The Compliance Goal Problem: When 'Following Directions' Isn't a Real Skill

Compliance is not a skill, it is a side effect. How to tell when a 'following directions' goal is socially significant, from a BCBA-led CEU.

Key takeaway

Take Tony. He is 13, in a specialized school, and his team has worked on "following directions" for years. He still bullies peers. Not because he missed a compliance lesson.

Watch the full CEU recording

The Ethics of Socially Significant Goal Selection - Applied 2023

Kaelynn Partlow · 1 CEU · 57 min
Watch on openceu.com →

Take Tony. He is 13, in a specialized school, and his team has worked on "following directions" for years. He still bullies peers. Not because he missed a compliance lesson. He bullies because a peer says something that lands wrong, he spits in their face before he can think, the peer reacts, and Tony has no way to climb back out of the hole. His real deficit is a recovery skill, not obedience. Every "comply with adult instructions" goal on his sheet was teaching the wrong thing.

That is the compliance goal problem in one kid. A Board Certified Behavior Analyst (BCBA) writes "will follow directions in 4 out of 5 trials" because it is easy to measure and easy to defend at peer review. But compliance is not a skill. It is a side effect of a hundred other skills working together. When we target it directly, we usually teach the learner to mask, not to cope.

Why compliance shows up on so many goal sheets#

Compliance goals are everywhere because they fit the system. They are easy to write, easy to graph, and easy to show a funder. A four-year-old who sits when told gets a check mark. A teenager who answers "yes ma'am" gets a check mark. The data looks clean.

The problem is that the goal sheet starts to drift away from the kid. We pick the next target on the assessment because the box is open, not because the skill changes anything in the learner's actual day. The 2018 Ferguson review of every Journal of Applied Behavior Analysis study from 1968 through 2018 found that only 12 percent measured social validity at all. Of those, only 12 percent looked at whether the goal itself mattered. That is 17 papers out of more than 1,200 asking, "is this thing we are teaching worth teaching?"

When the field has barely looked at the question in print, it makes sense that our goal sheets default to compliance. It is the path of least resistance.

What "compliant" actually means (and who benefits)#

Strip the word down and "compliant" means the learner does what an adult says, when the adult says it. That is useful for the adult. It is sometimes useful for the kid. It is almost never useful for the kid in five years.

Here is the test. Picture the learner at 18, in the place their family hopes they will be. A community college class. A grocery job. A roommate situation. Walk through their day. How many moments require pure compliance, and how many require something more layered, like reading a room, asking a clarifying question, recovering from a mistake, or saying no?

The answer is almost always "more layered." Adult life is not a clinic. Nobody is running a 4-to-1 positive-to-negative feedback ratio at the deli counter. If our learner only knows how to follow directions, every unscripted moment is a cliff edge.

You better believe in public high school, they are not doing four-to-one positive to negative ratios. Can they independently navigate the school? Do they know how to get lunch? Can they handle the instructor-to-student ratio? Can they learn behaviors from their peers? Can they accept feedback from adults that they're not familiar with? From the talk — Kaelynn Partlow

That list is the real curriculum. None of those bullets are "compliance." Every one of them is a separate skill we can teach.

The five real skills usually hiding inside a compliance goal#

When a team writes "Johnny will comply with adult instructions in 8 out of 10 opportunities," they almost always mean one of these five things. Pulling them apart gives you a real target.

  1. Following a multi-step direction. This is a listening and working-memory skill. Teach it like that, with growing chain length and fade prompts. It is not obedience. It is processing.

  2. Tolerating a "no" or a wait. This is frustration tolerance and self-regulation. The kid is not "non-compliant" when they push back on "wait five minutes." They are practicing a coping skill they do not have yet.

  3. Asking for help or clarification. A learner who freezes when an instruction is unclear does not need more compliance trials. They need a script for "can you show me?" or "what do you mean?"

  4. Recovering from a mistake. This is Tony's missing skill. When the learner does the wrong thing, what do they do next? Apologize? Ask what to try? Walk away and reset? That is a teachable repertoire, and it is more important than the original misstep.

  5. Choosing to cooperate because it pays off. This is the long game. The learner cooperates because the relationship is good, the activity is fun, and the reward is worth it. You build this by stacking reinforcement and respecting assent, not by drilling "do this" until they break.

Pick one of those five for your goal sheet. Drop the word "compliance."

Tony's case: writing a real replacement instead#

Back to Tony. His old goal probably read something like "Tony will respond to adult redirection without verbal aggression in 4 out of 5 opportunities." That goal taught Tony nothing. It only counted how often he failed.

A real replacement starts with the function. Tony bullies when he feels socially cornered and cannot get out. The skill he needs is conflict recovery: a way to interrupt the spit-or-yell loop, slow down, and try again.

Tony's bullying behaviors often arise when he perceives conflict. A peer says something that sets Tony off. Without thinking, Tony immediately responds by spitting in their face. His classmate retaliates with corrective feedback. Tony realizes he's made a significant social mistake and has no way to recover from it. From the talk — Kaelynn Partlow

A goal that actually helps Tony might look like:

When Tony notices he is upset with a peer, Tony will use one of three taught recovery strategies (request a break, name the feeling out loud, or use a repair phrase) before responding, in 3 of 5 logged conflict moments per week.

That goal has a function. It has a context. It teaches a skill Tony will use at 13, at 23, and at 53. And the next time peer review asks why it is not "compliance with adult correction," the answer is right there in the goal: because compliance with correction does not stop the spitting. Recovery does.

If a problematic situation is not moving the learner toward their goals and is in fact moving them further away from their values, what skill do I need to teach that, if it were present, would diffuse or prevent the situation from occurring in the first place? From the talk — Kaelynn Partlow

Screenshot that one. It is the cleanest goal-writing question in the field.

How to talk to a team that insists on a compliance target#

Sometimes the parent wants compliance on the sheet. Sometimes the school does. Sometimes a supervisor does. Pushing back without sounding preachy is its own skill.

Start with the value behind the request, not the wording. A parent who asks for "more compliance" usually wants their kid to be safe in public, easier to take to the grocery store, and welcome at family gatherings. Those are real values. Restate them out loud. "It sounds like the goal here is that he can go to Grandma's without a meltdown. Did I get that right?"

Then offer the upgrade. "I want that too. The way I would write it is, 'Will accept a redirection from a familiar adult and return to the activity within two minutes,' because that is the actual skill we need at Grandma's. Pure compliance would help in the clinic, but it usually does not travel."

This works because you are not arguing about ethics. You are arguing about what will actually move the needle in their life. Bridge the gap between the language they brought and the language that works.

If a Behavior Intervention Plan (BIP) is in play, this gets easier. The BIP forces you to name the function of the behavior. Once the function is on paper, "more compliance" stops being a serious answer because it does not address the function.

Three compliance goals to retire this quarter#

Go through your active goal sheets. Find these three and rewrite them.

  1. "Will follow one-step directions in 8 of 10 trials." Rewrite as a specific listening or tolerance skill. What direction? In what setting? Why does the learner need this skill in their life this year?

  2. "Will say 'good game' after losing in 4 of 5 opportunities." Almost no real teenager says "good game" sincerely. Replace with a regulation goal ("Will name one thing they enjoyed after the game ends, in 3 of 5 sessions") or drop the target. Kaelynn flagged this one specifically. Even in four-year-olds, it is hard to defend.

  3. "Will comply with adult redirection without protest." Rewrite as a self-advocacy or recovery goal. The protest is information. The protest is usually the kid telling you the demand is too high, the task is unclear, or the reinforcer is not worth it. Teach them to say that with words.

Three swaps. Same data system. Better outcomes.

Frequently asked questions#

Isn't safety compliance still important?

Yes, and it is also not what we are talking about. Stopping at a curb, taking a hand at a parking lot, getting out of a bathtub when told. Those are non-negotiable safety routines, and they should be taught with strong rapport and clear reinforcement. The line is whether the goal exists to protect the learner from a real, immediate danger. If yes, keep it and call it a safety skill. If the "safety" framing is being used to justify drilling generic obedience, that is the compliance goal problem in a different hat.

What about instructional control, is that the same thing as compliance?

No. Instructional control is the relationship-and-reinforcement state where the learner wants to work with you because the activity is fun and the payoff is real. You build it by pairing, by following the learner's interests, and by making demands worth meeting. Compliance is the output you get when control is in place. You cannot teach instructional control by teaching compliance any more than you can teach trust by demanding it.

How do I write a goal that builds cooperation without building compliance?

Anchor the goal to a context the learner actually cares about. "Will accept a teacher prompt to clean up before recess in 4 of 5 opportunities" is about cooperation in a moment the kid wants something. "Will accept any prompt from any adult" is about obedience in a vacuum. The first one builds a skill that travels. The second one builds masking.


If you want to hear Kaelynn walk through Tony's case in her own words, plus AnnaMarie Stoudemire's full framework for picking goals that actually serve the learner, the talk is one continuing education unit (CEU) and it is free.

Watch The Ethics of Socially Significant Goal Selection