Head Start Prevents Foster Care? To Be Decided

The jury is decidedly out on the academic track record of Head Start, the education-oriented pre-school program for low-income families invented in the 1960s and federally proliferated in the early 1980s.

Critics will point to large impact studies that show early academic gains fade by third grade. Proponents will say that those gains would stick if the students ended up in better public schools.

But Youth Services Insider had never seen Head Start mentioned as a possible preventer of foster care, until a recent study authored by Sacha Klein, Lauren Fries and Mary Emmons. (Click here for the synopsis; the whole study requires subscriber access.)

The study did not get much attention in the media; just a few stories out of Lansing, Mich., home to study author Klein of Michigan State University. Those stories honed in on the finding that among a sample of youth whose parents had previous contact with the child welfare system, those participating in Head Start were 93 percent less likely to be placed in foster care than kids who received no early childhood education program.

Head Start nearly doubles the odds that a system-involved child won’t land in foster care? File that under “big if true.”

But a read of the study’s details shows that we are far from certifying that connection. The study used the National Survey of Child and Adolescent Well-Being to establish a universe of 1,995 children, under 5 years of age, who lived with a biological or adoptive parent and had been the subject of at least one maltreatment report.

Of that group, just 32 of the 1,995 received Head Start. And of those 32 kids, only one ended up in foster care within 18 months. After controlling for other variables that might influence foster care outcomes, that translated to 93 percent lower odds of being placed in foster care than young kids in our child welfare sample who were not in any type of early care and education.

Perhaps of equal interest: the kids who received other early childhood programs were more likely to land in foster care than those who received no program.

Klein, the study’s author, agrees that the size of the Head Start subset cannot be ignored, but also told YSI that the outcome warrants further attention.

“It is a limitation of the study,” Klein said of the tiny fraction of kids who went to Head Start. “I’d feel more confident if that were a bigger number. But it is statistically significant, it’s not a spurious finding. When you have a small sample size … the smaller the size, the harder it is to get a statistically significant finding. So the fact that it was significant even with a small sample size is meaningful.”

Klein said she hopes to spark some interest in a more robust study. A randomized control study would be difficult, because it would inherently require the denial of Head Start to Head Start-eligible children. Klein said the most likely approach would be to use propensity score matching to isolate and measure the impact Head Start has on child welfare cases.

The basic question is, does Head Start protect against foster care placement? If so, the follow-up question would be why.

Klein’s hypothesis is that adult involvement in Head Start could be a factor.

“The big thing about Head Start is that it involves parents, and not just that, but helps parents,” Klein said. “There are ancillary services … that invest in the human capital of the parent. With other early childhood programs, it’s often just the kid.”

She views Los Angeles County, the single largest local child welfare system in the country, as the best potential place to conduct a larger test of Head Start’s child welfare impact. “It’s a huge system, so it would be possible to get numbers,” Klein said.

Another L.A. advantage: the county’s Department of Children and Family Services has a unit dedicated to connecting families with early childhood education and childcare. Klein assisted in the evaluation of that program, the Los Angeles Child Welfare-Early Education Partners Initiative, which was started in 2011 with a three-year federal grant.