Elicited constructed value of information scores for American kestrel uncertainties in the United States from a 2023 structured decision making workshop

The American kestrel (Falco sparverius; hereafter kestrel) is a widespread but declining raptor species in North America. We organized a rapid prototyping workshop to use decision analysis to develop research priorities for understanding environmental drivers of population decline in the United States for the kestrel. The virtual workshop spanned 2.5 days and was held in-person in Albuquerque, NM and online in January–February 2023. Workshop participants, comprised of species experts/researchers and wildlife biologists, chose to focus the decision analysis process on developing management-relevant hypotheses that described uncertainties relevant to maintaining abundance and maximizing persistence probability for the kestrel. We used constructed value of information to prioritize uncertainties (stated as alternative hypotheses developed by participants in breakout groups) that could serve as the basis for subsequent coordinated research studies. Constructed value of information (CVoI) is a newly-developed decision analysis tool that scores uncertainties in three areas: (1) Magnitude of uncertainty, which reflects the strength of theoretical foundation and empirical support of the hypothesized relationship; (2) Relevance to a management decision, which indicates how likely the preferred management alternative is to change if the uncertainty were resolved; and (3) Reducibility, which is the degree to which the uncertainty could be resolved through research and monitoring. Magnitude is scored on a scale from 0–4 while Relevance and Reducibility are scored on a scale from 0–3. These data are the anonymized workshop participant scores from two scoring rounds (n=33 and n=26, respectively) for thirteen hypotheses focused on potential environmental drivers of population decline for the kestrel. The data are contained in a .csv file that can be opened using a spreadsheet program such as Microsoft Excel, or read into a statistical analysis program such as Program R.