Subjective modeling choices and the robustness of impact evaluations in conservation science

A comprehensive analysis of Virunga National Park reveals that subjective modeling choices can significantly influence the estimated impacts on tree cover loss and carbon storage, despite consistent findings across models. The study evaluated over 5000 candidate models to assess how variations in statistical matching algorithms, parametrization, and covariates affected results. While all valid models confirmed a decrease in tree cover loss and an increase in carbon storage from 2000 to 2019, the magnitude of these changes varied by up to threefold. Notably, the choice of matching algorithm emerged as a critical factor affecting point estimates, underscoring the need for structured robustness checks in conservation science to ensure credible findings.

Desbureaux, S. (2021). Subjective modeling choices and the robustness of impact evaluations in conservation science. Conservation Biology, 35(5), 1615–1626. Portico. https://doi.org/10.1111/cobi.13728