Mastery criteria are the standards a child has to meet before an ABA goal counts as mastered. They are written into the goal in advance and usually combine three things: a level of accuracy (for example, 80% or 90% of trials), a number of sessions in a row, and sometimes a number of different people or settings. The best criteria also ask whether the skill holds up over time. This guide explains how those pieces fit together, what the research says about choosing them, and what questions parents and new clinicians can ask.

TL;DR

• A mastery criterion is a performance threshold set before teaching starts, such as a percentage of correct trials across a set number of sessions.
• 80% is a common benchmark, but research on criterion level found that a 90% criterion produced more stable maintenance than lower ones.
• Mastery is a milestone, not a finish line. A good plan includes generalization and maintenance checks.
• Criteria should fit the child and the skill. Safety and self-help skills often warrant stricter targets than some social goals.
• Parents can ask to see the criterion written on any goal, and how the team will check that the skill lasts.

Table of Contents

• What mastery criteria are, in plain terms
• The parts of a well-written criterion
• Is 80% mastery?
• Mastered once is not mastered
• How the data are collected changes what mastery means
• What the ethics code expects
• Examples of mastery criteria in action
• A note on performance goals vs. mastery goals
• Perspective: what we look for before closing a goal
• Questions to ask your child’s team
• Where we can help
• FAQ
• Sources

What mastery criteria are, in plain terms

When a BCBA writes a skill acquisition goal, the goal needs an endpoint. “Child will request a snack” is a direction. “Child will request a snack independently in 90% of opportunities across three consecutive sessions with two different staff” is a goal you can actually close, because everyone knows what the data must show.

A 2024 study of behavioral target mastery describes criteria as defined by the percentage of trials, the number of trials, and the number of successful sessions. It also notes that 80% accuracy remains a common benchmark in clinical research, while stressing that it is always subject to clinical judgment about the specific behavior and learner.

That last part matters. There is no single number that works for every child and every skill.

The parts of a well-written criterion

Most criteria are built from the same few ingredients. Knowing them lets you read any goal sheet.

Accuracy level

This is the percentage of opportunities where the child responds correctly or independently. Common choices run from 80% to 100%. A review of the mastery criteria literature cites recommendations from practitioners of 80% to 100% correct for two or three consecutive sessions, and another of 80% to 90% over two to three days with at least two different teachers, adapted to the learner.

Number of sessions

One great session can be luck, a good mood, or a favorite reward. Requiring the same performance across consecutive sessions makes it less likely that a single good day closes a goal.

Number of trials

Percentages mean little on tiny samples. Getting 4 of 5 trials right is 80%, and so is 16 of 20, but the second tells you far more. Many teams set a minimum number of opportunities per session so a percentage is not based on a handful of trials.

Independence

A response that needs a prompt is not yet independent. Criteria usually specify whether prompted responses count as correct. In most programs they do not.

People and settings

Requiring more than one instructor, and sometimes more than one location, builds generalization into the goal. If a child performs only for one therapist at one table, that is useful information, and the goal is not done.

Is 80% mastery?

Sometimes, and the honest answer is that it depends. Eighty percent is a widely used starting point, and it is reasonable for many skills. But it is not a rule.

Research gives a reason to think harder about it. A parametric analysis of mastery criterion level compared criteria of 50%, 80%, and 90% accuracy. Skills taught to a 90% criterion were maintained more reliably: after the maintenance interval, performance on those skills ranged from 70% to 100% accuracy, versus 20% to 80% for the lower criteria. The study was small and preliminary, so it is a signal, not a mandate. But it is a good reason not to treat 80% as automatically enough.

A descriptive analysis of applied research procedures also points out a gap between research and practice. Researchers often use stricter accuracy over fewer observations, while practitioners often use looser accuracy over more observations. The authors caution that lower standards can trade long-term maintenance for quicker progress reports. They also report that only two studies had systematically tested how any dimension of mastery criteria affects maintenance, so there is still a lot we do not know.

A simple way to think about how strict to be

This table is a way of reasoning, not a standard from a rulebook. The BCBA on your child’s case decides, and should be able to explain why.

Mastered once is not mastered

The most common mistake in goal tracking is closing a target the moment the data hits the line. The goal is not the number. The goal is a skill your child can use at snack time, at school, and with grandma.

The 2024 study on target mastery examined maintenance failure as well, a reminder that skills can slip after a goal closes. Experts also caution against relying blindly on data software that closes targets automatically. If a child “masters” a goal but cannot do it when the instruction, the setting, or the person changes, the criteria may have been too narrow, or the generalization step was skipped.

A solid plan therefore includes:

1. Acquisition: the child reaches the written criterion.
2. Generalization: the skill is probed with new materials, new people, and new places.
3. Maintenance: the skill is rechecked after a gap, with no teaching, to see if it stuck.
4. Reopening: if the skill has faded, the team adjusts the teaching or the criterion.

The CDC’s overview of autism treatment is a useful parent-level reference for the idea that autism interventions should be tailored to the individual child.

How the data are collected changes what mastery means

Mastery depends on measurement. If you only record the first trial of each session, a skill can look mastered faster than it really is. A comparison of continuous and discontinuous measurement in programs for children with autism evaluated how different data collection frequencies affected acquisition and maintenance. Continuous, trial-by-trial data is generally preferred because it gives a fuller picture of the child’s learning. We cover what that looks like in our guide to trial-by-trial data in ABA.

If you want a parent-friendly look at the numbers behind the graphs, see how behavior analysts collect data. For telling a real trend from a bad day, our piece on ABA progress tracking is a good companion, and the role of data collection in ABA therapy covers the basics.

What the ethics code expects

For Board Certified Behavior Analysts, this is not just good practice. The BACB Ethics Code for Behavior Analysts expects analysts to collect data correctly, monitor interventions continuously, and take corrective action when the data show that desired outcomes are not being reached. In practice that means a mastery criterion is a decision tool. If a child sails through targets, the criteria may be too easy. If a child stalls on a target for months, the criteria, the teaching procedure, or the target itself may need to change.

For people training to become behavior analysts, writing measurable goals is part of the exam blueprint. The BCBA Test Content Outline (6th ed.) lists developing intervention goals in observable and measurable terms.

Examples of mastery criteria in action

These are illustrative examples, not clinical recommendations for any child.

Requesting. “Alex will independently request a preferred item using a word or approximation in 90% of opportunities, across three consecutive sessions, with at least two staff members.” Maintenance probe: one week and then one month after mastery, with a parent presenting the opportunities at home.

Hand washing. “Maya will complete all steps of the hand-washing routine without prompts on four of five consecutive days.” Because this is a daily-living skill, the criterion is stricter and is checked in both the clinic bathroom and the home bathroom.

Greeting peers. “Sam will greet a peer within five seconds of arrival in 80% of opportunities, across two settings.” Because the social context varies, the team counts different peers and locations rather than a single scripted drill.

Notice what each includes: a level, a time frame, a way of handling prompts, and a plan for what happens after mastery.

A note on performance goals vs. mastery goals

The phrase “mastery goal” shows up in two worlds. In ABA, it means a skill target with a written criterion. In education and psychology more broadly, a mastery goal contrasts with a performance goal: it is about learning something for its own sake rather than to look good or earn a reward. The two ideas overlap in spirit, since both want real learning, but a clinical goal sheet is about measurable behavior.

Perspective: what we look for before closing a goal

At Building Block Resolutions, we have spent more than 20 years watching what happens after the graph hits the line. The honest truth is that the number is the easy part. The hard part is making sure the skill belongs to the child and not to the session.

We write criteria individually. Our plans are built around each child’s challenges, and that includes how strict a criterion should be for that skill. We want to know whether a skill shows up with a different person, in a different room, and on a Tuesday when the child is tired. That is why we involve parents and caregivers in both teaching and checking: families see the skill in the places that matter most.

We also try to be candid when a goal closes too fast or a target will not move. Neither is a failure. Both are data. A criterion is a hypothesis about what “learned” looks like for this child, and we are always willing to revise it.

— Jennifer

Questions to ask your child’s team

You do not need a behavior analysis degree to ask good questions. These are a few that tend to open useful conversations:

• What is the written mastery criterion for this goal, and why was that level chosen?
• Do prompted responses count toward the criterion?
• Will the skill be tested with other people and in other places before the goal closes?
• How and when will you check that the skill is maintained after mastery?
• What happens if my child masters goals very quickly, or very slowly?
• Can I see the trial-by-trial data, not just the summary graph?

A team that welcomes these questions is a team that is thinking about real progress.

Where we can help

If you would like to see how individualized goals and criteria work in practice, our ABA autism therapy programs pair tailored treatment plans with parent coaching, so skills carry over to home. You can also browse our full list of services, including remote ABA therapy for families who prefer sessions at home. Reach out through our site to book an appointment and talk through your child’s goals.

FAQ

Can you give an example of a mastery goal?

Yes. “The child will independently label 20 common objects with 90% accuracy across three consecutive sessions with two different instructors” is a typical shape. It states the skill, the accuracy, the number of sessions, and the number of people. Many teams then add a maintenance check weeks later.

Is 80% considered mastery?

It is a common benchmark, and recent clinical research still uses it, but it is not universal. Complex or safety-related skills often call for higher criteria. Your BCBA should choose the level based on the skill and your child.

What are the 5 steps of DTT?

Discrete trial teaching is usually described as five parts: the instruction (antecedent), the child’s response, the consequence, a brief pause, and then the next trial. Different programs word the steps a little differently, and prompting is often folded into the first or second part. Ask your child’s team how they run trials and record each one.

How do I know a mastered skill will last?

Look for a plan that tests the skill after mastery: with new people, in new settings, and after a break from teaching. If performance drops, a good team revisits the teaching or the criterion rather than leaving the goal closed. Asking to see maintenance data is completely reasonable.

Sources

• A Preliminary Analysis of Mastery Criterion Level: Effects on Response Maintenance
• Mastery Criteria and Maintenance: a Descriptive Analysis of Applied Research Procedures
• Predicting Behavioral Target Mastery With Age, Intensity and Duration, Open Targets, and Maintenance Failure With ABA
• Mastery criteria and the maintenance of skills in children with developmental disabilities (Behavioral Interventions, 2021)
• Evaluating Progress in Behavioral Programs for Children with Autism Spectrum Disorders Via Continuous and Discontinuous Measurement
• Ethics Code for Behavior Analysts (BACB)
• BCBA Test Content Outline, 6th ed. (BACB)
• Treatment and Intervention for Autism Spectrum Disorder (CDC)

Recommended

• How ABA Improves Daily Living Skills for Kids
• ABA Progress Tracking: Spot Reliable Data, Not Session Noise
• Treatment Fidelity in ABA: Six Steps Supervisors Can Use in Practice
• Clinic to Home Generalization Checklist: ABA for Clinicians & Caregivers