Conservation interventions often rely on rules that restrict and regulate activities harmful to biodiversity. To be effective, such interventions require reliable information about the prevalence of rule-breaking behaviours, alongside sufficient understanding of the factors that affect compliance. However, this is challenging given the secretive nature of rule-breaking. Subsequent data may be heavily biased and thus unreliable, resulting in inappropriate conservation interventions which represent a waste of participant’s time and research resource. My thesis aims to increase understanding of rule-breaking behaviour in conservation with a particular focus on developing and advancing quantitative and qualitative methods. I use a combination of fieldwork with people living around protected areas in two highly biodiverse but socially and economically dissimilar countries (Indonesia and Tanzania), and systematic reviews of existing research.

First, I explore different approaches for understanding if and when a topic is sensitive and develop a psychometric scale to measure topic sensitivity. Then I investigate how Specialised Questioning Techniques, a suite of methods developed by social scientists to encourage people to answer sensitive questions more accurately, might contribute to obtaining robust information about sensitive rule-breaking behaviours in conservation. I use an experimental approach to critically assess the performance of Specialised Questioning Techniques (specifically the Randomised Response Technique, the Unmatched Count Technique, the Crosswise model and the Bean method) when asking about a common rule-breaking behaviour: wildlife hunting. Results suggest these methods do produce more accurate data, but that respondents do not always feel more at ease using them. I then focus on one method, the Randomised Response Technique, and provide conservation scientists with detailed guidance on robust study design and application, based on a systematic review of studies using this approach to quantify rule-breaking. Building on these lessons, I apply the RRT to quantify compliance in the Ruaha-Rungwa ecosystem of Tanzania and assess how well it performed. Results highlight that Randomised Response Techniques cannot overcome participants’ lack of trust in the research process. Lastly, I examine factors that motivate compliance with protected area rules. Drawing on previous conservation research and using factorial survey experiments, I explore how the administration of rules by law enforcers around protected areas can influence compliance. Importantly, results show that people want rules to be administered fairly, and that the abuse of power by law enforcers, for example through the acceptance of bribes, is not tolerated.

This thesis enhances conservation science by providing empirical evidence and detailed guidance on how to study rule-breaking.