Abstract: This paper analyzes the determinants of the probability of USA citizens being victims of terrorist attacks occurred in European countries, taking into account uncontrolled heterogeneity of the data. The analysis employs ITERATE data from February 1968 to December 2002 to ascertain significant characteristics that influence the aimed probability (e.g. location, type of casualties, type of attack, and type of terrorists). To deal with the unobserved heterogeneity a random-parameter logit model (mixed logit) is used. Some policy implications are presented.