This function calculates the effective number of parties with incomplete data, as introduced by Taagepera (1997). It can be used to estimate the effective number of groups in a population.
Population vector, the residual category (‘other’) needs to be last
This function calculates the number of effective parties if data are incomplete, i.e. if there is a residual category (‘other’). This function can be used to enumerate the effective number of parties, but also the effective number of groups in a population or representative body.
Note that the residual category is assumed to be the last item in the population vector. The best estimate (
Neff) may be incorrect if certain small parties or groups are listed for historical (or other) importance. In this case, the error range would be underestimated.
Returns a list:
Neff gives the best estimate (in the absence of other information),
simple does not square the residual category,
low treats the residual category a group on its own,
least assumes the groups in the residual category to be at least as big as the smallest group given, and
high excludes the residual category.
Taagepera, R. (1997) Effective Number of Parties for Incoplete Data. Electoral Studies 16(2): 145-51.