RESUMO
Focus group discussions (FGDs) and individual interviews (IIs) with community members are common methods used in evaluations of all kinds of projects, including those in international development. As resources are often limited, evaluators must carefully choose methods that yield the best information for their particular program. A concern with FGDs and IIs is how well they elicit information on potentially sensitive topics; very little is known about differences in disclosure by methodology in the domain of justice. Using FGDs (n = 16) and IIs (n = 46) from a USAID project in Haiti, we systematically coded responses based on a shared elicitation guide around access to and engagement with the formal and informal justice systems and performed thematic and statistical comparisons across the two methods. We introduce the continuous thought as the novel standard unit for statistical comparison. Participants in IIs were statistically more likely to provide themes relevant to genderbased violence. Importantly, sensitive themes extracted in IIs (e.g., related to sexual violence, economic dimensions, and restorative justice) did not emerge in FGDs. Given these results and other limitations to the FGD, prioritizing interviews over focus group modalities may be appropriate to guide targeted, effective programming on justice or other socially sensitive topics.