Thursday, April 3, 2008

Junk Science

The Blog article about Dr. DeVore includes this statement:

Only Dr. Devore, by discrediting all the other information provided to him, and by using a very questionable technique and "junk science" was able to give the prosecution exactly what they wanted. He was a very important part of a prosecution strategy that sent an innocent man to death row.

One of the comments to the article asks this question:
Please describe in your own words what the term "junk science" means to you.

In my own words I would describe junk science as the use of unproven techniques and the dishonest manipulation of facts and figures to obtain a required result.

Another definition from is faulty scientific data and analysis used to advance special and, often, hidden agendas.

SII, including information from several sources, discusses junk science:

Junk Science is one of the major factors in wrongful convictions.

Its deadly potential comes from the trust the Courts and the Jurors place in "expert testimony," when they generally can't tell a good expert from a bad expert.

Junk Science results from a number of factors:

The theory or method is untested.

Peer review is generally the way that new theories and methods undergo the critical analysis necessary
to ensure they are scientifically sound.

The theory or method is scientifically sound, but the expert using the theory or method is not sufficiently skilled in its use, or is careless in its use.

The theory or method is scientifically sound, but the results are exaggerated to mean much more than they really do.

Experts yield to the pressure to please their client, to interpret the data to produce the results the client wants.

The general public seems content with the myth that "paid experts,"i.e., experts that are paid to deliver the desired outcome, result only from rich defendants who can afford to pay for such, and that every expert who testifies for the State is well-qualified, highly ethical, and only wants to present the truth.

Another comment on the blog notes that junk science does not follow the Daubert requirements for scientific evidence:

* The theory or technique must be non falsifiable, non refutable, and testable.
* The theory or technique must have been subjected to peer review and publication.
* The theory or technique must have a known or quantifiable error and to be valid requires the existence and maintenance of standards concerning its operation.
* The theory or technique must be generally accepted by a relevant scientific community.

Junk science is considered a major factor in wrongful convictions by the Center on Wrongful Convictions at Northwestern University School of Law, and by the Innocence Project.

The Center on Wrongful Convictions of Northwestern University School of Law identifies four types of false or unreliable evidence :
~ False testimony by informant or "snitch" witnesses
~ Incorrect eyewitness identification
~ False confessions
~ False or unreliable forensic evidence or "junk science"

The Innocence project
Since forensic evidence is offered by "experts," jurors routinely give it much more weight than other evidence. But when misconduct occurs, the weight is misplaced.

If you would like to continue a discussion about this blog article instead of posting a comment, please sign up at:

Tuesday, April 1, 2008

Dr. DeVore's Curious Calculations

Dr. Brian Peterson, the forensic pathologist who performed autopsies for Contra Costa County, testified at Scott Peterson’s preliminary hearing that he thought Conner Peterson was a 9 month/full term fetus, based on a crown rump measurement of 32 centimeters. However, he knew that this measurement was just an estimate, and for a more accurate assessment of the age of the fetus at the time of death he consulted Dr. Alison Galloway, a forensic anthropologist.

Dr. Galloway concluded that Conner was 35-36 weeks old at the time he died. With a standard deviation of 2 weeks plus or minus, she said the earliest he died was 33 weeks, the latest 38 weeks. To reach her conclusions Dr. Galloway made 11 measurements: length of the bones from the upper leg (femur), the lower leg (tibia), the upper arm (humerus), the parietal cord (height, perimeter height, cord width, perimeter width), right frontal cord (height, perimeter height, cord width, perimeter width); and she verified the age range with well accepted formulas from three different textbooks.

At the time of Laci’s last prenatal visit to the Hera Medical Group on December 23, 2002, Conner’s age was 32 weeks 1 day. (Dr. Tow-Der, rounding the number off, noted 32 weeks on Laci’s medical chart). At the time of the second ultrasound on September 24, Laci’s due date had been revised from February 10 to February 16. Several measurements made by Dr. Yip at that time (abdominal circumference, head circumference, biparietal diameter and femur length) indicated that Conner was 6 days younger than originally estimated.

The calculations of Dr. Galloway, who testified for the prosecution, indicated that Conner had died on December 29 (fetus age 33 weeks) at the very earliest.

This was not what the prosecution wanted to hear; so they went looking for another expert to support their theory that Scott Peterson had killed his wife between 8:30 p.m. on December 23 and 10:08 a.m. on December 24.

And they found Dr. Greggory DeVore, who was willing to see things their way, and who was willing to manipulate numbers and information to come up with the calculations the prosecution required.

Dr. DeVore was hired by the prosecution in February 2004. On February 8, he went to the Contra Costa County Coroner’s office where he was given Conner’s femur bone. He placed the bone on a cup in a water bath (like a fish tank) and measured it 3 times with his portable ultrasound machine. This was the only test he did. On the basis of this, he produced a report on February 19 which stated unequivocally that Conner had died on December 25. The prosecution did not like this estimate either.

So, on the day he testified, Dr. DeVore changed his story. He said that December 25 was a mistake, that he had been adding up the numbers on his hands.

In testimony, Dr. DeVore insisted that December 23 was the date of Conner’s death. Although he said he had created a spread sheet to verify this, the only thing he provided in court to support his theory was a free hand graph. His miraculous calculation, based on measuring one bone 3 times in a tank of water, was exactly the number the prosecution required. To arrive at this number, Dr. DeVore completely dismissed much of the information from Laci’s doctors including the revised due date, and discounted the opinions of the forensic pathologist, the forensic anthropologist and various ultrasound and x-ray experts. He relied on the measurement of only one bone—a scientifically unsound method, and chose the only formula which could match the date he wanted.

Dr. Charles March, testifying for the defense, considered all the information available to him: doctor’s records and information, autopsy records, forensic analysis, interviews with family and friends and he agreed with the calculations of Dr. Galloway and Dr. Yip. He reached the conclusion that the earliest date Conner could have died was December 29. One of the things he considered was the date of Laci’s first positive pregnancy test, June 9, 2002. This date was confirmed by family members and friends in investigative reports and in an interview with Katie Couric on the year anniversary, June 9, 2003. The date of Laci’s first positive pregnancy test was 6 days later than might have been expected, and this was verified by her second ultrasound and the decision of Laci’s doctors to change her due date from Feb 10 to Feb 16.

Only Dr. Devore, by discrediting all the other information provided to him, and by using a very questionable technique and “junk science” was able to give the prosecution exactly what they wanted. He was a very important part of a prosecution strategy that sent an innocent man to death row.

To see a video of Dr. DeVore talking with Greta Van Susteren “On the Record,” December 17, 2004, go to the following link. The other videos on this link show the strong media attention given to Dr. DeVore’s testimony and the very superficial analysis done by most of the reporters.