In 1976, Ollie George, a 19-year-old college student drove her brother’s car on a routine shopping trip.  After shopping, Ollie could not start the car.  She phoned her family, but by the time her family arrived atto the shopping center, Ollie was gone.  Two days later, Ollie’s body was found in a rural area of Sacramento County.  She had been raped and murdered.  After a lengthy investigation that provided very little evidence, the police ended the hunt. 

Several years later, Dennis Louis Nelson was convicted for the rape of several women, and his DNA profile was then entered into the state’s convicted offender database. 

In 2001, California allocated money to Sacramento County that was earmarked for DNA testing to solve sexual assault cases.  After developing a DNA profile from semen found on Ollie’s sweater, the police compared the profile with DNA profiles in California’s convicted offender database.  Dennis Nelson’s  profile matched the DNA profile found on Ollie’s sweater. 

In the case against Nelson for Ollie’s murder, the prosecution’s experts estimated the likelihood of a person from the general population having that particular DNA profile was one in 930 sextillion (93 followed by 22 zeros).  Nelson argued that the DNA evidence was inadmissible, because the method used to calculate the likelihood of the match had not achieved scientific acceptance standard under People v. Kelly.  The court agreed with the prosecution, finding that the “‘rarity statistic”’ was admissible. 

Once such a match is found, forensic investigators use the method known as the “product rule” to calculate the likelihood of the sample coming from the relevant population.  Nelson acknowledges that the product rule is an acceptable technique, but he argues that when it is applied to a finite database of individuals for whom the state already has DNA data, the product rule is deceiving.  He asserts that running DNA against a group of individuals presupposes the product rules use of randomly selected individuals.  In other words, the likelihood of finding a DNA match, if compared to everyone in the world, would at most be the number of people on the earth, which is seven billion (7 with 9 zeros)–the logic being that somebody had to do it.  To conclude that the likelihood of a match is 930 sextillion, then, Nelson argues is ridiculous. 

Last month, the California Supreme Court rejected this notion.  The court found that the Kelly test measures reliability and that scientists agree that the product rule is reliable, even though they disagree about its relevance.  In short, the rarity statistic was admissible, at least in terms of reliance. 

Even if evidence is matched within a compiled database, post-Nelson courts will admit DNA evidence.  The product rule measures rarity, and the fact that the DNA profile from the crime scene was randomly run through a DNA profile database does not affect the evidence’s admissibility.  This is a huge success for California’s prosecutors, who can now bolster their cases by citing the rarity statistic even though a DNA profile is run through a finite database.

To read the recent CA Supreme Court decision, you can download a Word document here from the Court’s website.

-Seth Erickson

Comments are closed.