When the Parenting Research You’re Handed Doesn’t Hold Up

Somewhere in the stack of training materials, agency handouts, therapist recommendations, and well-meaning advice you’ve collected as a foster parent, adoptive parent, or stepparent, there is research. Some of it has shaped how you think about your child. Some of it has shaped what you’ve been told to do, what to watch for, and what your child’s behavior means.

I want to tell you something that took me too long to figure out. A lot of that research is shakier than the researchers will admit.

In April 2026, a research team published the largest replication project ever attempted in the social and behavioral sciences. They tested 274 claims drawn from 164 papers. Just over half held up when independent researchers tried to reproduce the results. You can read the headline study here. When findings did replicate, the effect sizes were typically about half the size originally reported.

That number, fifty-five percent, lines up with a decade of similar projects. About half of social-science research, on average, does not survive independent replication. And the half that fails is not random. The most exciting findings, the ones that get cited and built into training programs and packaged into parenting books, are often the ones least likely to hold up.

I’m not telling you this to make you cynical. I’m telling you because Plan B parents, foster parents and adoptive parents and stepparents alike, have spent years being handed research-backed frameworks as if they were settled science. Some of those frameworks are sound. Some are oversimplified versions of sound research. And some are not as well-supported as the people teaching them seem to believe.

What Replication Means and Why It Matters for Your Family

A study replicates when an independent team uses the same methods on a fresh sample and gets substantially the same answer. When studies fail to replicate, the original finding was probably exaggerated, found by chance, or built on choices that bent the data toward a particular result.

The reasons are understandable. Researchers face career pressure to produce striking findings. Journals reward novelty. Out of dozens of small analytical choices in any study, the ones that produce significant results get reported, and the ones that don’t, often don’t. None of this requires bad faith. It only requires the ordinary human tendency to keep going when the answer looks good and stop when it doesn’t.

For you, the practical question is this. When a caseworker, a therapist, or a training program tells you that “research shows” something about your child, what do you actually know? You know that someone, somewhere, found a result that got published. You don’t know whether anyone has tried to reproduce it. You don’t know how big the effect was. You don’t know whether the study even involved children like yours.

That is not a reason to dismiss research. It is a reason to keep your own observations and your own judgment in the picture.  You may not have a lot of fancy degrees, but you are the expert about your child.

The ACE Score You Were Probably Handed

If you have been a foster parent, adoptive parent, or stepparent for any length of time, someone has probably told you about ACEs, the Adverse Childhood Experiences score. The framework comes from a 1998 study that asked adults to recall ten categories of difficult childhood experiences and tracked their adult health outcomes. The big finding, which other studies have replicated, is real. At the population level, more childhood adversity correlates with worse adult health.

That part of the science is solid. However it does not translate into the current fad of “ACEs scores” for individual children.

Researchers never designed the ACEs analysis to predict what would happen to a particular child. They designed it to identify a population-level pattern. The senior author of the original ACE study, Robert Anda, published a 2020 commentary explicitly warning against using the score as a clinical screening tool.  No one has ever standardized it as a measurement of stress exposure, and we should not apply population-level patterns to individuals.

A 2020 study using two long-running birth cohorts tested whether ACE scores could actually predict individual outcomes. They couldn’t, not reliably. The score forecast group averages reasonably well and individual children’s outcomes poorly. There is also a 2019 meta-analysis that compared adults’ retrospective recall of childhood maltreatment with what was actually documented at the time. Agreement was 0.19 on a scale where 1.0 would be perfect. The two are largely measuring different things.

This matters because some agencies, schools, and even therapists treat the ACE score as if it tells them something specific and predictive about a child. It doesn’t. If your child has a high ACE score, that does not mean they will have a particular outcome. If your child has a low ACE score, that does not mean nothing is wrong. The score is a research instrument, not a diagnosis, not a prognosis, and certainly not a roadmap for how to parent the child in front of you.

What Actually Holds Up: The Other Side of the Same Research

Here’s where the research gets more useful.  In recent years, researchers have started measuring not just adverse childhood experiences but positive ones. Those are things like feeling that family supported you, having a sense of belonging at school, and having two non-parent adults who genuinely cared about you. The framework is sometimes called Positive Childhood Experiences, sometimes Benevolent Childhood Experiences. A 2019 study found these experiences predicted adult mental and relational health in a dose-responsive way, and the effects held up even when the same individuals reported significant adversity.

The adverse experiences did not erase the protective ones. They operate on their own. That is genuinely useful research, because it points toward what we can actually do.

For a foster parent, that means the consistent supportive presence you provide is doing real work even if the child also has serious adversity in their history. For an adoptive parent, it means the belonging and stability you build is not undone by what came before. For a stepparent, and this is something I think gets too little attention, it means that becoming one of those caring non-parent adults in a child’s life has independent value, even when the bond is complicated and even when you are not the primary parent. The research does not require you to fix the past. It suggests that what you provide now matters in its own right.

I don’t have a tidy formula for how this plays out day to day. What I can tell you is that the research aligns with what most experienced Plan B parents eventually figure out anyway. Showing up consistently, building belonging slowly, and being one of the steady adults in a child’s life matters in a profound way. These things accumulate, even when they don’t feel like enough.

What About the Bigger Frameworks? Trauma-Informed Care, SEL, and the Rest

Beyond the ACE score, a whole industry of training programs and parenting frameworks promises to help us raise children from hard places. Some are genuinely useful. Others are weaker than their marketing suggests.

Trauma-informed care is the clearest example.  A 2024 federal systematic review concluded that evidence was insufficient to make clear determinations about its effectiveness on health outcomes. A 2019 review of trauma-informed approaches in schools, conducted under one of the most rigorous review methodologies available, found zero studies meeting basic methodological inclusion criteria. The framework gets used everywhere. The evidence base behind it is thinner than that suggests. The underlying instincts are sensible. The brand-name programs do not deliver the transformations their marketing implies, and you do not need to feel inadequate when they don’t.

Other categories have more evidence, though the effects are modest. A 2023 meta-analysis found social-emotional learning programs produce consistent positive effects when implemented as designed. A 2019 meta-analysis found that one-on-one mentoring produces small but real positive effects, with relationship quality mattering more than duration. The pattern is the same across categories. Steady, consistent, well-delivered support helps. Magic-bullet programs are mostly marketing.

What This Looks Like for You as a Plan B Parent

When you’re handed a research-backed claim about your child, your child’s behavior, or how you should parent, here are the questions worth asking. Not out loud, necessarily. In your own head, as you decide how much weight to give what you’ve just been told.

1. Has anyone tried to reproduce this finding? If a therapist or trainer is working from a single dramatic study, that is much weaker than working from a result that has held up across many studies and many research teams.
2. How big is the effect? Many genuine findings in this area are small. They matter at the population level but don’t predict much about a particular child.
3. Was the study about children like mine? A finding from a study of suburban middle-class families may not generalize to a child whose life has included instability, foster care, or significant adversity. A finding about young children may not apply to teenagers. A finding from one cultural context may not transfer to another.
4. Does this match what I actually see in the child? You know your child in ways that researchers cannot. If a framework keeps telling you something that does not match what you observe, the framework is not necessarily right and you are not necessarily wrong.

I am not suggesting you become a research methodologist. I am suggesting that you view expert claims with the same calm skepticism you have probably already learned to bring to a lot of advice that comes your way as a Plan B parent. Trust your observations. Ask questions. Notice when an answer feels more confident than the evidence supports.

The Bottom Line

About half of social-science research does not replicate. The ACE framework is real at the population level and not particularly useful as an individual screening tool. Positive childhood experiences and steady caring adult relationships have research support that maps directly onto what foster parents, adoptive parents, and stepparents actually do. The big-name trauma-informed and behavioral interventions vary widely in evidence quality, with implementation fidelity mattering far more than brand.

You don’t need to be alarmed by this. You don’t need to throw out everything you’ve been taught. What you can do is keep your own judgment, your own observation of the child, and your own accumulated experience in the picture. The science of childhood adversity is real. The packaging it sometimes comes in is less reliable than it looks. The children we are parenting deserve our best thinking, not just the latest framework.