The calculation of volatility poses problems in party systems where there is high discontinuity in electons (ie. parties and any other “electoral units”, most importantly coalitions), either because of breakthroughs of new ones or innovation amongst old ones (i.e. splits and mergers). Dichotomous classification of such partially novel electons as new or old can both be misleading and have a disproportionate effect on volatility scores. The paper proposes a new measure of congruence (or novelty) combining on three dimensions: (a) an electon’s name and organization, (b) leader, and (c) top-ranking candidates. To illustrate the proposed index and an innovative technique of weighting vote shares for calculating volatility, the paper discusses five diverse cases of (partially) novel electons: Kadima (2006, Israel), Unity (2010, Latvia), Danish People’s Party (1998), Pro Patria and Res Publica Union (2007, Estonia) and Electoral Action Solidarity (break-up, 2001, Poland). The proposed approach to the calculation of volatility is computationally slightly more intensive than the standard approach, yet feasible, and the data is easy to collect for the last two decades.
