Results and summary tables for the Bunch study, with only independent pairwise comparisons included. 95% bootstrap intervals are provided for quantities estimated via simulation. To allocate the inconclusives and eliminations, we used the rule stated in Bunch and Murphy (2003): FBI examiners can exclude only on class characteristic mismatches. It should be noted that in none of the 287 536 simulations which matched the other set criteria did we find a set which had more than 109 class characteristic mismatches, suggesting that there may have been some eliminations made on individual characteristics, but as this cannot be confirmed, estimates of eliminations only include class characteristic mismatches
| A comprehensive validity study for the forensic examination of cartridge cases Bunch and Murphy (2003) . | ||||
|---|---|---|---|---|
| Study type . | Test set . | Participants . | . | |
| . | # Knowns . | # Unknowns . | . | . |
| Open set | variable | variable | 8 FBI examiners | |
| Independent comparisons | ||||
| Mean (95% CI) | ||||
| Identification | Inconclusive | Elimination | Source Total | |
| Same source | 28.7 [26, 31] | 0 | 0 | 28.7 [26, 31] |
| Diff source | 0 | 156.5 [134, 181] | 42.5 [22, 64] | 199.1 [184, 217] |
| Conclusion Total | 28.7 [26, 31] | 156.5 [134, 181] | 42.5 [22, 64] | |
| Independent comparisons: conclusion-specific probabilities | ||||
| Identification | Inconclusive | Elimination | Source Probability | |
| Same source | 1.000 | 0.000 | 0.000 | 0.126 [0.110, 0.142] |
| Diff source | 0.000 | 1.000 | 1.000 | 0.874 [0.858, 0.890] |
| Conclusion probability | 0.126 [0.110, 0.142] | 0.687 [0.597, 0.779] | 0.187 [0.095, 0.278] | |
| Independent comparisons: source-specific probabilities | ||||
| Identification | Inconclusive | Elimination | Source probability | |
| Same source | 1.000 | 0.000 | 0.000 | 0.126 [0.110, 0.142] |
| Diff source | 0.000 | 0.787 [0.682, 0.891] | 0.214 [0.109, 0.318] | 0.874 [0.858, 0.890] |
| Conclusion probability | 0.126 [0.110, 0.142] | 0.687 [0.597, 0.779] | 0.187 [0.095, 0.278] | |
| Independent comparisons: error rates | ||||
| Opt. | Meaning | Missed identification | Missed elimination | Total |
| 2 | FTE error | 0.000 | 0.000 | 0.000 |
| 3 | Process error | 0.000 | 0.787 [0.682, 0.891] | 0.687 [0.587, 0.779] |
| A comprehensive validity study for the forensic examination of cartridge cases Bunch and Murphy (2003) . | ||||
|---|---|---|---|---|
| Study type . | Test set . | Participants . | . | |
| . | # Knowns . | # Unknowns . | . | . |
| Open set | variable | variable | 8 FBI examiners | |
| Independent comparisons | ||||
| Mean (95% CI) | ||||
| Identification | Inconclusive | Elimination | Source Total | |
| Same source | 28.7 [26, 31] | 0 | 0 | 28.7 [26, 31] |
| Diff source | 0 | 156.5 [134, 181] | 42.5 [22, 64] | 199.1 [184, 217] |
| Conclusion Total | 28.7 [26, 31] | 156.5 [134, 181] | 42.5 [22, 64] | |
| Independent comparisons: conclusion-specific probabilities | ||||
| Identification | Inconclusive | Elimination | Source Probability | |
| Same source | 1.000 | 0.000 | 0.000 | 0.126 [0.110, 0.142] |
| Diff source | 0.000 | 1.000 | 1.000 | 0.874 [0.858, 0.890] |
| Conclusion probability | 0.126 [0.110, 0.142] | 0.687 [0.597, 0.779] | 0.187 [0.095, 0.278] | |
| Independent comparisons: source-specific probabilities | ||||
| Identification | Inconclusive | Elimination | Source probability | |
| Same source | 1.000 | 0.000 | 0.000 | 0.126 [0.110, 0.142] |
| Diff source | 0.000 | 0.787 [0.682, 0.891] | 0.214 [0.109, 0.318] | 0.874 [0.858, 0.890] |
| Conclusion probability | 0.126 [0.110, 0.142] | 0.687 [0.597, 0.779] | 0.187 [0.095, 0.278] | |
| Independent comparisons: error rates | ||||
| Opt. | Meaning | Missed identification | Missed elimination | Total |
| 2 | FTE error | 0.000 | 0.000 | 0.000 |
| 3 | Process error | 0.000 | 0.787 [0.682, 0.891] | 0.687 [0.587, 0.779] |
Results and summary tables for the Bunch study, with only independent pairwise comparisons included. 95% bootstrap intervals are provided for quantities estimated via simulation. To allocate the inconclusives and eliminations, we used the rule stated in Bunch and Murphy (2003): FBI examiners can exclude only on class characteristic mismatches. It should be noted that in none of the 287 536 simulations which matched the other set criteria did we find a set which had more than 109 class characteristic mismatches, suggesting that there may have been some eliminations made on individual characteristics, but as this cannot be confirmed, estimates of eliminations only include class characteristic mismatches
| A comprehensive validity study for the forensic examination of cartridge cases Bunch and Murphy (2003) . | ||||
|---|---|---|---|---|
| Study type . | Test set . | Participants . | . | |
| . | # Knowns . | # Unknowns . | . | . |
| Open set | variable | variable | 8 FBI examiners | |
| Independent comparisons | ||||
| Mean (95% CI) | ||||
| Identification | Inconclusive | Elimination | Source Total | |
| Same source | 28.7 [26, 31] | 0 | 0 | 28.7 [26, 31] |
| Diff source | 0 | 156.5 [134, 181] | 42.5 [22, 64] | 199.1 [184, 217] |
| Conclusion Total | 28.7 [26, 31] | 156.5 [134, 181] | 42.5 [22, 64] | |
| Independent comparisons: conclusion-specific probabilities | ||||
| Identification | Inconclusive | Elimination | Source Probability | |
| Same source | 1.000 | 0.000 | 0.000 | 0.126 [0.110, 0.142] |
| Diff source | 0.000 | 1.000 | 1.000 | 0.874 [0.858, 0.890] |
| Conclusion probability | 0.126 [0.110, 0.142] | 0.687 [0.597, 0.779] | 0.187 [0.095, 0.278] | |
| Independent comparisons: source-specific probabilities | ||||
| Identification | Inconclusive | Elimination | Source probability | |
| Same source | 1.000 | 0.000 | 0.000 | 0.126 [0.110, 0.142] |
| Diff source | 0.000 | 0.787 [0.682, 0.891] | 0.214 [0.109, 0.318] | 0.874 [0.858, 0.890] |
| Conclusion probability | 0.126 [0.110, 0.142] | 0.687 [0.597, 0.779] | 0.187 [0.095, 0.278] | |
| Independent comparisons: error rates | ||||
| Opt. | Meaning | Missed identification | Missed elimination | Total |
| 2 | FTE error | 0.000 | 0.000 | 0.000 |
| 3 | Process error | 0.000 | 0.787 [0.682, 0.891] | 0.687 [0.587, 0.779] |
| A comprehensive validity study for the forensic examination of cartridge cases Bunch and Murphy (2003) . | ||||
|---|---|---|---|---|
| Study type . | Test set . | Participants . | . | |
| . | # Knowns . | # Unknowns . | . | . |
| Open set | variable | variable | 8 FBI examiners | |
| Independent comparisons | ||||
| Mean (95% CI) | ||||
| Identification | Inconclusive | Elimination | Source Total | |
| Same source | 28.7 [26, 31] | 0 | 0 | 28.7 [26, 31] |
| Diff source | 0 | 156.5 [134, 181] | 42.5 [22, 64] | 199.1 [184, 217] |
| Conclusion Total | 28.7 [26, 31] | 156.5 [134, 181] | 42.5 [22, 64] | |
| Independent comparisons: conclusion-specific probabilities | ||||
| Identification | Inconclusive | Elimination | Source Probability | |
| Same source | 1.000 | 0.000 | 0.000 | 0.126 [0.110, 0.142] |
| Diff source | 0.000 | 1.000 | 1.000 | 0.874 [0.858, 0.890] |
| Conclusion probability | 0.126 [0.110, 0.142] | 0.687 [0.597, 0.779] | 0.187 [0.095, 0.278] | |
| Independent comparisons: source-specific probabilities | ||||
| Identification | Inconclusive | Elimination | Source probability | |
| Same source | 1.000 | 0.000 | 0.000 | 0.126 [0.110, 0.142] |
| Diff source | 0.000 | 0.787 [0.682, 0.891] | 0.214 [0.109, 0.318] | 0.874 [0.858, 0.890] |
| Conclusion probability | 0.126 [0.110, 0.142] | 0.687 [0.597, 0.779] | 0.187 [0.095, 0.278] | |
| Independent comparisons: error rates | ||||
| Opt. | Meaning | Missed identification | Missed elimination | Total |
| 2 | FTE error | 0.000 | 0.000 | 0.000 |
| 3 | Process error | 0.000 | 0.787 [0.682, 0.891] | 0.687 [0.587, 0.779] |
This PDF is available to Subscribers Only
View Article Abstract & Purchase OptionsFor full access to this pdf, sign in to an existing account, or purchase an annual subscription.
