Laserfiche WebLink
<br />December 2003 <br /> <br />Figure 4 and Figure 5 summarize the ranks assigned by stakeholders to each option by the <br />Direct and Swing Weighting methods respectively. For each stakeholder, options ranked 1 <br />through 3 are colored blue, 4 through 6 are yellow, 7 through 9 are white, and 10 through 12 <br />are red. Be aware that there are some boundary problems with this approach; in some cases <br /> <br />calculated scores show very little different in the value of two options, but if one is ranked 3rd <br /> <br />and another 4th they are assigned different colors, potentially exaggerating differences in <br />value. <br /> <br />Overall, options 7 and 10 appear to be the strongest candidates. No participants liked Options 1 <br />or 2 by either the direct or swing weighted methods. While options 4, 8, 9 and 11 have quite a <br />bit of support in direct ranking, they do not score well by weights. <br /> <br />On the basis of these results, the TWG may want to consider eliminating Options 1 and 2 (MLFF <br />and SASF) from further consideration. <br /> <br />Some alternatives are dominated or nearly dominated by others, and the TWG may want to <br />consider eliminating them from further consideration as well: <br /> <br />Option 3 (P) is dominated by 12 (P+Mech) for nearly all stakeholders by both methods <br />(exception is TWG 13 who slightly prefers P by Direct and TWG 14 who slightly prefers P by <br />Swing Weighting). See Table 1 in Appendix D for a direct comparison of these two options. <br />The trade-off to consider is quite simple: a loss of S 1.0 million per year in mitigation costs <br />for a gain in LCR Chub EU (i.e., a reduction in the probability of extinction from 10 to 5%). <br /> <br />Option 8 (P+ TCD8) is dominated by 9 (P+ TCD8+Mech) by both methods for nearly all <br />stakeholders. See Table 2 in Appendix D. Again, the key trade-off to consider is the <br />mitigation cost versus the gain for LCR Chub. <br /> <br />On swing weights, Option 4 (MLFF+FSF+BHBF) is dominated by 6 (P+ MLFF+FSF+BHBF); however <br />several participants preferred 4 by direct ranks. Further consideration could be given to <br />whether option 4 can be eliminated. However, there are complicated performanc.e trade-offs <br />involving Chub, Rainbow, Cultural Sites, Sand Deposition, Recreation and Costs. See Table 3 in <br />Appendix D. <br /> <br />Options 7 and 10 (BHBF Anytime and P+SF IBH+ TCD8+Mech) consistently performed higher by <br />swing weighting than they did by direct ranks, suggesting that they may merit further <br />consideration. Options 9 and 11 consistently under-performed by swing weights relative to <br />direct ranks, suggesting that some participants may have anchored on these alternatives, <br />without adequate review of the estimated performance (or that some aspect of performance is <br />not captured - e.g., a missing attribute, a hidden threshold effect, unidentified competing <br />hypotheses) . <br /> <br />5.0 Next Steps and Outstanding Issues <br /> <br />The proposed next step is to review and refine the consequence table, the options and the <br />attributes, to have some dialogue about the relative importance assigned to each attribute, <br />and to repeat the weighting exercise. <br /> <br />15 <br />GCDAMP MATA: <br />December 2003 Workshop Report <br /> <br />02289 <br />