New party programs, policy, personnel or electorate: What does ‘a new party’ really mean?

Party novelty is a complicated issue to define, some scholars perceive as new parties with a new origin or competing in an election for the first time (Bolleyer 2013, Hug 2001), competing on novel issues (Lucardi 2010, Zons 2015), or having new leadership and personnel (Sikk 2005), the others point out that party novelty is a multidimensional issue and can be disaggregated – parties can be new in some areas (Barnea & Rahat 2011) and to some degree (Sikk & Köker 2019). One method of verifying whether a party is new or not is to conduct an expert survey, but if party novelty is a multidimensional variable, what do experts take into consideration when classifying parties as new: programs, policy, personnel (candidates, elites, leaders), the electorate, or some other features? The aim of this presentation is to compare the outcomes of the expert survey with the results obtained from the analysis of the party novelty based on empirical data from other sources: party documents, electoral data, etc. Such comparison is supposed to answer the main research question of the paper – to what extent the perception of party novelty or continuity by experts is congruent with diverse data-based measures. The level of such congruence might provide the answer for the more general question – what features are decisive for parties to be perceived as new. On the basis of an expert survey of genealogical relationships between post-transition Polish political parties, conducted in 2022 on the group of over 50 experts, we test four hypotheses regarding factors underlying expert perceptions of continuity: that they correlate with (H1) personal continuity, (H2) ideological (programmatic) continuity, (H3) policy continuity, and (H4) continuity of electoral support patterns. To measure the four explanatory variables, we introduce and compute novel indices that incorporate big data analysis and machine learning methods.