Bias

Question 1

What can influence evaluation?

Answer 1

1. Context:
   - Corroborative inflation/bias snowball effect/investigative echo chamber
   - Confirmation bias
   - Forensic confirmation bias  Belief in guilt influences collection, perception and interpretation of evidence
2. Human:
   - Hindsight bias
   - Primacy effect
   - Elasticity of forensic evidence
   - Overreliance on salient data
   - Under-utilization of base-rates
   - Failure to analyse co-variation
   - Limited amount of information can be takin in  The more data considered, the less accurate the judgement
   - No evidence that experience increases accuracy
   - Barnum effect

Question 2

How can these be remedied?

Answer 2

1. Work linear, not circular
2. Blind-testing
3. Double-blindness
4. Line-up of evidence
5. Use technology
6. Train judicial decision-makers
7. Randomize lists of suspects
8. Know about bias
9. Avoid premature abandonment of useful decision rules
10. Do not focus on salient material
11. List alternative options and look for their evidence
12. Use base-rates
13. List contradictory evidence
14. Make a deliberate effort to obtain feedback

Question 3

Biases in CT scans

Answer 3

1. More influence when info is suggestive and CT scan is ambiguous
2. Suggestive info made it seemingly easier for experts to understand/interpret CT scans

Question 4

Clinical v. actuarial judgement

Answer 4

1. Actuarial:
   - Conclusions are based strictly on empirical data and established relations between data and criterion
   - Fixed, structured and standardized
   - More accurate
   - ‘Broken leg’ issue
   - Automatic/mechanical
2. Clinical:
   - Estimation of risk based on personal judgement
   - Malleable depending on situation
   - Not as accurate especially when too much data is involved
   - Prone to human errors
   - Does not detect ‘broken legs’