Washington, June 26 : Of the many inmates on death row, there are only a few who actually get to see the execution chamber. And in order to know who will be spared, scientists have developed a new computer system to gauge the chances of life and death for such prisoners and the results that came out have been chillingly accurate.
And these cold calculations have confirmed suspicions that the people most likely to be executed are those who have had the least schooling, instead of those who have committed the most heinous crimes.
While US is the only western democracy to retain the death penalty, but executes only a few of the people it sentences to death. For instance, just 53 of the 3228 inmates on death row were executed in 2006. And how they make that choice is what made the scientists to go for the study.
"We couldn't see any clear patterns in the data," said computer scientist Stamos Karamouzis, who has been investigating this question with criminologist Dee Wood Harper at Loyola University in New Orleans, Louisiana.
After tasting initial failure through a direct approach, the researchers turned to an artificial neural network (ANN) - an intelligent computer system, modelled after the human brain - that is able to deduce how various factors within a jumble of data relate to each other. The system can then take what it has learned and make predictions about a new set of data.
Already, Karamouzis has used ANNs to predict the likelihood that juveniles given parole will reoffend, and to pinpoint the students most likely to drop out of college courses. "ANNs surprise us by revealing non-obvious patterns," he said.
For knowing which factors might be linked to executions, they first "trained" their ANN by entering the profiles of 1000 death row inmates between 1973 and 2000. Half of this sample of prisoners had been executed and the other half had survived. Each profile contained 18 factors, including the inmate's sex, age, race, marital status, educational level and information on their capital offences.
Later, they fed in profiles for 300 more inmates from the same period and asked the ANN to predict what had happened to them. Surprisingly, it correctly predicted the fates of more than 90 per cent of those inmates.
While the neural network had clearly hit on a strong relationship between the inmates' profiles and their likelihood of execution, the team wanted to know which factors were most important. For this, they repeatedly retrained the ANN from scratch, withholding information about one of the factors each time.
And they discovered that gender was the most significant factor - women are rarely executed. However, Harper said that while race was implicated to be a key factor in sentencing criminals to death, it did not turn out to be an important factor when it came to the decision to execute.
The most striking factor by far was educational level - the number of years the inmate had spent in high school. This may be crucial because it indicates how well an inmate can manage their appeal process.
"This finding confirms that being executed is not about what you've done, but more about your ability to defend yourself," said Simon Shepherd of Death Watch International, a group that campaigns against the death penalty worldwide.
However, the researchers do not expect their work to have much effect on policy.
"Until US public opinion shifts, no amount of scientific evidence is going to make any difference," said Harper.
"It raises lots of tricky questions. If it says someone is likely to be executed, would the lawyers give up on saving that individual?" said Karamouzis.