Package changes from previous ElectDecomp version 0.0.1-1

The package contains three new functions called inequality(), traditional_indexes() and malapportionment_index(), and includes some changes in the output of the distorsion() function. Furthermore, it has been created a plot method.

The new function inequality() computes the inequalities in votes-to-seats conversion, including the conversion ratios (Kedar et al., 2016) for each combination of district and party, from which it can be calculated the Representational Inequality Index (RI) and drawn the representational inequality (Lorenz) curve of the election.

The new function traditional_indexes() provides as output the values of a list of indexes proposed in the literature as deviations from proportionality, including: the MAL index (previously available in distorsion()), the Gallagher index (Gallagher, 1991) or the Representational Inequality Index (Kedar et al., 2016), among others.

The new function malapportionment_index() computes the malapportionment index as defined by Samuels and Snyder (2001).

Three changes has been made to the outputs of the distorsion(). First, a mismatch between labels and numbers of party distortion effects in the output party.distortions of the function distorsion() has been corrected. In the previous version, the correct order of labels had been “Total deviation, D_i”, “Electoral system effect, A_i”, “Population effect, B_i”, “Mean effect, d_i”, “Variance effect, e_i”, ” Malapportionment effect, f_i” and “Turnout effect, g_i”. In the new version, we have maintained the order of the labels and reorder the rows. Second, the MAL index is not longer available in the output aggregate.distortions of the function distorsion(). Third, a new output district2party.contributions, which contains a matrix of order kxp with the district contributions to the total seat-to-vote deviation of each party, has been included.

A plot method has been included to draw the seat-to-vote descomposition for every party from the distorsion() output and the district contributions to the total seat-to-vote deviation of each party from the distorsion() output, as well as the representational inequality (Lorenz) curve of the election from the inequality() output.