Dive Deeper: Same Seed, Two Soils

Five sources behind our post — what they found, why they matter, and where the evidence has limits.

This is the research companion to Same Seed, Two Soils. Read the main post first.

The parenting-intervention literature usually intervenes on the home OR the school. ParentCorps, the spine of Post 18, did both. Five other studies show why the both-and matters more than either alone.

When You Saturate, the Curve Moves

Dobbie, W., & Fryer, R. G. (2011). Are high-quality schools enough to increase achievement among the poor? Evidence from the Harlem Children’s Zone. American Economic Journal: Applied Economics, 3(3), 158-187.

DOI: 10.1257/app.3.3.158

What they found: The Harlem Children’s Zone is the most ambitious attempt in modern American policy to coordinate everything around the developmental experience of a child in a defined geographic area. Charter schools. Parenting programs. Health services. College counseling. After-school enrichment. All inside a few square miles in Harlem. Dobbie and Fryer used the school admissions lottery to identify what the package did compared to the same kids without it. The kids who won the lottery and got the HCZ charter schools closed the Black-white achievement gap in mathematics over three years of middle-school enrollment, with smaller and less robust gains in English Language Arts.

Why this matters for you: Dobbie and Fryer themselves argue that the school inputs drive most of the effect -- in their own words, community programs alone are “neither necessary nor sufficient.” That is a real finding, and it cuts against any easy “saturation” reading. The FCD reading is more specific. The Promise Academy schools were themselves coordinated environments: they bundled instruction with parenting classes, health services, and behavioral norms inside the school’s walls. The coordination did not have to live across a separate neighborhood program. It lived inside the school. Post 18 is the parent-facing version of that argument, with one extension: when the school cannot do it alone, the alignment between school and home becomes the next layer of the soil.

What it doesn’t answer: Which specific components of the HCZ model drove how much of the effect. The evaluation cannot decompose the school effect from the surrounding-services effect, because the kids got both. The honest version of the finding is that the package works. Disaggregating the package is hard.

School Reform That Required Parents, and Worked Because of It

Comer, J. P. (1988). Educating poor minority children. Scientific American, 259(5), 42-48.

DOI: 10.1038/scientificamerican1188-42

What they found: James Comer described the New Haven School Development Program, a comprehensive school reform model he and his Yale team built in the 1960s and refined for two decades. The model required three things to operate at the same time. A school planning team that included parents in real decision-making, not just as audience. A mental-health team that treated kids’ emotional development as a coequal goal alongside academic instruction. A parent program that integrated parents into the daily life of the school structurally, not as an occasional event. Schools that implemented all three saw measurable gains in achievement, attendance, and behavior over multiple years.

Why this matters for you: The Comer model is forty years old and it remains the cleanest articulation of what coordinated school reform actually requires. Parents are not asked to volunteer. They are structurally included in the governance of the school. If you have ever felt like the school treats parents as a problem to be managed rather than a partner to build with, you are reading the system correctly. Comer showed forty years ago what the alternative looks like.

What it doesn’t answer: Why most American school districts that have piloted Comer-style coordination have failed to sustain it. The implementation literature points to political turnover, funding instability, and the demand on educators to do work the system was not designed to compensate. Comer’s own writing acknowledges the gap.

What Multi-Component Prevention Looks Like in the Wild

Conduct Problems Prevention Research Group. (2010). Fast Track intervention effects on youth arrests and delinquency. Journal of Experimental Criminology, 6(2), 131-157.

DOI: 10.1007/s11292-010-9091-7

What they found: FFast Track was a multi-component prevention trial that combined classroom social-emotional curriculum, parent training, home visits, peer coaching, and academic tutoring for high-risk kids starting in first grade and continuing through tenth grade. 891 kids randomized across four sites. By age nineteen, the intervention group showed significantly lower severity-weighted arrest activity, lower rates of first juvenile arrests, and lower self-reported antisocial behavior than controls. The outcomes were measured roughly three years after the intervention ended in tenth grade.

Why this matters for you: A decade of coordinated prevention -- first grade through tenth grade -- produced kids who, as young adults, were measurably less likely to end up in the criminal justice system. This is what “the system has to tell the same story” looks like when it is funded and sustained. Most school-only or home-only or therapy-only interventions cannot generate this kind of effect because they are operating on one part of the developmental environment while the other parts are sending contradictory signals. Coordination is the variable.

What it doesn’t answer: Whether Fast Track is cost-effective at the population level. The intervention is expensive. The benefit-cost calculations are favorable for the highest-risk kids, less clearly favorable for universal application. The math depends on the targeting question.

When Teachers Pick Up the Phone, Engagement Moves

Kraft, M. A., & Dougherty, S. M. (2013). The effect of teacher-family communication on student engagement: Evidence from a randomized field experiment. Journal of Research on Educational Effectiveness, 6(3), 199-222.

DOI: 10.1080/19345747.2012.743636

What they found: Kraft and Dougherty ran a randomized field experiment inside a mandatory summer-school credit-recovery program for sixth and ninth graders. Teachers in the treatment condition made a daily phone call home plus a written message to each family across the five-day session, with specific updates on each kid’s classroom experience. Control teachers used standard contact. Students whose teachers reached out daily had a forty-two percent higher odds of completing homework, twenty-five percent fewer teacher redirections to task, and fifteen percent higher class participation. Teachers in the treatment arm maintained eighty-six percent compliance on planned contacts.

Why this matters for you: This is the smallest, cleanest, most replicable version of the Post 18 argument. When the school and the family hear from each other regularly about the specific kid, the specific kid does better. Daily contact home for a short window, with documented teacher compliance above eighty percent. That is the protocol. The effect beats most other things schools spend money on. The reason this is not standard practice is not that we do not know it works. It is that the teacher’s contract and the school’s incentive structure do not pay for it. If you want to know what to ask the school for, this is the protocol with the strongest evidence base.

What it doesn’t answer: How long the engagement gains persist after the structured outreach stops. The Kraft-Dougherty intervention ran for six weeks. The follow-up window is short. Whether the effects survive a return to normal practice is less clear.

Coordination at the Healthcare-Family Seam

Olds, D. L., Eckenrode, J., Henderson, C. R., Jr., Kitzman, H., Powers, J., Cole, R., Sidora, K., Morris, P., Pettitt, L. M., & Luckey, D. (1997). Long-term effects of home visitation on maternal life course and child abuse and neglect: Fifteen-year follow-up of a randomized trial. JAMA, 278(8), 637-643.

DOI: 10.1001/jama.1997.03550080047038

What they found: Olds followed up on the original Nurse-Family Partnership randomized trial fifteen years after the children were born. Low-income first-time mothers had received regular home visits from registered nurses through pregnancy and the kid’s second birthday. Thirteen years after the intervention ended, the families that had received nurse visits showed substantially lower rates of child abuse and neglect, fewer subsequent unintended pregnancies, longer intervals between births, less reliance on public assistance, and fewer maternal arrests and convictions. The mothers’ criminal-justice outcomes are the 1997 paper’s headline finding; the parallel effects on the children’s own delinquency were reported in a companion Olds 1998 follow-up.

Why this matters for you: Healthcare coordinating directly with the family on the developmental work produced effects fifteen years later. This is not the school version of coordination. It is the healthcare version, operating on the same principle. The system that touched the family touched the rest of the family’s developmental trajectory, because the system spoke the same language the family was trying to speak. Most pediatric care in the United States does not operate this way. The Nurse-Family Partnership shows what the alternative looks like, with fifteen years of receipts.

What it doesn’t answer: Whether NFP effects are replicable in current implementation conditions. The original trials ran in well-controlled academic-affiliated settings. Recent large-scale rollouts have produced more modest effects, raising the implementation question that bedevils every coordinated intervention. The model works. The conditions for the model are hard to maintain at scale.

Coming Up

Phase 2 wraps with this post. Phase 3 opens next week with the Coming Up tease still to be set by Victor and Gabriele; that note will be inserted before publish.