This project aims to answer the following research question: What explains the success of transitional justice mechanisms (TJMs)? In recent years, a number of impact studies have attempted to answer similar questions but they have been largely inconclusive due to three limitations. First, they use countries as units of analysis rather than TJMs. Their underlying assumption is that transitional justice (TJ) processes should be distinguished based on the number, timing, and/or sequencing with which TJMs are adopted, claiming that 'holistic' TJ approaches are preferred to 'minimalist' ones. However, qualitative studies show that TJMs of the same type - such as truth commissions, trials, purges, or reparation measures - can achieve very different things. Second, despite the proliferation of impact assessment studies, their focus has been on the goals stated by the actors involved rather than on the actual outcome of a specific TJM. This raises problems regarding the possibility to identify the causal link between TJ and peace, reconciliation, trust, and democratization, among other things. Finally, contextual factors are usually reduced to the mode of transition and the type of regime, yet there are a number of other contextual factors that should be taken into account, particularly regarding the political dimension of transitional justice and the politicization of the past. This project offers to build a new large dataset using TJMs as units of analysis and covering European, Latin American, African and Asian post-authoritarian and post-conflict democracies. This strategy will enable to differentiate between TJMs of the same type by taking into account aspects such as their mandate, leadership, target, and partiality, among others. A second dataset will be created to measure TJ outcome through expert surveys. Using this method will allow for the collection of information on aspects that have been largely neglected in the literature, such as levels of mediatization, social mobilization, politicization, and international repercussion, among others, thus making a significant theoretical contribution. The two new datasets will enable us to answer the research question through a two-step strategy. First, the project will use descriptive statistics to create a TJ typology, and second, the project will use statistical analysis to identify major explanatory factors that can contribute to the different types. The ultimate aim of the project is to build a new dataset that can be used by the academic and the non-academic community. This is important because international organizations and practitioners have been advocating the implementation of certain measures to the detriment of others on the basis of the findings produced by impact assessment studies. Finally, the project offers to create a TJ Observatory to monitor TJ developments in the world focusing on the same aspects highlighted above and not only on the mere implementation of new TJMs.