ch 4 (MT 2)

Question 1

what is forensic science (Canadian Society of Forensic Science) + 3 caveats?

Answer 1

“application of science and the scientific method to the judicial system”

1. not all forensic sciences have scientific foundations
2. all forensic sciences req human decision making
3. pattern matching disciplines req subjective judgment

Question 2

what % of first 250 DNA exonerations used forensic evidence? (Garret, 2011)

Answer 2

74%

Question 3

what are pattern matching disciplines? what principle was it derived from?

Answer 3

-compare imprints left by perpetrator to known imprints of a suspect
-Locard’s Exchange Principle: criminals leave trace of themselves at crime scene + take trace of crime scene with them

Question 4

what did the National Academy of Sciences Report (2009) say about forensic sciences + their reliability?

Answer 4

-many were based on experience, not science
-limited peer reviewed studies
-limited (if any) data on error rates

-reliability of many disciplines questioned: only nuclear DNA analysis capable of individualization

Question 5

what is individualization? why is DNA capable of it?

Answer 5

-testifying that a forensic science technique has produced a 1-1 link between the defendant + trace evidence from the crime scene

-sharing DNA is rare
-population stats are available for DNA

Question 6

what are population statistics?

Answer 6

the prevalence of an attribute in a reference group
-(eg: 37% of all humans have O+ blood)

Question 7

what is random match probability?

Answer 7

-the risk of trace evidence matching an innocent person by chance
-eg: O+ blood found at crime scene, perp is a White man
-population stats: 33% of White men have O+ blood
-random match probability is 1/3

Question 8

what is DNA profiling?

Answer 8

-comparing trace DNA to DNA of known source
-done by counting # of tandem repeats (short lengths of DNA that repeat within a gene)
-most of DNA strand is identical across people, but tandem repeats consistently vary across people at certain locations
-population stats available for tandem repeats

Question 9

what did the La Trobe study find?

Answer 9

-considerable variability in error rates for document examination examiners
-477 examiners, overall error rate 3-4%
-one analyst: 0% error rate for genuine, disguised, simulated signatures
-another: 67% for genuine, 91% for disguised, 0% for simulated

Question 10

What happened in the Robert Lee Stinson case? (why was bite mar testimony admitted? what went wrong in the case?)

Answer 10

-precedent – admitted in 19 jurisdictions
-evidence needed to only be “relevant and useful”
-expert witnesses were recognized odontologists

-initial sketch didn’t match Stinson’s teeth, but sketch wasn’t admitted
-no one asked about error rates - scientific sounding methods considered evidence of reliability
-odontologist falsely claimed bite mark was conclusive

Question 11

what are 3 problems with bite mark testimony?

Answer 11

1. no evidence surface of teeth are unique
2. no evidence skin records unique bite marks
3. no evidence bite marks can be matched to teeth

Question 12

what is hair microscopy + 3 problems with it?

Answer 12

-compare hairs under microscope, used before dev of DNA analysis
1. population stats on hair variability don’t exist
2. great variability in hair from one person
3. subjective

Question 13

how many DNA exonerations (of first 329) involved faulty hair evidence? what is the Santae Tribble case?

Answer 13

-74 / 329 (~23%)

-FBI claimed RMP was 1/10Mil - DNA test later showed one hair wasn’t even human

Question 14

what are 2 types of fingerprint comparison?

Answer 14

1. identity verification: 2 prints obtained in controlled conditions
2. crime scene identification (known print vs crime scene print):
   -one print (known) obtained in controlled conditions
   -other recovered from crime scene (latent)

Question 15

what is the ACE-V method of fingerprint examination?

Answer 15

A- analyze: are prints suitable for comparison?
C - compare: how similar?
E - evaluate: match, mismatch, inconclusive?
-
V - verification: 2nd opinion

Question 16

what happened in the Brandon Mayfield case (2004)?

Answer 16

-Mayfield recently converted to Islam, was on FBI watch list
-3 FBI examiners claimed crime scene print was 100% match to his
-Spanish authorities later discovered the actual bomber

Question 17

what are 3 problems with fingerprint examination?

Answer 17

1. ACE-V is subjective + error ctrl methods are inadequate
2. limited data on examiner reliability: FBI study found that 1/1000 positive matches are errors
3. No population stats for partial print patterns

Question 18

What are the findings of Dror, Charlton, Peron’s Fingerprint Bias Study (2009)?

Answer 18

-took a pair of fingerprints examiners had in the past said were a match
-created non matching context (leading them to believe fingerprints weren’t a match)
–> 20% accurate: 1/5 examiners stuck to prev decision
–> 80% inconsistent: 3/5 said it wasn’t, 1/5 said they couldn’t tell

Question 19

Discuss reliability + bias in DNA profiling

Answer 19

-under ideal conditions, DNA evidence is highly reliable
-IRL, they’re not

-susceptible to bias: when results are ambiguous, judgment largely depends on interpretation by a human analyst

Question 20

what are 5 recommendations for forensic science?

Answer 20

1. minimize irrelevant contextual info
2. separate crime labs + law enforcement
3. meaningful proficiency tests for forensic analysts
4. only admit forensic experts in court if reliability of testimony is known
5. create a national agency to regulate the forensic sciences

Question 21

a) Define the following three terms: individualization, population statistics, random match probability (3 points)

Answer 21

-individualization: forensic science technique produces a 1:1 link between defendant + trace evidence from a crime scene
-population statistics: the prevalence of an attribute in a reference group
-random match probability: the risk of trace evidence matching an innocent person by chance

Question 22

b) Discuss individualization, population statistics, and RMP in the context of DNA Profiling, Forensic Odontology, Hair Microscopy, and Fingerprint Examination (4 points)

Answer 22

1. DNA profiling: compares trace DNA to the DNA of a suspect, which is done by counting the tandem repeats (how often they occur in known areas of variability). This is highly reliable under the right conditions, and is capable of individualization since a) sharing DNA is rare and b) population statistics are available for DNA. the random match probability is low
2. forensic odontology: individualization is difficult bc there’s no evidence that surfaces on teeth are unique (no population statistics); therefore can’t determine random match probability
3. hair microscopy: individualization is difficult because there’s no population statistics on hair variability; random match probability
   -fingerprint examination: individualization difficult bc we don’t actually know that fingerprints are all unique; no population statistics for partial print patterns; random match = can partially match to someone else in the population

Question 23

c) Describe any additional problems (i.e., beyond individualization, population statistics, and random match probability) in the areas of Forensic Odontology, Hair Microscopy, and Fingerprint Examination (7 points)

Answer 23

-forensic odontology:
-no evidence that skin records unique bite marks
-no evidence that bite marks can be matched to teeth
-high false positive rate
-“6 degrees of certainty” is recommended; however there are no precise standards for what each degree is
-hair microscopy:
-substantial variability in hair features from the same person
-subjective
-many wrongful convictions
-fingerprint examination:
-threshold of how similar 2 fingerprints are is often “intuitive” → bias
-limited data on fingerprint examiner reliability
-many fingerprints might partially match another
-prints often smudged/distorted