dropstop

Eli Broad has put a ton of philanthropic energy and money into education; his foundation's site has some interesting statistics:

1. 1.2 million students drop out of high school every year, 44% of dropouts under 24 are jobless
2. Dropouts pay less taxes (costing the Feds potential revenue), and 65% of US convicts are dropouts
3. It costs $12,956 more per year to house an inmate ($22,600) than educate a student ($9,644)

Assuming dropouts only spend two years in jail over their adult lives, then each dropout costs the government $45,200 + $? in arrest/court costs + $? in lost tax revenue = $100k (guessing $5,000 in arrest/court + $50k in lost taxes). I think the actual costs are much higher, but let's be conservative and say it's just $100,000 per dropout over their lifetime.

This means every year, the American government (American taxpayer actually) takes on an increased burden of $78 billion dollars just from the 65% of dropouts who will become convicts. Just for the dropouts since the year 2000, this burden over their lives is over $700 billion. To put that in context, the annual operating budget for education is $450 billion, which serves almost 50 million students. [See update below.] I couldn't find the data I wanted, but still believe that the burden on society is significant enough to warrant the investment below. 

Why do students drop out of high school? The three most frequently-cited reasons are: family issues ~25% (pregnancy or becoming a parent), employment ~32% (couldn't combine school and work, or needed to get a job), and motivation ~69% (didn't care, no-one pushed them). How about this: two thirds said that they would have tried harder if more was expected from them (i.e. someone cared).

The National Dropout Prevention Center proposes fifteen strategies for improving graduation rates (about twelve too many); if I were to pick three, besides the obvious ones (centralization and teacher improvement), my third pick would be mentorship, or as I proposed, "minternships."

I recall hearing about a Big Brothers Big Sisters study that said effective mentorship requires a minimum 18 month commitment; less can actually do more harm than good. The quality of the mentor is of course also a critical factor. But if you're good and you stick around, the results are strong and positive - the mentee is more likely to complete and graduate school, etc.

Is there a common thread among the three main reasons kids drop out? They're not being counseled well about family planning, birth control, safe sex, etc.; there are no clear options for financial support for children in high school; there's no-one who believes in them, encourages them, and provides them with a model of how to achieve gainful employment. There doesn't seem to be one, but I think we can create it :-).

Let's build a formalized mentorship program (using the Teach for America concept, we could call it "Reach for Americans") where students are able to receive not just counseling and support, but also a stipend (for community service or internships that pays enough to prevent the need to get a job). The mentorship must focus on both family planning/sex ed and of course motivation, support and inspiration to stay in school, graduate and succeed. Let's not make Reach for Americans (or whatever it'd be called) a .org - let's instead make it a .gov, that way it's not beholden to fund-raising, and students can count on the counselors being there always.

The program should begin early - say at age 8, and continue through high school graduation - this would get expensive (but I bet we'd not only improve graduation rates, but also student performance, and hopefully in a generation be able to claim that families were elevating themselves out of poverty because their children were graduating high school (potentially going on to community college or university), and earning a family wage. This organization might also become a student advocacy group, and argue for student rights across the board, but particularly in low-income communities.

The business case is clear, the outcomes could be significant - I think we should pick a state and give it a try.

UPDATE - I realize (thank you everyone for your emails and comments!) that I made a false assumption above - 65% of criminals are dropouts, not 65% of dropouts are criminals. I will correct this this evening.