Theory and Experiments on Agenda Access Costs’, in Thomas Palfrey (ed.) Experimentation in Political Science (Ann Arbor: University of Michigan Press).Google Scholar
- Rutherford, Andrew (2001) Introducing ANOVA and ANCOVA: A GLM Approach (London: Sage).Google Scholar