Ways to understand and measure bushmeat hunting to improve targeting of conservation interventions: a case study of the GolaMA project, Liberia

Over-hunting is a major driver of biodiversity loss and threatens people’s food security and livelihoods worldwide. I describe a rural hunting system at a conservation project site in Gola, Liberia, and explore how social science tools can help conservationists understand and influence human behaviour. In marketing, the technique of ‘audience segmentation’ is used to identify which specific group of people will be targeted by a campaign. I applied audience segmentation to differentiate hunters, identifying distinct hunter types that differed according to livelihood portfolios, hunting methods, citizenship (indigenous locals or non-locals) and previous experience of law enforcement. Results suggested that interventions should seek to target specific groups, for instance, programmes to support income from cocoa farming could be appropriate for local trappers with cocoa farms, but not for non-local gun-hunters who did not own plantations. Measuring people’s behaviour is challenging where activities are illegal and could be under- reported. I evaluated hunting and trading over a two-year period using the bean method, a technique designed to encourage truthful reporting by ensuring people’s answers remain anonymous. Results indicated a decrease in bushmeat trading from 36% to 20% of households, but little change in hunting. Most respondents (>90%) gave direct answers that were consistent with bean method answers, suggesting questions were not sensitive despite conservation interventions aiming to reduce hunting. The technique was low-cost and straightforward to implement. Harvest datasets are fundamental for understanding hunting systems but are susceptible to sampling and measurement biases. I compared catch-per-day from two methods, hunter recall interviews and village-based monitoring, revealing a two-fold difference in estimates. Results suggest non-random sampling of hunters’ or hunting trips could be a major source of bias in harvest datasets. Conservation interventions can be improved by better targeting of interventions, supported by robust tools to measure resource-use behaviour.