A comparison of partisan-gerrymandering measures.

摘要: We compare and contrast fourteen measures that have been proposed for the purpose of quantifying partisan gerrymandering. consider that, rather than examining shapes districts, utilize only vote distribution among districts. The considered are two versions bias; efficiency gap several its variants; mean-median difference equal weight standard; declination one variant; lopsided-means test. Our primary means evaluating these is a suite hypothetical elections we classify from start as fair or unfair. conclude most successful measure in terms avoiding false positives negatives on considered. include an appendix extreme outliers each historical congressional state legislative elections.