Op-Ed in Post-Gazette: Why Forensic Analysis of Crime Scenes is not as Reliable as you Think

SAMSI was featured in an op-ed piece in the Post-Gazette that was written by Lucas Mentch, Maria Cuellar, William C. Thompson and Clifford Spiegelman, all whom are participating in the SAMSI program on forensics this year.

The piece focuses on the Netflix mini-series, “Making a Murderer,” that raised questions about the actions and motives of law enforcement. Read their piece here.

Five things I learned at the Forensics Tutorial

fingerprint analysis lab

Participants got to try their hand at doing a fingerprint analysis.

I don’t know about you, but I am fascinated by forensics. Crime scenes are puzzles waiting for someone to solve. So, I, along with some other staff members, sat in on part of the Forensics tutorial held in August.

Here are five things I learned while I listened in on the tutorial:

1)Unlike “CSI” or whatever your favorite cop show might be (mine is “Dexter”), the methods used at the crime scene are not nearly as scientific as you would expect. That’s where statisticians and applied mathematicians come in. There is a plethora of areas where adding statistical or mathematical methodologies to the field of forensics can make a tremendous impact in helping law enforcement catch the right people.

2)A lot of forensic evidence comes from inductive inferential processes, which is not a very good/reproducible way to gather evidence. Bill Tobin, from Forensic Engineering International, told us that they do not share error rates and there is no “peer review” for the evidence.See one of his presentations here.

3)There are many factors that are often overlooked or are not explored in firearm toolmark evidence (Firearm toolmarks are used to associate or eliminate a particular firearm as a murder weapon comparing characteristics imported to bullets and cartridge cases when it cycles through a gun.) but there are ways that forensic experts could use a more scientific approach to make this a more credible piece of evidence. See Cliff Spiegelman’s talk for more information about this.

Andy Parker dusting for fingerprints

Andy Parker shows the group how he dusts for footprints.

4)Andy Parker from the Wake County Crime Bureau Investigation team is awesome! He talked about what his job entailed and showed us several slides (sorry, they cannot be shared) of dead bodies and had us guess if it was a murder, suicide or death by accident. Then he explained how one could conclude which was correct. Later, he had us all gather around and showed us how they dust for prints, including footprints. It was really neat!

Herbert David Sheets

Herbert David Sheets, Canisius College and University at Buffalo, talks about bitemarks.

5)The uniqueness of bite marks is not been scientifically established, and the uniqueness of a bite mark to make a unique pattern in someone’s skin has also not been scientifically established. See Herbert David Sheets’ presentation for more info.

Overall, this is a fascinating topic that applied mathematicians and statisticians have traditionally not been involved with, but that needs your attention. Saving an innocent person from being in prison and/or catching a criminal who may still be free is vital to keeping our society whole. You can still get involved by attending one of SAMSI’s subsequent workshops which will be posted on the SAMSI website sometime in the near future.