[Multicenter Study on Cultural Attitudes and Norms towards Violence (ACTIVA project): methodology]

The purpose of this article is to describe the conceptual bases and methods used in the ACTIVA multicenter study, as well as the process employed in the different stages of the research. General information is given on the approach, design, and methods that were applied in the cities that participated in the study: San Salvador, El Salvador; San José, Costa Rica; Cali, Colombia; Caracas, Venezuela; Salvador, Bahia, Brazil; Rio de Janeiro, Brazil; Santiago, Chile; and Madrid, Spain. The objectives of the study were to: (1) analyze and compare among cities the prevalence of verbal and physical aggressive behaviors toward children, the spouse or other partner, and persons outside the family; and (2) identify personal, environmental, and socioeconomic factors associated with these violent behaviors. The purpose of the ACTIVA project was to generate information to help guide violence-prevention policies and programs in the cities and to provide criteria for decision-making, as well as to obtain baseline data to use in evaluating prevention policies and programs. To achieve the proposed objectives, a single, common questionnaire was prepared and validated. The questionnaire was used in face-to-face interviews with residents of private urban homes who were between 18 and 70 years old. The sample was selected in each city through stratified multistage sampling of clusters proportionate to the socioeconomic strata of the general population and without any substitutions. The final sample consisted of 10,821 people. In general, the response rates were within the established limits. Exceptions included San José, Rio de Janeiro, and Madrid, where the size of the final sample was slightly smaller than anticipated. More young men of higher socioeconomic strata declined to participate, so that the samples had an overrepresentation of women from medium and lower strata. The article concludes with a description of the scope and limitations of the study, in terms of both the approach and methods. With their comparative analyses, studies of this type can contribute to identifying the differences and similarities between cities. The article shows that with this method it is possible to analyze how violent behaviors are associated with different risk factors and individuals' characteristics, at a point in time. However, this design is limited in terms of establishing causal relationships, and it did not allow an evaluation of the context of every specific instance of violence.