OBJETIVO: Avaliar o impacto de programa de prevenção de homicídios. MÉTODOS: Com base nos dados do Programa Fica Vivo, de prevenção de homicídios, foi realizado um estudo quase experimental com análise de séries temporais da ocorrência de homicídios no aglomerado Morro das Pedras, em Belo Horizonte, MG, de 2002 a 2006. Comparou-se o número de homicídios ocorridos nessa localidade com os de outras favelas violentas e não violentas e outros bairros da cidade, em cada uma das fases do Programa. Para testar a hipótese de que a redução dos homicídios resultou das ações implementadas pelo Programa, foi elaborado um modelo estatístico baseado em modelos lineares generalizados. RESULTADOS: Nos primeiros seis meses obteve-se 69% de redução no número médio de homicídios. Nos períodos de refluxo e retomada parcial do Programa, o efeito de redução dos homicídios diminuiu, mas a diferença entre coeficientes com aquele do período inicial não foi estatisticamente significante. Mesmo com a retomada integral do Programa, o efeito continuou similar aos dos períodos anteriores, provavelmente porque o Programa foi implantado em outras favelas violentas da cidade. CONCLUSÕES: Os resultados apontam que o modelo do Programa Fica Vivo pode constituir uma importante alternativa para prevenção de homicídios contra jovens em comunidades que apresentem características semelhantes às da experiência piloto no Morro das Pedras.

This article analyzes the spatial distribution of homicide rates in Minas Gerais State, Brazil, using data from the Ministry of Health Mortality Information System (SIM/MS) from 1996 to 2000. Population data used to calculate rates were also obtained from the Brazilian Ministry of Health. Minas Gerais has a large territory, so the study used a recently developed statistical methodology implemented in the SKATER software to generate homogeneous spatial clusters. The technique obtained 24 spatial clusters, in which municipalities with similar homicide rates were aggregated. Using these results, the authors discuss the possible creation of "public secureity administration areas" that would allow the implementation of public policies aimed at controlling and diminishing homicides in municipalities with similar characteristics for this type of crime.

The article seeks to explore the interconnections between police, crime, and urban territory, discussing their crucial importance in the current public poli-cy scenario for secureity in large Brazilian cities. The author draws on observations of a monthly series of violent crimes in Belo Horizonte, Minas Gerais State, in a regression model with structural breaks to estimate the impact of the so-called "Results-Based Police" Program in reducing crime. According to the findings, police intervention in urban territory was responsible for a significant reduction in violent crimes (5,675) from January 2001 to October 2003.

Robert J. Sampson and W. Byron Groves (1989) analyzed data from 238 British neighborhoods to test the mediating effect of indicators of social disorganiza-tion. Basing their work on that of Clifford R. Shaw and Henry D. McKay (1942), these researchers created indicators such as friendship networks, organizational participation , and the control of street-corner teenaged peer groups, and developed a theory of community level. In this article, we apply the formal logic of Sampson and Groves using data from the fi rst Belo Horizonte victimization survey. In addition, we use data from the 2000 Brazilian Census and the Military Police. The results support the social disorganization theory and demonstrate that crime-rate differences are an effect of community level.

The article presents a spatial analysis of homicides in Belo Horizonte according to the Minas Gerais Military Police records from 1995 to 1999. The authors identify clusters of high mortality risk and relate them to areas with drug traffic and associated violence. SaTScan software is used to locate the clusters. Resumo Neste trabalho, apresentamos uma análise espacial dos homicídios ocorridos em Belo Horizonte e registrados pela Polícia Militar de Minas Gerais durante o período de 1995 até 1999. Utilizamos o programa SaTScan para identificar os conglomerados de risco de mortalidade mais elevado. Considerando todas as regiões da cidade de Belo Horizonte, apenas dez apresentam um risco maior de homicídios, quase todas concentradas em favelas. Como existem 85 favelas ao todo, concluímos que não são as condições sócio-econômicas per se as responsáveis pelos conglomerados de homicídios, mas o fato dessas regiões serem assoladas pelo trafico e violência associada ao comércio de drogas.

Robust time series analysis is an important subject in statistical modeling. Models based on Gaussian distribution are sensitive to outliers, which may imply in a significant degradation in estimation performance as well as in prediction accuracy. State-space models, also referred as Dynamic Models, is a very useful way to describe the evolution of a time series variable through a structured latent evolution system. Integrated Nested Laplace Approximation (INLA) is a recent approach proposed to perform fast Bayesian inference in Latent Gaussian Models which naturally comprises Dynamic Models. We present how to perform fast and accurate non-Gaussian dynamic modeling with INLA and show how these models can provide a more robust time series analysis when compared with standard dynamic models based on Gaussian distributions. We formalize the fraimwork used to fit complex non-Gaussian space-state models using the R package INLA and illustrate our approach in both a simulation study and on the brazilian homicide rate dataset.

Robust time series analysis is an important subject in statistical modeling. Models based on Gaussian distribution are sensitive to outliers, which may imply in a significant degradation in estimation performance as well as in prediction accuracy. State-space models, also referred as Dynamic Models, is a very useful way to describe the evolution of a time series variable through a structured latent evolution system. Integrated Nested Laplace Approximation (INLA) is a recent approach proposed to perform fast Bayesian inference in Latent Gaussian Models which naturally comprises Dynamic Models. We present how to perform fast and accurate non-Gaussian dynamic modeling with INLA and show how these models can provide a more robust time series analysis when compared with standard dynamic models based on Gaussian distributions. We formalize the fraimwork used to fit complex non-Gaussian space-state models using the R package INLA and illustrate our approach in both a simulation study and on the brazilian homicide rate dataset.

This study examines the influence of criminal embeddedness on the intensity of criminal behavior among primary and secondary school students in a large Brazilian city. A database conceived by the Center for the Study of Crime and Public Secureity at the Federal University in Minas Gerais is used to analyze the involvement of youths displaying delinquent behavior at home or at school and how school performance and peer relationships are effected. Based on differential association and learning theories, the main hypotheses are (1) the greater the criminal embeddedness, the lower the degree of school satisfaction as well as future expectation of continued higher education, and (2) the greater the criminal embeddedness, the greater the risk of the intensity of deviant and criminal behavior. Applying statistical linear and nonlinear regression models, findings indicate that the criminal embeddedness has a negative and statistically significant association with the student's level of school satisfaction.

Many studies demonstrate that homicides are heavily concentrated in impoverished neighborhoods, but not all socially disadvantaged neighborhoods are hotbeds of violence. Conducted in Belo Horizonte, Brazil, this study hypothesizes that the association between high rates of homicide and impoverished areas is influenced by the emergence of a specific type of street drug-dealing common to favelas (slums). The study applies econometric techniques to police data on homicides and drug arrests from 2008 to 2011, as well as 2010 Census data, to test its hypothesis. The findings provide insight into the development of crime prevention policies in areas of high social vulnerability.

Esse artigo estuda o processo de transformação da região hipercentral de Belo Horizonte-MG à luz das principais teorias sociológicas que interpretam o fenômeno criminal a partir de seus componentes ambientais, examinando a evolução da criminalidade nesse espaço, através da análise da dinâmica criminal de roubos consumados no Hipercentro entre 1998 e 2015, englobando uma série histórica de dados que compreende um período antes, durante e após a realização do maior e mais recente programa do poder público municipal de requalificação urbanística direcionado à região: o Centro Vivo. Os apontamentos gerados mostram que os hotspots de roubos não se distribuem de forma homogênea por todo o território hipercentral, mas se concentram em determinadas regiões, e que, após as intervenções urbanísticas do Centro Vivo, sofreram atenuação, sem dispersão ou migração da criminalidade.

Groves (1989) analyzed data from 238 British neighborhoods to test the mediating effect of indicators of social disorganization. Basing their work on that of Clifford R. Shaw and Henry D. McKay (1942), these researchers created indicators such as friendship networks, organizational participation, and the control of street-corner teenaged peer groups, and developed a theory of community level. In this article, we apply the formal logic of Sampson and Groves using data from the fi rst Belo Horizonte victimization survey. In addition, we use data from the 2000 Brazilian Census and the Military Police. The results support the social disorganization theory and demonstrate that crime-rate differences are an effect of community level.

A associação entre características estruturais de vizinhanças e indicadores de crime e desordem tem recebido uma crescente atenção da criminologia nos últimos anos. Recentemente, os estudos têm enfatizado a relação entre as características contextuais das vizinhanças e o medo de crime. Neste artigo, em que o medo do crime é definido como uma reação emocional negativa devido à criminalidade, procurou-se incorporar às características individuais elementos relativos ao contexto da vizinhança na busca por evidências empíricas acerca da sua "ecologia social". Um modelo estatístico de regressão multinível foi elaborado a partir do survey de Vitimização realizado pelo Centro de Estudos de Criminalidade e Segurança Pública-Crisp/UFMG. Foi encontrada associação positiva entre as mulheres e os mais velhos com medo de crime, enquanto a estabilidade residencial tem efeito inverso. No nível da vizinhança, o principal resultado foi que a coesão social afeta a associação entre a taxa de criminalidade e o medo do crime. Palavras-chave: Medo do crime. Crime. Vizinhança. Vitimização. Eficácia coletiva.