“…It was the fourth Presidential election since Chile's return to democracy, held at a time of continuing economic growth and high popularity of the incumbent Concertación government, and with Pinochet relegated to a marginal role in the national political scene (Bonilla, 2002;Angell and Reig, 2006). Also, for the first time in its history, the two main partners of the Alianza por Chile, UDI and RN, presented independent candidates who adopted relatively different electoral strategies: while Lavín (UDI) adopted an aggressive campaigning style aimed at consolidating the vote among his right-wing supporters, the candidate of the National Renewal Party, Sebastián Piñera, took a more moderate stance, distancing himself from the traditional right and the legacy of the military regime in order to capture the support of centre in view of the almost certain second-round runoff between the candidate of the Concertación and one of the two conservative candidates (Angell and Reig, 2006;Gamboa and Segovia, 2006). Together In order to analyze the relative influence of socio-demographic, ideological and political variables on voter choice at the individual level, we specify and estimate a Bayesian multinomial probit model that explicitly accounts for the multi-party character of the election by letting voters evaluate all competing candidates simultaneously and to 'group' alternatives they consider similar when choosing for which candidate to vote.…”