Examples of errors and biases in criminal justice process

[Elaedith]

I hope it's ok to put this here rather than education because I don't think anyone who might have suggestions will see it there.

I am redesigning a series of lectures which are taught to (mainly) criminal justice students and focus on the theme of explaining errors and biases within the justice system. I already have lectures covering forensic confirmation bias, eyewitness testimony, false confessions, detecting deception, offender profiling, and jury decisions. I would like a couple more topics although I can just split the existing ones across more than one lecture. I was considering biases about public perception of crime (where people think crime rates are always rising when they are not), but it doesn't really fit with the general theme that revolves around criminal justice process.

I was wondering, for example, about police biases that enter the investigative process and not already covered in other topics. So far I have found a bit of material on investigative decision biases that occur at 'tipping points' (such as decision to arrest) but haven't yet found a lot of material that fits together well. There needs to be a clear theme for each lecture or students get confused. This isn't about specific cases although they can be used as illustrations. I need to extract some type of issue that occurs across different investigations, and might be used to explain failures in the process. Is there anything obvious (or not so obvious) that is missing? Something that perhaps occurs across multiple cases of wrongful conviction, that I can look into and make into a clear topic?

 

[Chris_Halkides]

domain irrelevant information: DNA mixtures and hair comparisons

This is not exactly what you are looking for, but it is in the general area and may be helpful. There is something called "domain irrelevant information" that has the potential to bias a forensic investigator. There is an article in the forensic literature about DNA mixtures by Dror and Hampikian that was later summarized in The Economist magazine. "However, of the 17 examiners Dr Dror and Dr Hampikian approached—who, unlike the original two, knew nothing about the context of the crime—only one thought that the same suspect could not be excluded. Twelve others excluded him, and four abstained."

An example of this problem in the area of hair analysis was addressed in Roger Koppl's 2007 article CSI For Real. "Larry S. Miller demonstrates an excellent example of cognitive bias at play. He asked a group of 14 students trained in hair analysis, all of whom met the basic requirements for expert testimony on human hair identification in courts of law, to examine four cases each. For each student, two cases were presented the usual way: They were given two samples and told that one was from the crime scene and the other from the suspect. The other two cases were presented through a forensic line- up. The known sample from the imaginary crime scene was compared to five suspect-known hair samples. In all 56 cases, there were in reality no true matches. The first group of cases yielded an error rate of 30.8 percent; the second group an error rate of only 3.8 percent.12"

Perhaps more directly related to your interest is Chapter Twelve (on the decision to institute charges in a rape case) in the book Race to Injustice. The book is about the Duke lacrosse case, and this chapter is by Michael Siegel, the editor of this book. Mr. Siegel is especially interested in the question of whether or not the a grand jury could have served as a check on DA Nifong's highly questionable prosecution, as well as the more general question of whether or not grand juries should be reformed. The decision of the grand jury is a clear tipping point in the process.

 

[Chris_Halkides]

the proximity effect

One thing that crops up in accepted or possible wrongful convictions or accusations is proximity to the victim. The proximity may be because the person accused or convicted is the spouse of the victim (Michael Morton and David Kamm), the father of the victim (Billy Wayne Cope), or the roommate of the victim (Amanda Knox--who along with her boyfriend Raffaele Sollecito were present when Meredith Kercher's body was found). Again, I am not sure whether or not this is what you are seeking, but it might be that proximity is a biasing influence on the police, and one that might become more acute when there is pressure to solve a case quickly.

 

[Samson]

I was typing a lengthy post and accidentally deleted the lot, but it is about appeal court reasoning and focus in the Mark Lundy case and the Brendan Dassey case, so will redo when I get time.

 

[Vixen]

The police are by training proactively biased. A woman is raped and murdered. They go from door to door questioning all men in the neighbourhood.

They've assumed it was a guy who lived nearby.

How else are they supposed to do their job except by:

1. looking at similar historical cases and detecting a pattern (for example, most women are killed by their partner and there is often a pattern of domestic abuse)

2. Obtaining alibi and checking up on them

3. the last known person at the scene of the crime.

Sure, it's irritating to be considered a possible suspect when you are completely innocent but you accept that the police are not to know that, given the lengths a criminal will go to conceal their crime, mostly committed in conditions of privacy and secrecy.


It would be wasteful economically and in terms of time, and make the case unwieldy to suspect absolutely everybody, in the name of 'fairness' and 'lack of bias'.

 

[Elaedith]

This is not exactly what you are looking for, but it is in the general area and may be helpful. There is something called "domain irrelevant information" that has the potential to bias a forensic investigator. There is an article in the forensic literature about DNA mixtures by Dror and Hampikian that was later summarized in The Economist magazine. "However, of the 17 examiners Dr Dror and Dr Hampikian approached—who, unlike the original two, knew nothing about the context of the crime—only one thought that the same suspect could not be excluded. Twelve others excluded him, and four abstained."

An example of this problem in the area of hair analysis was addressed in Roger Koppl's 2007 article CSI For Real. "Larry S. Miller demonstrates an excellent example of cognitive bias at play. He asked a group of 14 students trained in hair analysis, all of whom met the basic requirements for expert testimony on human hair identification in courts of law, to examine four cases each. For each student, two cases were presented the usual way: They were given two samples and told that one was from the crime scene and the other from the suspect. The other two cases were presented through a forensic line- up. The known sample from the imaginary crime scene was compared to five suspect-known hair samples. In all 56 cases, there were in reality no true matches. The first group of cases yielded an error rate of 30.8 percent; the second group an error rate of only 3.8 percent.12"

Perhaps more directly related to your interest is Chapter Twelve (on the decision to institute charges in a rape case) in the book Race to Injustice. The book is about the Duke lacrosse case, and this chapter is by Michael Siegel, the editor of this book. Mr. Siegel is especially interested in the question of whether or not the a grand jury could have served as a check on DA Nifong's highly questionable prosecution, as well as the more general question of whether or not grand juries should be reformed. The decision of the grand jury is a clear tipping point in the process.

I already have some of this in the forensic confirmation bias lecture; we look at biasing effects on interpretation of fingerprint, DNA admixture and other types of evidence and the Brandon Mayfield case as an example. In fact, it was attending a seminar by Dror that sparked my interest in that area. Thanks for the additional references which I will look at. I also noticed that Dror and others have a lot of new material and it may be that I can extend this topic into two lectures as it's popular. I would just need to find some meaningful way to split the material into two sub-topics.

I should say that these are UK students and one complication is differences in legal procedures when referring to material from the US. This comes up also in the false confessions area where lots of published research is based on the US procedures. We don't have issues with Reid technique and police being allowed to lie to suspects here, but still use this material as a comparison (and because you can never be sure nobody will suggest adopting US approaches here in the future).

 

[Elaedith]

One thing that crops up in accepted or possible wrongful convictions or accusations is proximity to the victim. The proximity may be because the person accused or convicted is the spouse of the victim (Michael Morton and David Kamm), the father of the victim (Billy Wayne Cope), or the roommate of the victim (Amanda Knox--who along with her boyfriend Raffaele Sollecito were present when Meredith Kercher's body was found). Again, I am not sure whether or not this is what you are seeking, but it might be that proximity is a biasing influence on the police, and one that might become more acute when there is pressure to solve a case quickly.

Although there is probably not enough material on proximity alone to make that a topic, I could use this type of idea to form a lecture around biases that affect early stages of investigations.

I am thinking of the Claremont murders case which I am familiar with from living in the area at the time. The police narrowed down on the wrong suspect, Lance Williams, and although they could not get any credible evidence for a charge they persisted with him as a suspect for years. They kept him under surveillance and told the victims' families they were not looking for anyone else. Obviously confirmation bias comes into play here but I am interested in any additional factors that might affect why police would narrow down on a suspect so early and refuse to consider alternative hypotheses.

 

[Rolfe]

That's a very common situation I think, and if you've been reading the Luke Mitchell thread you'll see that was the fundamental problem in that case. I guess it is just a type of confirmation bias but it's such a major issue that I think it does deserve its own topic.

Identifying the wrong suspect at an early stage can bias the investigation beyond repair, as it causes the police to fail to collect evidence that doesn't fit their initial theory, and if that evidence is missed at the time it can be way too late for the defence to get hold of it. (An example of this from the Mitchell case is that the full records from Luke's mobile phone would have shown whether he was telling the truth about the arrangement being to meet at the west side of the path rather than the east, and might have given vital information that would have helped establish whether he actually did go to the east end of the path or not. By the time he was charged and a defence team was gathered together it was too late to get that data.)

 

[Rolfe]

Actually Sandra Lean has a book out called No Smoke which covers a number of these points. It isn't available as an eBook unfortunately but Amazon have it in hard copy.

 

[Chris_Halkides]

cajoling or coercing witnesses

One thing that shows up in known and putative wrongful convictions is that the witnesses changed their statements at some point. Sometimes this involves eyewitness testimony. Witnesses who are themselves potentially facing charges can be coerced, but witnesses who are not can be persuaded to do so for the purpose of obtaining a conviction. The Timothy Hennis case has at least one example of the latter. It seems to me that this alteration of might be done by police either consciously or unconsciously.

 

[Rolfe]

That was a serious issue in the Lockerbie case. An eyewitness who saw one of the terrorists gave about 21 statements to the police and over the course of these - and this developed further in court - his "recollection" got closer and closer to what the police wanted him to say. He even had a copy of an article in a magazine which helpfully detailed the discrepancies between the police case and his story, and the police only took it from him four days before he went to do the identification parade (yes it had a photo of the accused as well!)

Someone actually got his PhD analysing these statements and their discrepancies. https://ethos.bl.uk/OrderDetails.do?uin=uk.bl.ethos.678640

(I'm surprised they're asking for a scan fee to access this. I discovered a few years ago that my own PhD thesis which dates back to 1985 has been scanned and is free to download. Different university though.)

 

[Darat]

I hope it's ok to put this here rather than education because I don't think anyone who might have suggestions will see it there.

I am redesigning a series of lectures which are taught to (mainly) criminal justice students and focus on the theme of explaining errors and biases within the justice system. I already have lectures covering forensic confirmation bias, eyewitness testimony, false confessions, detecting deception, offender profiling, and jury decisions. I would like a couple more topics although I can just split the existing ones across more than one lecture. I was considering biases about public perception of crime (where people think crime rates are always rising when they are not), but it doesn't really fit with the general theme that revolves around criminal justice process.

I was wondering, for example, about police biases that enter the investigative process and not already covered in other topics. So far I have found a bit of material on investigative decision biases that occur at 'tipping points' (such as decision to arrest) but haven't yet found a lot of material that fits together well. There needs to be a clear theme for each lecture or students get confused. This isn't about specific cases although they can be used as illustrations. I need to extract some type of issue that occurs across different investigations, and might be used to explain failures in the process. Is there anything obvious (or not so obvious) that is missing? Something that perhaps occurs across multiple cases of wrongful conviction, that I can look into and make into a clear topic?

What "institutional" bias, so for example institutional racism. There was an inquiry in the UK that uncovered the institutional racism in the Metropolitan police the report is available as a PDF https://assets.publishing.service.g.../uploads/attachment_data/file/277111/4262.pdf

There is also the type of institutional bias that seeks to "protect" the justice system from criticism and scrutiny. A classic example was Lord Denning the Master of the Rolls (senior appeals judge in the English justice system): in summary 6 men "Birmingham 6" were convicted of a terrorist bombing, they were wrongly convicted (took a campaign spanning decades to right that wrong) yet this senior judge went on the record to say

We shouldn't have all these campaigns to get the Birmingham Six released if they'd been hanged. They'd have been forgotten, and the whole community would be satisfied... It is better that some innocent men remain in jail than that the integrity of the English judicial system be impugned.​

 

[Elaedith]

That's a very common situation I think, and if you've been reading the Luke Mitchell thread you'll see that was the fundamental problem in that case. I guess it is just a type of confirmation bias but it's such a major issue that I think it does deserve its own topic.

Identifying the wrong suspect at an early stage can bias the investigation beyond repair, as it causes the police to fail to collect evidence that doesn't fit their initial theory, and if that evidence is missed at the time it can be way too late for the defence to get hold of it. (An example of this from the Mitchell case is that the full records from Luke's mobile phone would have shown whether he was telling the truth about the arrangement being to meet at the west side of the path rather than the east, and might have given vital information that would have helped establish whether he actually did go to the east end of the path or not. By the time he was charged and a defence team was gathered together it was too late to get that data.)

I agree, although I'm having trouble finding a lot of academic research on this compared to other topics. There is some experimental cognitive research which tries to get at the underlying mechanisms, but is rather removed from actual police procedures. But I will make a topic on confirmation bias in investigation more generally and use this to introduce confirmation bias which comes up throughout the other topics, (forensic examination, interrogation etc).

 

[Elaedith]

What "institutional" bias, so for example institutional racism. There was an inquiry in the UK that uncovered the institutional racism in the Metropolitan police the report is available as a PDF https://assets.publishing.service.g.../uploads/attachment_data/file/277111/4262.pdf

There is also the type of institutional bias that seeks to "protect" the justice system from criticism and scrutiny. A classic example was Lord Denning the Master of the Rolls (senior appeals judge in the English justice system): in summary 6 men "Birmingham 6" were convicted of a terrorist bombing, they were wrongly convicted (took a campaign spanning decades to right that wrong) yet this senior judge went on the record to say

We shouldn't have all these campaigns to get the Birmingham Six released if they'd been hanged. They'd have been forgotten, and the whole community would be satisfied... It is better that some innocent men remain in jail than that the integrity of the English judicial system be impugned.​

We look at the Birmingham Six under false confessions, but I wasn't aware of that quote. I will add it in, should provoke some reaction.

 

[Nessie]

Although there is probably not enough material on proximity alone to make that a topic, I could use this type of idea to form a lecture around biases that affect early stages of investigations.

I am thinking of the Claremont murders case which I am familiar with from living in the area at the time. The police narrowed down on the wrong suspect, Lance Williams, and although they could not get any credible evidence for a charge they persisted with him as a suspect for years. They kept him under surveillance and told the victims' families they were not looking for anyone else. Obviously confirmation bias comes into play here but I am interested in any additional factors that might affect why police would narrow down on a suspect so early and refuse to consider alternative hypotheses.

I would suggest the Shirley McKie fingerprint case in Scotland, as an example of how the police decide on a suspect and then work the evidence to fit. She was a police officer, but that did not stop other police officers from deciding she was guilty and then working the evidence. Another is the murder of Joanna Yeates and how her landlord Christopher Jefferies became the prime suspect.

I would suggest the reasons why police will narrow in on a suspect as quickly as possible are;

- pressure from senior management to get cases solved as quickly as possible. Murder enquiries are expensive.

- pressure from the media. The media move fast and hate a slow story.

- pressure from the public. People get scared by serious crimes in their neighbourhood and want fast resolutions.

Then there are issues within the police itself from;

- arrogant cops who think they have a sense for who did it, the detective's hunch. If a senior and/or charismatic detective decides that so and so is the criminal, then the team follow and start to work the evidence to back that up. The idea that detective is wrong is hard to take and exculpatory tends to be ignored or buried.

- career cops who are after a detection for their next promotion. A detection is a detection and a good detection can lead to promotion. It does not matter if it turns out to be wrong, especially since that could take years and by then the cop has been promoted. They can always find reasons to say the jury was wrong and continue to believe the suspect was guilty.

- thick cops who just do not understand evidencing, beyond reasonable doubt, innocent until proven guilty, exculpatory evidence, rules of disclosure and the details of the law.

- lazy cops, who just want an easy life, so the first person to make a complaint is believed and the minimum of evidence is gathered.

- corrupt cops, who have been planted or been bribed to disrupt major enquiries or any enquiry into certain organised crime groups or crime lords.

 

[Elaedith]

That was a serious issue in the Lockerbie case. An eyewitness who saw one of the terrorists gave about 21 statements to the police and over the course of these - and this developed further in court - his "recollection" got closer and closer to what the police wanted him to say. He even had a copy of an article in a magazine which helpfully detailed the discrepancies between the police case and his story, and the police only took it from him four days before he went to do the identification parade (yes it had a photo of the accused as well!)

Someone actually got his PhD analysing these statements and their discrepancies. https://ethos.bl.uk/OrderDetails.do?uin=uk.bl.ethos.678640

(I'm surprised they're asking for a scan fee to access this. I discovered a few years ago that my own PhD thesis which dates back to 1985 has been scanned and is free to download. Different university though.)

The Lockerbie eyewitness evidence would be an ideal example as it illustrates everything not to do when collecting eyewitness testimony. There is a very simple article by Loftus which shows how most of the issues that we look at apply in the case. It's just a matter of squeezing it in as eyewitness evidence is already a huge topic.

 

[thines]

Biases.

In the book "The Psychology and Sociology of Wrongful Convictions" edited by W. Koen and C. M. Bowers (Academic Press, 2018) chapter 7 by Koen is titled "Confirmation Bias in Forensic Science." It's an excellent chapter in an excellent book.

 

[Elaedith]

In the book "The Psychology and Sociology of Wrongful Convictions" edited by W. Koen and C. M. Bowers (Academic Press, 2018) chapter 7 by Koen is titled "Confirmation Bias in Forensic Science." It's an excellent chapter in an excellent book.

Thanks, I will put it on the reading list.

 

[Chris_Halkides]

Forensic Science Reform

The book Forensic Science Reform Protecting the Innocent was also edited by Wendy Koen and C. Michael Bowers. Each chapter focuses on a different discipline and has a case study. Link

 

[psionl0]

I would suggest the Shirley McKie fingerprint case in Scotland, as an example of how the police decide on a suspect and then work the evidence to fit. She was a police officer, but that did not stop other police officers from deciding she was guilty and then working the evidence. Another is the murder of Joanna Yeates and how her landlord Christopher Jefferies became the prime suspect.

I would suggest the reasons why police will narrow in on a suspect as quickly as possible are;

- pressure from senior management to get cases solved as quickly as possible. Murder enquiries are expensive.

- pressure from the media. The media move fast and hate a slow story.

- pressure from the public. People get scared by serious crimes in their neighbourhood and want fast resolutions.

Then there are issues within the police itself from;

- arrogant cops who think they have a sense for who did it, the detective's hunch. If a senior and/or charismatic detective decides that so and so is the criminal, then the team follow and start to work the evidence to back that up. The idea that detective is wrong is hard to take and exculpatory tends to be ignored or buried.

- career cops who are after a detection for their next promotion. A detection is a detection and a good detection can lead to promotion. It does not matter if it turns out to be wrong, especially since that could take years and by then the cop has been promoted. They can always find reasons to say the jury was wrong and continue to believe the suspect was guilty.

- thick cops who just do not understand evidencing, beyond reasonable doubt, innocent until proven guilty, exculpatory evidence, rules of disclosure and the details of the law.

- lazy cops, who just want an easy life, so the first person to make a complaint is believed and the minimum of evidence is gathered.

- corrupt cops, who have been planted or been bribed to disrupt major enquiries or any enquiry into certain organised crime groups or crime lords.

That explains the real problem here. It is not about biases so much as pressure.

It is easier to build a case against somebody (anybody) than to track down the real culprit.