Over the past decade we’ve had a rich debate on how to expand opportunity for underprivileged children. But we’ve probably made two mistakes.

First, we’ve probably placed too much emphasis on early education. Don’t get me wrong. What happens in the early years is crucial. But human capital development takes a generation. If you really want to make an impact, you’ve got to have a developmental strategy for all the learning stages, ages 0 to 25.

Second, we’ve probably put too much weight on school reform. Again, reforming education is important. But getting the academics right is not going to get you far if millions of students can’t control their impulses, can’t form attachments, don’t possess resilience and lack social and emotional skills.

So when President Obama talks about expanding opportunity in his State of the Union address on Tuesday, I’m hoping he’ll widen the debate. I’m hoping he’ll sketch out a stage-by-stage developmental agenda to help poor children move from birth to the middle class.

Such an agenda would start before birth. First, children need parents who are ready to care for them. But right now roughly half-a-million children are born each year as a result of unintended pregnancies, often to unmarried women who are not on contraception or are trying to use contraceptives like condoms or the pill. As the University of Pennsylvania’s Rebecca Maynard and Isabel Sawhill and Quentin Karpilow of the Brookings Institution have argued, if these women had free access to long-acting reversible contraceptives like I.U.D.’s, then the number of unintended births might decline and the number of children with unready parents might fall, too.

Once born, children are generally better off if they grow up within a loving two-parent marriage. It would be great if we knew how to boost marriage rates, but we don’t.

For the time being, we probably should spend less time thinking about marriage and more time thinking about parenting skills. As Richard Reeves, also of Brookings, points out, if we could teach the weakest parents to behave like average parents — by reading more to their kids, speaking more, using consistent, encouraging discipline — then millions of children might have more secure attachments, more structure and better shots at upwardly mobile careers. Programs like Nurse-Family Partnerships and the Baby College in the Harlem Children’s Zone seem to be able to teach these parenting skills.

Once they get to elementary school, children need to learn how to read and write. But that can’t happen in schools where 15 percent of the students are disruptive, where large numbers of students live with so much stress that it has stunted the development of the prefrontal cortexes, sent their cortisol levels surging, heightened their anxiety responses and generally made it hard for them to control themselves.

Therefore, we probably need more programs like Pamela Cantor’s Turnaround for Children, which works in schools to help teachers and administrators create “fortified environments,” in which overstressed children can receive counseling and treatment, in which the psychic traumas that go with poverty are recognized and addressed.

According to work done by Sawhill and others, a significant number of kids stay on track through the early years, but then fall off the rails as teenagers. Sawhill set a pretty low bar for having a successful adolescence: graduate from high school with a 2.5 G.P.A., don’t get convicted of a crime, don’t get pregnant. Yet only 57 percent of American 19-year-olds get over that bar. Only one-third of children in the bottom fifth of family income do so.

Over the next few years, we’ve got to spend a lot more time and money figuring out how to help people from poorer families chart a course through the teenage years. There’s evidence that Career Academies help adolescents navigate the teenage rapids. There’s some evidence that New York’s “small schools of choice” yield measurable results. We as a nation have made awesome progress in reducing teenage pregnancies, so it is possible to change teenage behavior, even in the face of raging hormones.

But it is harder to find successful programs geared toward teenagers than it is to find successful programs geared toward younger children. It feels like less money has been raised to help teenagers, fewer innovative programs have been initiated.

Robert Putnam of Harvard argues that when we design early education programs, they need to be “wrap-around.” They need to have formal and informal programs that bring parents in and instill communal skills. With teenagers, we need more guidance counselors to help them become savvy, so they know how to work the system, and to respond when their needs aren’t being met.

Putnam is emphasizing skills — for toddlers or teenagers — that are hard to see and measure. But that’s the next frontier of human capital development: Building lifelong social and emotional development strategies from age 0 to 25. I’m hoping President Obama goes there.