Fig 2 - uploaded by Matias Enrique Mastrangelo
Content may be subject to copyright.
Flowchart showing the main methodological steps included in the tool. First, sparse and fragmented evidence is collected through literature reviews and integrated into qualitative matrices. Then, experts assess qualitative matrices and score each link to produce quantitative matrices. Matrices of longchain links obtained through literature review and expert opinion (primary) and matrix multiplications (resultant) are compared to test the reliability of the tool. Finally, integrated and synthesized evidence is depicted in balloon/triangle plots and impact-influence diagrams, where dissimilar stakeholders' vulnerabilities are visualized.
Contexts in source publication
Context 1
... direct drivers and benefits are mediated by landscape types and ecological conditions (Fig. 1). The five matrices of short-chain links are primary, which means that they are filled following a two-step procedure by which quantitative matrices arise from expert scoring based on qualitative matrices built out of evidence from literature review (Fig. 2). The three matrices of long-chain links have two versions: i) a primary version (p) filled following the two-step procedure, and ii) a resultant version (r) that ensues from the sequential multiplication of quantitative matrices (Fig. 2). Next, we firstly describe the eight primary matrices, and then describe how the tool produces the ...
Context 2
... matrices arise from expert scoring based on qualitative matrices built out of evidence from literature review (Fig. 2). The three matrices of long-chain links have two versions: i) a primary version (p) filled following the two-step procedure, and ii) a resultant version (r) that ensues from the sequential multiplication of quantitative matrices (Fig. 2). Next, we firstly describe the eight primary matrices, and then describe how the tool produces the three resultant matrices and the assessment of stakeholders' ...
Context 3
... about each short-and long-chain link included in the eight primary matrices. Then, pieces of evidence for each link are compared and integrated into text that summarizes the knowledge available about each link. In this way, qualitative matrices filled with synthesis texts are constructed before quantitative ones filled with expert scores (step 3) (Fig. 2). 3) Construction of quantitative matrices: once qualitative matrices are constructed based on literature reviews, these are presented to a set of local experts with diverse disciplinary backgrounds. Based on the text synthesizing the available knowledge about each link and its own expert opinion, experts assign a quantitative score to ...
Context 4
... is scarce or absent. Where available, experts also score long-chain links and produce primary versions of these matrices (M6p, M7p, and M8p). Whether the tool yields a reliable estimation of the available knowledge on long-chain links can be tested by looking at the differences between scores in primary and resultant versions of M6, M7, and M8 (Fig. 2). 5) Estimation of confidence levels: to inform managers and policymakers about the level of confidence associated with each quantitative score, a confidence level is defined and assessed for each ...
Context 5
... we applied the sequence of matrix multiplications described above and depicted in Fig. 1. We compared the scores of long-chain links contained in the primary and the resultant versions of matrices 6, 7, and 8. The difference between these scores was not different from zero for 92 out of the 108 long-chain links contained in matrices 6, 7, and 8 ( Fig. S2), indicating that matrix multiplications reliably estimated the magnitude and sign of long-chain ...
Context 6
... of long-chain links is scarce, as in AGC, expert quantitative scoring and matrix multiplications are important steps to estimate social impacts and vulnerability. However, matrix multiplications tend to smooth the magnitude of relationships, and resultant matrices capture ecological or social trade-offs less accurately than primary matrices (Fig. S2). Thus, multiplications are less needed where evidence of long-chain links is abundant, although quantification may still be helpful to communicate information in easy-to-visualize plots and diagrams. Also, where knowledge holders and/or users are unfamiliar with quantitative scores, qualitative matrices alone sufficiently integrate ...