-. (1997). Maggioritario per caso. Le elezioni politiche del 1996.

De Sio, L. (2009). Oltre il modello di Goodman. La stima dei flussi elettorali in base a dati aggregati. Polena, 9–35.

The 2012 municipal election in Palermo produced an unexpected outcome. In the Sicilian city - for a long time a conservative stronghold - the center-right candidate, Massimo Costa did not succeed to reach the second ballot and the election was won by the former Major Leoluca Orlando, supported by a radical left coalition. Orlando prevailed with a sensational 72% of the vote share against the winner of the center-left primary elections, Fabrizio Ferrandelli. What happened in the 2012 Palermo municipal election? Does the Sicilian capital moved suddenly toward the left? Which factors fostered this sharp and unpredictable electoral change that altered the long-time-established political landscape of the City? To answer these questions, the article analyzes the results of the 2012 municipal election in Palermo through an electoral geography approach and the use of a quantitative methodology with ecological data. In particular, the article makes use of both the territorial study of turnout and election results and the voting ecological estimates generated with the traditional Goodman model. The empirical analysis shows that this election was strongly influenced by factors linked to the local context more than by authentically political ones. In other words, Palermo did not move toward the left. Moreover, the internal electoral segmentation of the City between central and peripheral neighborhoods persisted as the main determinant of the vote choice.

D’ALIMONTE, R. D. R., & CHIARAMONTE, A. (1995). Il nuovo sistema elettorale italiano: le opportunità e le scelte. In S. Bartolini & R. D’Alimonte (Eds.), Maggioritario ma non troppo. Le elezioni politiche del 1994 (pp. 37–81). Bologna: I...

This article aims to rediscover a variable that has been rather neglected by the Italian electoral studies on the so called «Second Republic»: demographic size of municipalities. Is there a difference between a citizen who votes in a small municipality of North-east and another one who votes in Milan? Between voting in a rural village or in an urban metropolis? In other words, is territory – considered as centrality or peripherality of the municipality where vote is cast – important to understand Italians’ electoral choices? And if so, how much it matters? May it even become a decisive dimension for the electoral results? Moving from these questions, the article analyzes the results of 2008 Italian general election by dividing the more than 8.000 Italian municipalities in 5 classes of demographic size (0-5.000, 5.001-15.000, 15.001-50.000, 50.001-100.000, above 100.000) and the territory of our country in 4 geo-political sub-units (North-west, North-east, Red belt and South) in order to develop a complete mapping of the incidence of demographic variable on the vote. This study concerns the 2008 vote to main Italian parties, coalitions and electoral blocs and uses the analysis of variance to calculate the tightness of the association between the above variable and the vote through a synthetic index. The findings are very interesting and in some ways surprising. Demographic size matters, especially in some areas (North) and for some parties (Northern League, Pd, Udc, Idv). In particular, three possible behaviours occur: some parties, definable as «city oriented», tends to achieve increasing electoral results whenever the size of municipality grows (eg. Pd, Idv); other parties, labelled as «village oriented», show an opposite trend, that is strongly rooted in small towns and a systematic loss of votes when demographic size increases (Northern League, Udc); the third type of behaviour is given by some «all around» political forces (Pdl, La Destra, Mpa) that show indifference to the variable. An even more pronounced effect could be found in coalitions and blocs analysis, with the centre-left collecting a strictly urban vote and the centre-right stronger in small towns.