An abundant empirical literature on corruption relying on survey research has emerged since the mid-1990s. The predominant line of inquiry concerns perceptions of corruption with respect to institutions and processes. Another, separate line of inquiry that has enjoyed less attention concerns reports about individuals’ participation in corruption. These two dimensions of corruption, however, are typically conflated, leading to error and confusion. This article explores the relationship between the two and seeks to differentiate the two. Using data at the country and individual levels, analysis shows how the two may be only weakly related to one another – though causality remains unclear – and respond to distinct sets of determinants and generate distinct outcomes. The analysis underlines the need to specify the findings in the literature: that the causes and consequences of corruption relate more to ‘perceived’ corruption rather than actual corruption.