José Alexandre da Silva

When is statistical significance not significant

The article provides a non-technical introduction to the p value statistics. Its main purpose is to help researchers make sense of the appropriate role of the p value statistics in empirical political science research. On methodological grounds, we use replication, simulations and observational data to show when statistical significance is not significant. We argue that: (1) scholars must always graphically analyze their data before interpreting the p value; (2) it is pointless to estimate the p value for non-random samples; (3) the p value is highly affected by the sample size, and (4) it is pointless to estimate the p value when dealing with data on population.

Does Size Matter? Electoral Performance of Small Parties in Brazil

What is the impact of small parties on electoral outcomes? This articles aims at contributing to the literature on party systems by proposing a new method to classify political parties. The methodology is applied to Brazil by focusing on the description of the election results of small parties. Cluster analysis is employed to classify political party size based on their percentage of votes in the Brazilian states. The main findings indicate that classifying parties through cluster analysis is more objective than previous classifications. As a result of this method, the article shows that small parties exert little effect on electoral volatility in Brazil as well as small parties benefit less from the disproportionality between votes and seats than the larger ones.

A rosa púrpura do cárcere: O encarceramento de mulheres no Brasil (2003-2007)

O encarceramento feminino se tornou uma preocupacao nao apenas de organizacoes nao governamentais, mas tambem do Estado. Institucionalmente, estabelecer um sistema efetivo de protecao as mulheres e um dos principais desafios enfrentados pelos gestores governamentais. O principal objetivo desse trabalho e identificar o perfil da mulher encarcerada no Brasil. Em termos metodologicos, utilizamos estatistica descritiva, correlacao de Pearson e analise espacial para analisar os dados compilados pelo Sistema Integrado de Informacoes Penitenciarias (Infopen). Os resultados apontam que: (1) o perfil da mulher encarcerada indica que ela possui Ensino Fundamental incompleto, esta na faixa etaria entre 18 e 30 anos e cumpre pena em regime fechado de ate oito anos por crime de trafico de entorpecentes; (2) a populacao carceraria feminina tem aumentado ao longo do tempo; e (3) existem conglomerados de estados geograficamente proximos que apresentam valores similares de encarceramento feminino, sugerindo dependencia espacial das observacoes.

Capacidade institucional e corrupção: Tribunais de Contas Estaduais em perspectiva comparada

This paper analyzes the relation between the institutional capacity of the State Courts of Accounts and corruption. Methodologically, we combine descriptive and multivariate statistics to analyze a database compiled from Melo, Pereira and Figueiredo (2009) and Ferraz and Finan (2011). In particular, we use principal component analysis to estimate an indicator of institutional capacity as the percentage of the state budget, budget transparency and seniority of each Court. We also employ a linear ordinary least squares (OLS) model to estimate the effect of institutional capacity on the incidence of corruption. The results suggest that: (1) in comparison, Sao Paulo presents the Court with greater institutional capacity, while Mato Grosso do Sul presents the worst performance; (2) on average, the larger the budget of the Court, the lower the level of transparency and (3) the higher the institutional capacity of the Court, the higher the probability of detecting corruption.