Erroneous Convictions in Criminal Justice
Interview with Jon Gould, Ph.D., Director of the Washington Institute for Public and International Affairs Research, American University.
Dr. Gould discusses:
- Bottom line findings from the study "Predicting Erroneous Convictions: A Social Science Approach to Miscarriages of Justice"
- Ten statistically significant factors related to wrongful convictions
- The role of systemic error and tunnel vision
What does your recent study tell us about errors in the criminal justice system?
Jon Gould I’ve been studying wrongful convictions now for almost a decade and one of the things that continues to concern me is that the criminal justice system does not do what we expect and have happen in other professions and that is look at these problems after they have occurred. So, those of us who take airplanes can be secure in the knowledge that if anything happens on a plane, that the National Transportation Safety Board is going to investigate what happened to make sure that that error doesn’t happen again.
Doctors do this, too, in hospitals. If there is an error, doctors convene a group to figure out what went wrong to prevent it. Only in the criminal justice system, where the stakes are so high, do we not see professionals come together when there are errors to try to learn what happened to prevent it. Not to point fingers, not for legal liability, simply to make sure that we have the best criminal justice system we can have and to make sure that the innocent aren’t being convicted and that the guilty are being convicted.
If there’s anything that this research can help move forward, I hope it is that. That, if we are truly committed to professionals in the criminal justice system, they need to be willing to look at when errors occurred and try to rectify them.
What was the research project?
So, this is social science research about the criminal justice system. We have these two sets of cases, the wrongful convictions and the near misses. What we're doing in this research is we are comparing them, what happens to a defendant in one set of cases versus the other.
Someone comes into the criminal justice system; their case can go one way or the other. What we're trying to figure out is what explains why a case goes this way as opposed to going that way. Why does a defendant get wrongly convicted as opposed to having a near miss? The one set of cases, the wrongful convictions, what happens in these cases is someone who is factually innocent ends up being indicted. That means he’s going to be prosecuted. He ends up being convicted of a crime he does not commit and he does time for that.
Eventually, he ends up being exonerated. That means that, eventually, he’s able to show evidence that he did not commit the crime. That’s one set of cases.
In the other set of cases, what we call the “near miss” cases, again, someone who is factually innocent ends up being indicted. The prosecution is going to go forward and begin this case. But, before he’s convicted, the prosecution and the defense, between the two of them, realize that the person is factually innocent and he has his case dismissed or, at trial, the jury concludes that this person is factually innocent. They acquit him, but they acquit him not on the basis of a legal technicality; they acquit him because they conclude he didn’t do the crime.
What does the term “near miss” mean?
We use the term “near miss” because what we’re really interested, in this research, is how the criminal justice system identifies someone who’s factually innocent to prevent the wrongful conviction. This isn’t simply trying to understand how wrongful convictions happen. That’s looking at a lot of the problems in the system.
We also want to look at what the system does well and how the system is able to identify factually innocent people to prevent their conviction.
Using quantitative and qualitative research methods.
We used both quantitative and qualitative methods for this research. Quantitative research is what people generally know as statistical research. So, we’re looking for statistical patterns between both sets of cases. Qualitative research is where you actually look at the case facts and try to figure out what’s going on in an individual case. There, we were also — we had the assistance of an expert panel. There we had experts from the criminal justice system, prosecutors and police officers and the like who are able to help us look at these cases as well. In order to do the quantitative research, we had to collect information on all of these cases. We had approximately 60 variables of information on every single case. We have over 400 and something cases in this data set and we have over 60 variables on each one of these. This took years of research to be able to even get that information.
[End of video clip]
What did the study find?
Gould The results from the quantitative findings were actually really interesting because they identified 10 factors that help explain why a defendant is wrongly convicted as opposed to a near miss. I just want to make sure that we get this down correctly.
These 10 factors are, number 1, the age of the defendant. Number 2, the defendant’s criminal history. Number 3, the strength of the prosecution’s case. Number 4, the quality of the defense that the defendant had. Number 5, whether the defense brings or puts on a family witness. Number 6, whether the prosecution fails to turn over exculpatory evidence. Number 7, whether this prosecution takes place in a state with a punitive culture on the death penalty. Number 8, whether there is lying by a non-eyewitness. Number 9, whether a misidentification that happens in this case is an honest misidentification. Then, number 10, whether the forensic evidence that is offered is presented accurately or not.
What we found was that having any criminal history made the defendant more likely to be wrongly convicted. So, simply being known to police or prosecutors, having any prior familiarity with the criminal justice system puts someone at greater risk for a wrongful conviction.
So, let’s take a moment here and let me explain how we used our expert panel to help us understand this. Because we have this quantitative finding, we have this finding from one half of the research that says criminal history is related to wrongful conviction, but not criminal history of this crime for which the person is being indicted for.
So, we brought the expert panel together and what they helped us to understand is that what might be going on here is simply being known to police officers means that someone’s name comes up in an officer’s mind as a potential suspect or someone’s picture is, then, in a mug shot book. If someone’s picture is in a mug shot book, it means that they are potentially at greater risk of being incorrectly picked out of that photo book by the victim or some eyewitness to the crime. So, simply being known to police officers or having any prior photo of the defendant, potentially makes him at greater risk of wrongly convicted.
Let me say a little bit more about another one of the variables and that is this forensic evidence variable. Prior research has shown that errors in forensic evidence explained wrongful convictions.
Instead, what our research shows is improper testimony about the forensic evidence is what ends up leading to a wrongful conviction.
How reliable are the findings?
We’ve talked about the 10 variables that are related to the wrongful convictions and the near misses and we talked about the expert panel. So, someone might naturally ask, “Well, is this credible?” I think these are highly credible results. These are results that I feel very confident about. That begs the question, well, why? Why do you feel so confident about what you found?
There are really two answers to that. The first one is we did a statistical test that we call an ROC curve. That’s highly technical. I don’t expect anyone to really be interested in that, other than a few statisticians. But, what that showed is that we have tremendous certainly in the results we found, that these 10 variables are statistically related to the outcome that we were looking for.
The other piece of this that makes me confident is that we had the expert panel to double check these results and to see whether they make sense in their practice in the criminal justice system. Not only did they tell us that these results made sense, but they were able to help us understand how they operate. So, I think these are highly credible results.
[End of video clip]
What did the study reveal about 'systemic error'?
Gould If there’s one big finding from the statistical evidence here, it is that there are three variables that work in concert to create systemic error. Those three errors are the strength of the prosecution’s evidence, the quality of the defense, and whether the prosecution turned over exculpatory evidence. That means evidence that it’s required under law to give to the defense when that evidence suggests that the defendant is actually innocent.
Here is how these three things work together. What we found is that when you look at the strength of prosecution’s evidence, weaker facts were associated with wrongful convictions. This goes against all common understandings. You would think that if there are weaker facts, they would lead to a near miss. But, here, they lead to a wrongful conviction.
So, the second piece here is the weaker defense and this is something we might expect. If I’m a defendant and I have a weaker quality of defense, of course, I would lead to a wrongful conviction. It would lead to a wrongful conviction as opposed to a near miss.
This third thing, the failure to disclose evidence by the prosecution — if the prosecution doesn’t turn over evidence that the defense needs, then the defense can’t do its job in trying to lead to the near miss.
What we think is going on here from these three is that the prosecution has weaker facts. It is not able to do as thorough an investigation on these facts because it just doesn’t have that much to go with. It doesn’t turn over the evidence to the defense because it doesn’t recognize sometimes what those facts are and, then, the defense can’t do its job. Really, if you put these three factors together, if you put the weak prosecution facts, if you put the weak defense and you put the failure to disclose, what you have is a perfect storm of systemic failure in the criminal justice system that leads to the wrongful conviction of an innocent defendant.
When I talk about systemic failure in the criminal justice system, what I mean is that these aren’t just isolated factors that lead to a wrongful conviction. It’s not simply that, for example, the age of a defendant is what leads to the wrongful conviction. We don’t say in the study, for example, that younger defendants automatically end up being wrongfully convicted. There’s something going on in a criminal justice system that is fluid, that has these factors interrelating to one another. You don’t just say, “oh, it’s this factor by itself.” It’s that they combine and that, in the end, an innocent defendant who comes into the criminal justice system gets wrongly convicted because the interplay of these many factors that show that prosecutors and defense lawyers are not doing their job in the system of adversarial justice to try to make sure that justice actually occurs.
The role of 'tunnel vision'
The take away from this study is, first, we need to decide where do we want to put our energies? So, if the question is what ends up having a defendant be indicted, then, it’s a number of the things that we’ve seen before: eyewitness identification, false confessions and misconduct by police officers and prosecutors.
If, though, the question is, how do we prevent a wrongful conviction once someone ends up being indicted, then, it seems to me, we have three big factors to be focused on. They are either, number one, making sure that the quality of defense is better, number two, making sure that police and prosecutors are turning over exculpatory evidence and, number three, making sure that weak facts do not end up leading to a prosecution that keeps forward to trial.
The way all three of these relate are under the concept of tunnel vision. Tunnel vision is something that, actually, all of us engage in every single day. It’s that we start to assume something is a one way because we have a few facts that we think lead to that.
What happens here in one of these cases, in a wrongful conviction, is the police have one fact about a defendant and they begin to focus on that defendant and start not paying attention to other facts in the case that suggest that the defendant is innocent. Again, this is not specific to police or prosecutors. We all do this. But, that is one of the big problems here. So, if we are trying to prevent wrongful convictions, we need to be focused on the strength of the prosecution’s facts. We need to be focused on the quality of defense, exculpatory evidence. But then how those lead to tunnel vision that have the professionals in the criminal justice system not looking at other facts that they ought to be paying attention to, that would suggest they have the wrong guy and that, as a result, the real criminal, the most dangerous person is still out there on the streets.
[End of video clip]
Opinions or points of view expressed in these recordings represent those of the speakers and do not necessarily represent the official position or policies of the U.S. Department of Justice. Any commercial products and manufacturers discussed in these recordings are presented for informational purposes only and do not constitute product approval or endorsement by the U.S. Department of Justice.
- Campus Sexual Assault Responses (CSAR): Informing Trauma-Informed Policies, Protocols, and Training
- Increasing Rigorous Evaluation of Interventions to Reduce Gender-based Violence Victimization
- Preparing for and Responding to Threats and Violence - Breakout Session, NIJ Virtual Conference on School Safety