Theoretical Expectations and Actual Outcomes of Electoral Systems: How to Measure the Size of the Deviation?

The article criticises current conceptual frameworks focused on the evaluation of the performance of electoral systems. It offers a new tool allowing researchers to measure the size of the deviation of electoral outcomes from theoretical expectations. The index d=log[ N S /( MS ) 1/6 ] is built on the Seat Product (Taagepera, Rein (2007b) Predicting Party Sizes: The Logic of Simple Electoral Systems . New York: Oxford University Press) and captures the deviations of electoral outcomes from predictions solely on the basis of two institutional factors – average district magnitude ( M ) and size of assembly ( S ). The theoretical background of index d is explained, and its reliability is further supported by conventional econometric methods based on empirical data.