In general, academic performance has a negative correlation with juvenile delinquency. Which makes sense -- we assume that academic excellence both shows a commitment by kids to work towards socially sanctioned goals, and that the pay-offs of excelling in school make criminal activity less attractive.
However, new research by University of Washington Professor Robert Cruchfield indicates that for kids in highly disadvantaged neighborhoods, this relationship is reversed: GPA actually has a positive correlation with delinquency.
This was an unexpected and obviously distressing finding, and folks are still trying to figure out potential causes. Several hypothesis were forwarded:
1) Kids with good grades in bad neighborhoods feel the need to "represent", countering the bad reputation they might get for being a nerd.
2) Kids with good grades feel like they have no potential for being rewarded for their accomplishments, and seek out the status that criminal activity brings.
3) Kids with good grades are more valuable to criminal organizations such as gangs, who value their intelligence and skill sets more than another dime-a-dozen brawler.
4) Having good grades acts as a "get out of jail free" card, causing authority figures to look the other way and making delinquent activity less costly.
All of these are worth considering, but I think #2 rings truest for me. One knock on the first explanation is, as one respondent noted, that it would seem simpler for smart kids to avoid the dilemma all together by underperforming in class. The fourth explanation I think would work better in high-income neighborhoods (where I definitely observe a "get out of jail free" effect). It's at best non-unqiue, at worst less effective, in poorer locales.
The third explanation seems slightly incongruous, but I think may be onto something insofar as we remember that modern gangs are rather sophisticated organizations which make rational "business" decisions. Still, the second example strikes me as the strongest. Kids who are high performers expect that they will see tangible rewards for their effort. Insofar as they don't see those rewards coming, either because they don't believe their work will pay off (i.e., even if they do well in school, they won't be able to afford college or can't imagine a good college will accept them), or because they don't believe they'll be around to see the benefits (i.e., they are fatalistic about street violence, or just assume they'll be taken in by the police on trumped up charges), the temptation to take immediately presented benefits rises dramatically.
Smart kids respond to rational incentives. The extent that criminal activity is the rational choice in many poorer neighborhoods worries me. A good anti-crime policy (indeed, a profoundly conservative one insofar as it deals with incentive structures) should be working holistically to make sure that other options are more fruitful than crime for disadvantaged youth.