Why ABA Skills Don't Transfer Home
Why ABA Skills Don't Transfer Home (and How to Plan for Generalization)
Have you ever watched a child nail a skill in session, week after week, and then heard a parent say, "He never does that at home"?
It's one of the most common frustrations in our work, and it's easy to wonder what went wrong. Maybe the parent isn't practicing. Maybe the skill wasn't really mastered. Usually, the answer is simpler than that. The skill was taught in one place, with one person, using one set of materials, and nobody planned for it to show up anywhere else.
That's a generalization problem. The good news is that our field has been studying how to solve it for almost 50 years.
What Generalization in ABA Actually Means
Generalization is when a skill shows up beyond the exact conditions where it was taught. That can look like:
Across settings: the skill learned at the clinic table also happens at the kitchen table.
Across people: the child responds to Mom, Grandpa, and a teacher, not only the therapist.
Across materials: a child who learned to wash hands at one sink can do it at a different sink.
Over time: the skill is still there a month after you stop teaching it (often called maintenance).
For families, generalization is the whole point. A skill that only lives in a therapy room doesn't change much about daily life.
Why Skills Don't Transfer on Their Own
In 1977, Trevor Stokes and Donald Baer published a paper in the Journal of Applied Behavior Analysis that still shapes how we think about this. They reviewed the research and noticed that the most common approach to generalization was to teach a skill, then hope it showed up elsewhere. They called it "train and hope."
Their central point was that generalization rarely happens by accident. It has to be programmed, the same way we program any other part of treatment. If we want a skill at home, the plan needs to say how it's going to get there.
Stokes and Baer laid out a set of strategies for doing that. Here's how they translate into everyday practice.
How to Plan for Generalization: 7 Strategies from Stokes and Baer (1977)
1. Teach skills that will be reinforced in real life
Choose and shape skills that the natural environment will keep reinforcing. Requesting a snack gets a snack. Greeting a sibling gets a smile back. When everyday life pays off the skill, it has a reason to stick around after you leave.
2. Use lots of examples
One example teaches one example. Teaching "wash hands" at three sinks, or "put on shoes" with sneakers, boots, and sandals, gives the child enough variety to handle the next new version on their own.
3. Train loosely
This is the strategy most programs skip. Training loosely means varying the things that don't matter: your exact words, your tone, where you sit, the time of day, the materials. Tight, identical teaching trials are great for acquisition, but they can teach a child that the skill only happens under those exact conditions. Real life is loose, so teaching should be too, once the skill is emerging.
4. Bring home into the session
Stokes and Baer called this programming common stimuli. Use the family's own cup, plate, toothbrush, or favorite toy during teaching. Better yet, teach in the actual routine. The more the teaching setting looks like the place where the skill needs to happen, the more likely it is to transfer.
5. Make reinforcement less predictable
If a child can tell exactly when reinforcement is coming, behavior can drop off when those cues disappear. Gradually thinning and varying reinforcement makes it harder to tell which moments "count," which helps the skill hold up in settings where reinforcement is less frequent.
6. Give the child a bridge
Some skills transfer better with a tool the child carries from place to place: a visual schedule, a short phrase they say to themselves, a picture card. That tool becomes a common link between the teaching setting and everywhere else.
7. Make generalization the goal
You can reinforce generalization itself. When a child uses a skill in a new way or a new place, notice it and celebrate it. Over time, the child learns that trying a skill in new situations is worth it.
Parents Are the Generalization Plan
Here's the part I want every BCBA to hear. The single most powerful generalization strategy you have is the parent.
Parents are present in the settings, routines, and moments where skills need to show up. When they can use the strategies confidently, every one of those Stokes and Baer strategies gets easier: the routine is real, the materials are the family's own, and the examples come built in.
Research supports this. A 2016 data-based case study by Crone in Education and Treatment of Children trained parents of children with autism to use a function-based behavior plan during mealtimes, comparing home-based and clinic-based parent training. Parents' use of the strategies increased and carried over into their real mealtime routines, and children's problem behavior during meals decreased.
The takeaway for practice: train parents in or around the routine where the skill needs to happen, and measure success in that routine, not just in session.
Why "Practice This at Home" Homework Often Falls Flat
Most of us have sent a family home with a list of things to practice. Most of us have also heard, the next week, that it didn't happen. Most parents want to do it. A vague instruction to "practice at home" leaves a parent guessing about when, where, and how.
What tends to work better:
One routine, one strategy. "During bath time, give two choices before each step" is doable. "Use choices throughout the day" is not.
Practice it together first. Model it, have the parent try it, and give warm, specific feedback before they leave.
Plan for the hard moment. Talk through what to do if the routine goes sideways.
Follow up on the routine itself. Ask how bath time went, not just whether they practiced.
For more on this, Getting Caregivers to Follow Through Outside of Sessions walks through follow-through in detail, and How to Make Parent Training Feel Less Like a Lecture covers moving from explaining to skill-building.
Three Signs a Skill Has Actually Generalized
How do you know it worked? Look for these:
It shows up without you. The skill happens with new people, places, or materials, without extra prompting.
It sticks after you fade. The skill holds steady after you reduce prompts and reinforcement.
The parent tells you about it first. "You won't believe what she did at the grocery store" is some of the best data you'll ever get.
The Whole Point of ABA
Every skill we teach is meant to make daily life easier for a child and their family, and generalization is how that happens. When we plan for it from the first session, and bring parents in as partners in that plan, the skills we teach have a real chance of changing family life.
Want to Go Deeper?
The Parenting with ABA CE Membership includes a full library of parent training CEUs, with practical tools for coaching caregivers so skills actually carry over into home routines. Explore the CEU library here.
