The first children of Head Start are old enough now to have children of their own. They’ve moved through high school — if they were able to get that far — and some much farther than that. They’ve entered the workforce and formed their own families.

That means it’s increasingly possible to track the long-term effects of a federal program, created in the mid-1960s, that sought to give 3- and 4-year-olds from struggling families an early lift out of poverty.

A new analysis from the Hamilton Project suggests that their lives today are measurably better in some important ways than those of poor children who never enrolled in the program. Their chances of finishing high school, attending college and earning postsecondary degrees or certificates were higher.

As adults, blacks in particular rated better on indicators of noncognitive skills such as planning and problem-solving. And as parents, the children of Head Start appear to invest more in their own children, suggesting that years after the government’s initial investment, the program could indirectly touch a second generation.

“The news is that these measures moved. That’s exciting,” says Diane Schanzenbach, the director of the Hamilton Project and one of the co-authors of the research, along with Lauren Bauer.

That result, though, complicates a large canon of Head Start analyses that have found over the years mixed and even disappointing results. And it reflects a curious possibility: that an intervention that didn’t yield immediate results in school preparedness or test scores might reveal its value years later.

“The fundamental problem with these programs where we care about their long-term impacts is that we don’t have a time machine,” Schanzenbach says. “It takes time between when somebody was in pre-K and when we can look at these long-term outcomes.”

Earlier studies have concluded that any gains in preschool for Head Start kids quickly recede, with the program showing little visible impact on cognitive and social-emotional outcomes by the third grade. Past research has also found no impact on children's math skills, and little evidence that their health improved (the program was designed to include health and nutritional services, too).

The new analysis, modeled off earlier waves of Head Start research, relied on data from the National Longitudinal Survey of Youth, which has tracked a nationally representative sample of men and women — and later, their children — since 1979. The survey captured several hundred Head Start families, including those where one child enrolled in the program and another didn’t.

The new analysis compares outcomes for Head Start children with their own siblings who did not attend the program, to more closely control for other possible factors that might explain differences between the two groups. Among children enrolled in Head Start between 1974 and 1994, Schanzenbach and Bauer found their likelihood of graduating high school was about 5 percentage points higher. The improvement was twice as large for Hispanic children and children whose mothers had never graduated high school themselves.

Those results aren’t as strong as what’s been found for the Perry Preschool program, a model high-quality preschool for poor children run in the 1960s, but they’re larger than what earlier Head Start studies have shown:

The effect of Head Start appeared even larger in boosting a child’s chances of completing some higher education, including associates’ degrees and license programs. Hispanic children who benefited from the program were about 15 percentage points more likely to do so than their siblings who didn’t attend Head Start:

These same Head Start children, when they became parents, were also more likely in the longitudinal survey to say they read aloud to their own children and taught them the alphabet, or had in the previous week praised them and spent time with them doing the child’s favorite activities.

That’s a very different set of outcomes than measuring kindergarten readiness or third-grade skills. But results like that reveal more about the potential of investments in young children to later break the intergenerational cycle of poverty.

In some ways, these findings mirror recent research about another major government intervention into the lives of poor children that initially appeared to show disappointing results. In the 1990s, the government gave some families housing vouchers to move to lower-poverty neighborhoods with better schools and less crime. But that opportunity didn’t significantly alter how children performed in school or how much their parents earned.

A new analysis released last year, however, found effects much later in life: By their mid-twenties, children who'd benefited from the Moving to Opportunity experiment were earning more relative to poor children who didn't make the same moves, and they were more likely to have attended college and were less likely to be single parents.

“This comes up over and over in the research literature: Measurable impacts fade, then they come back when they’re adults,” Schanzenbach says. “It’s happened enough that it should give us pause.”

It’s possible that some of that result with Head Start is explained by other changes over the life of the program. In the early years, poor children who didn’t attend Head Start had fewer alternatives. They were less likely to attend other kinds of preschool. Sesame Street didn’t exist. So some of the varied results comparing the two groups over years of research may reflect changes in the quality of the program and its counterfactuals.

But it’s also possible, Schanzenbach says, that Head Start has helped embed the kind of noncognitive skills — including self-control and self-esteem — that don’t necessarily translate into better reading scores but that may prove valuable later in life.

"That makes sense that that’s really the mechanism here,” she says. “It would be nice to understand more about that."

None of this means that Head Start, often criticized in political debate as ineffective and a waste of money, couldn’t be improved. But Schanzenbach argues it would be a mistake if we failed to recognize that investments in children pay off over a lifetime — and may take a lifetime to reveal their worth.