This is Austan Goolsbee of the University of Chicago on the factors determining the productivity of terrorists, and a warning about interpreting who terrorists are based upon who is caught:
Even for Shoe Bombers, Education and Success Are Linked By Austan Goolsbee, Economic Scene Commentary, NY Times: The fifth anniversary of 9/11 passed with a great deal of hand-wringing over all the people who want to kill Americans. Especially worrisome is the apparent rise of terrorists whose origins seem far from fanatical.
These terrorists are not desperately poor uneducated people from the Middle East. A surprisingly large share of them have college and even graduate degrees. Increasingly, they seem to be from Britain, like the shoe bomber Richard C. Reid and most of the suspects in the London Underground bombings and the liquid explosives plot.
This has left the public wondering, Why are some educated people from Western countries so prone to fanaticism? Before trying to answer that question, though, some economists argue that we need to think about what makes a successful terrorist and they warn against extrapolating from the terrorists we catch. It is a problem economists typically refer to as “selection bias.”
In their new study, “Attack Assignments in Terror Organizations and the Productivity of Suicide Bombers,” two economists, Efraim Benmelech of Harvard University and Claude Berrebi of the RAND Corporation, set out to analyze the productivity of terrorists in the same way they might analyze the auto industry. ...
They gathered data on Palestinian suicide bombers in Israel from 2000 to 2005 and found that for terrorists, just like for regular workers, experience and education improve productivity. Suicide bombers who are older — in their late 20’s and early 30’s — and better educated are less likely to be caught on their missions and are more likely to kill large numbers of people at bigger, more difficult targets... Experience and education also affect the chances of being caught. Every additional year of age reduces the chance by 12 percent. Having more than a high school education cuts the chance by more than half. ...
The research suggests ... that there may be a reason that the average age of the 9/11 hijackers ... was close to 26 and that the supposed leader, Mohammed Atta, was 33 with a graduate degree. As Professor Benmelech put it in an interview: “It’s clear that there are some terrorist missions that require a certain level of skill to accomplish. The older terrorists with better educations seem to be less likely to fail them. Perhaps it is not surprising, then, that terrorist organizers assign them to these more difficult missions.”
Among Palestinian suicide bombers, the older and better-educated bombers are assigned to targets in bigger cities where they can potentially kill greater numbers of people. That same idea means that the terrorists assigned to attack the United States are probably different from the typical terrorist. They will be drawn from people whose skills make them better at evading security.
We know relatively little about why the suspects arrested in the recent liquid explosives plot failed... Perhaps ... poor educations made it easier to uncover their plan. We do know, however, that they included a large number of South Asians raised in Britain. It’s only natural that terror groups would recruit native English speakers... It does not imply that the Muslim community is a more fertile ground for terrorists in Britain than in other countries.
Think of the extreme case. One of the people arrested in the liquid explosives plot ... was a woman with a baby. London newspapers have speculated that she was planning to carry her baby onto a plane with liquid explosives in his bottle. Even if true, that does not mean we should all start suspecting that women with babies are closet terrorists. That would be rather egregious selection bias. Objectively, we should be much more suspicious of other people. We see only the mother because the terrorists have an overwhelming incentive to find the one unusual terrorist who will outsmart our defenses.
And sadly, it seems that educated and intelligent terrorists are better at doing that than uneducated, fundamentalist lunatics. Oh, that it weren’t so. Like the old advertisement said, a mind is a terrible thing to waste.
Here's another example of selection bias that might help to clarify. Suppose there two sets of people, one living in city A and the other living in city Z, and you need to recruit a group of tall people to help you in a criminal venture involving impersonating basketball teams. Suppose further that the number people, the distribution of heights, and the willingness to commit crimes is identical in each city.
Now suppose you know your way around city A much better than city Z, and it's closer, so it's cheaper and safer to recruit in city A than in city Z. Then your group, once formed, will only have people from city A in it but they are no more likely to be tall and criminal than people in city Z (to make it like the example above, there could be several teams formed from the group of recruits and each team would be formed and assigned to a location based upon the age and education of team members). Thus, if and when the authorities catch people from your group and discover they are all tall and all from city A (and they are likely to be young and uneducated due to the assignment), it would be wrong to then conclude that young, tall, uneducated people in city A are more inclined toward crime than young, tall, uneducated people people in city Z.
[See also Civil Liberties and Terrorism.]