A method for assessing the cultural distance between countries has recently been proposed by De Santis, Maltagliati and Salvini (2015). Respondents from different countries are clustered on the basis of their answers to a set of questions. Countries whose respondents distribute similarly among cluster are (culturally) close, where “closeness” can be measured geometrically, as a Euclidean distance. The method is replicated here on Italian (Istat, Multiscopo 2013) data and improved in several respect, starting from the clustering method (here: EM, or Expectation Maximization). People from adjacent administrative regions, age groups and educational levels appear to be culturally close. Gender, conversely, appears to have no measurable effect.
Culturally close? Easier said than measured
DE SANTIS, Gustavo;MUCCIARDI, Massimo
2015-01-01
Abstract
A method for assessing the cultural distance between countries has recently been proposed by De Santis, Maltagliati and Salvini (2015). Respondents from different countries are clustered on the basis of their answers to a set of questions. Countries whose respondents distribute similarly among cluster are (culturally) close, where “closeness” can be measured geometrically, as a Euclidean distance. The method is replicated here on Italian (Istat, Multiscopo 2013) data and improved in several respect, starting from the clustering method (here: EM, or Expectation Maximization). People from adjacent administrative regions, age groups and educational levels appear to be culturally close. Gender, conversely, appears to have no measurable effect.Pubblicazioni consigliate
I documenti in IRIS sono protetti da copyright e tutti i diritti sono riservati, salvo diversa indicazione.