References in periodicals archive ?
The first is that judges will be "unaffected" (IPCI = 0).
The evidence was twenty times more consistent with the "judge IPCI = 0" hypothesis than the "judge IPCI = 10%" hypothesis.
The average lawyer IPCI was neither practically nor meaningfully different from zero (2%, [+ or -] 14%).
In both problems, the predicted student IPCI was greater than zero, and on average it was 12% ([+ or -] 14%).
More information can be extracted from the results, however, if one computes the likelihood ratios for competing hypotheses about the size of the "true" student IPCI. Figure 11 indicates that the experimental evidence is very slightly--1.5 times--more consistent with the hypothesis that the "true" student IPCI is "as large as" the public IPCI that with the "student IPCI = 0" hypothesis.
Figure 12 illustrates the use of this method to test competing hypotheses about the lawyer IPCI. The weight of the evidence against the hypothesis that lawyers will be affected as much as the public is quite strong: the "lawyer IPCI = 0" hypothesis is over 100 times more consistent with the evidence.
Generating a likelihood ratio of very close to one, the experimental results are effectively equally consistent with the "lawyer IPCI = 0" and "lawyer IPCI = 5%" hypotheses.
Neither the judge IPCI (-5%, [+ or -] 12%) nor the lawyer IPCI (2%, [+ or -] 14%) is statistically or practically different from zero.
Caption: Figure 8: Assessing the Weight of the Evidence: Competing Hypotheses for Public IPCI
Caption: Figure 9: Judge IPCI: Evidentiary Weight of Experimental Data
Caption: Figure 10: Lawyers & Law Students: Estimated IPCI
Caption: Figure 11: Student IPCI: Evidentiary Weight of Experimental Data
Acronyms browser ?
Full browser ?
- Ipcc Enterprise
- IPCC Fourth Assessment Report
- IPCC Policymakers Summary
- IPCC TAR Summary Conflict