Tuesday, September 14, 2010

On The Rate of Wrongful Conviction: Chapter 11.4

As I have mentioned many times previously, I am preparing a monograph on the rate of wrongful conviction. Each chapter will deal with one estimate of that rate, beginning with zero and ending beyond 10%. I am posting the draft chapters here, as I write them.  I have so far posted the following: 

Chapter 0.027: The Scalia Number
Chapter 0.5: The Huff Number
Chapter 0.8: The Prosecutor Number
Chapter 1.0: The Rosenbaum Number
Chapter 1.3: The Police Number
Chapter 1.4: The Poveda Number
Chapter 1.9: The Judge Number
Chapter 2.3: The Gross Number
Chapter 3.3: The Risinger Number
Chapter 5.4: The Defense Number
Chapter 9.5: The Inmate Number
Chapter 10.1: A Skeptical Juror Number
Chapter 11.1: A Skeptical Juror Number

As we draw near the end of this series, we return to another guess estimate. This one is based not on a survey of prosecutors, police, or judges. Instead it is based on a survey of commoners. You might find it an interesting read. It seems as if we commoners may be more savy than our protectors give us credit.

Chapter 11.4
The Common Man Number

One group of guessers has so far been overlooked. That group is the source of our juries, the American public. As you might recall, 50% of the prosecutors surveyed by Robert Ramsey guessed that our country’s wrongful conviction rate was less than 0.8%. In that same survey, police guessed the number to be less than 1.3% and judges guessed less than 1.9%.

Since 1999, those who conduct the Harris Poll have repeatedly asked the American public what they thought. More specifically, the pollsters asked the public to estimate the number of convicted murderers who were wrongfully convicted.  I quote from a summary of a March 2008 poll.
There is one issue almost all Americans agree on – 95 percent of U.S. adults say that sometimes innocent people are convicted of murder while only 5 percent believe that this never occurs. This is a number that has held steady since 1999. Among those who believe innocent people are sometimes convicted of murder, when asked how many they believe are innocent, the average is 12 out of 100 or 12 percent.
Apparently, 5% of the American public neither reads newspapers, watches television, nor has an internet account. Alternatively, 5% of our public consists of police and prosecutors who steadfastly refuse to acknowledge they might make an error.

Taking into account that 5%, the average estimate for the entire poll sample was 11.4%. That’s the percentage of people convicted of murder who are factually innocent, at least according to the best guess of Jane and Joe Q. Public. That number is surprisingly close to the 9.1% value I calculated for jury trials based on judge-jury agreement data. It’s even closer to the 10.1% value I calculated based on irreversibly-resolved capital murder cases. If the calculations just mentioned are correct, the commoner has far better insight into this serious social problem than do police, prosecutors, judges, and defense attorneys.


Not surprisingly, the common-man estimates were strongly influenced by race. Whites, who comprise a disproportionately small percentage of those in prison, believe that 9% of those convicted of murder are wrongfully convicted. Blacks, on the other hand, are imprisoned at 6 times the rate of whites. They believe that 25% of those convicted of murder are wrongfully convicted. The trends are shown in the table and plot below.

Another striking number appears in the box enclosed at the bottom-right corner of the table. It  is an average estimated value of wrongful convictions for murder, weighted by the ratio of percentage-prison-population to percentage-general-population. That weighted average of 16.5% is nearly identical to the Inmate Survey number of 17.5%. The comparison suggests that those most impacted by wrongful convictions are best at estimating its frequency.


Finally, consider if you will the following plot. It shows the relationship between estimates of wrongful conviction rates for murder and measures of prosecution / conviction errors for murder. I’ll detail the plot in the notes, but I want to focus here on the message. Those who are most confident we seldom convict the innocent are most prone to error. Police and prosecutors, for example, estimate that only 2% of our convictions are wrongful convictions; 22% of those they send to trial, however, are innocent (at least based on my judge-jury agreement analysis.) At the other extreme, the common man predicts a wrongful conviction rate six times greater, yet errs at one-third the rate.

Those who are most confident, are most likely to err. Those who are less confident, are less likely to err. Those who are the most directly impacted are most likely to perceive that we, as a society, have a serious problem.

1. For the Harris Poll summary, see Regina A. Corso, Harris Poll #28, March 18, 2008.

2. For the racial / ethnicity distribution of our general and prison populations, see Wikipedia

3. In the first plot, the estimated wrongful conviction rates for murder excluded those 5% of respondents who believe we have yet to wrongfully convict anyone of murder.

4. In the second plot, the estimated wrongful conviction rates are average numbers (mean values), not 50% numbers (modes) as I previously discussed in earlier chapters. While it is correct to use the mean numbers in the plot, the sense of the plot would have been little changed had I used some combination of mode and mean values for the estimated wrongful conviction rates.

5. In the second plot, the estimated wrongful conviction rates are for all crimes. The errors are murder related errors.

No comments:

Post a Comment