What Should You Measure in ABA Parent Training?

When BCBAs think about parent-training data, the first thing that often comes to mind is caregiver fidelity.

Did the parent follow the steps correctly?

Did they use the strategy in 80% of opportunities?

How much prompting did they need?

Those data can be useful. But they answer only one question:

Can the caregiver perform the skill?

Parent training can affect much more than procedural accuracy. A parent may learn to implement a strategy correctly while still feeling overwhelmed by it. The strategy may work during the session but be nearly impossible to use while the parent is making dinner, helping siblings, or getting everyone out the door. The child may improve while the parent becomes more stressed. Or the caregiver may become much more confident even before there is a large change in child behavior.

If those outcomes matter, they need to be measured too.

Research on caregiver interventions has included outcomes such as parenting stress, caregiver strain, parenting competence, family quality of life, treatment acceptability, caregiver implementation, and child behavior. The important question is not simply, “What data should I collect during parent training?”

It is:

“What are we trying to change?”

Caregiver fidelity is useful, but it is not the whole outcome

Treatment integrity tells you whether the caregiver is implementing a procedure as intended.

That matters when the caregiver goal is actually about implementing a procedure.

A recent scoping review of caregiver-implemented ABA interventions for challenging behavior specifically examined studies that included caregiver treatment-integrity data. Across the included literature, caregivers were often successfully taught to implement behavioral interventions, with many studies reporting high levels of integrity following training (O’Neill & Koudys, 2025).

But high fidelity does not tell you:

• whether the parent is using the strategy when the BCBA is not present;

• whether it fits the family's routines;

• whether the child is benefiting;

• whether the parent feels more confident;

• whether parenting is becoming less stressful;

• whether the family thinks the strategy is worth continuing.

A caregiver can score 90% on a fidelity checklist and still tell you:

“I can do this, but there is no way I can keep doing it every morning.”

That is important data.

Measure caregiver skill when caregiver skill is the target

If the goal is skill acquisition, measure the caregiver behavior directly.

Depending on the goal, that could include:

• percentage of procedure steps completed independently;

• correct responses during role-play or scenarios;

• number of naturally occurring opportunities implemented correctly;

• amount of BCBA prompting required;

• accuracy selecting an appropriate strategy;

• ability to identify when the strategy should be used;

• independent problem solving when a new situation comes up.

The measure needs to match the goal.

If Maria's goal is to recognize when to prompt Eli's break request, a checklist measuring treatment fidelity may not be the best first measure. It may be more useful to present several scenarios and measure whether Maria correctly identifies when prompting is appropriate.

If Daniel's goal is to carry out a multi-step teaching procedure, a fidelity checklist makes much more sense.

This is why measurable caregiver goals matter. For more examples of matching measurement to the caregiver behavior, see How to Write Measurable ABA Parent Training Goals.

Measure whether the skill is being used outside the session

Demonstrating a skill with the BCBA present and using it during daily life are not the same outcome.

A parent may perform a prompting procedure accurately during coaching but rarely use it during the week.

That difference is clinically useful.

Rather than treating it immediately as a consistency problem, measure what is happening.

You might track:

• number of relevant routines in which the strategy was used;

• number of days used;

• percentage of opportunities;

• caregiver self-monitoring;

• brief examples of when the strategy was and was not used;

• barriers that occurred when the caregiver intended to use it.

These measures do not need to become burdensome.

If recording every opportunity creates more work than the intervention itself, the data system probably needs to change.

The goal is to learn whether the skill is making it into the conditions where the family actually needs it.

Measure whether the child is benefiting

Parent training is not successful simply because the caregiver can perform the procedure.

The intervention should also produce the outcome it was intended to support.

If Maria learns to prompt functional communication accurately, you may also want to know whether Eli is communicating more independently.

If Daniel learns how to support transitions, you may want to know whether transitions are becoming easier.

If a caregiver learns to arrange communication opportunities, you may measure whether the child begins initiating more often.

Caregiver behavior and child behavior answer different questions.

For example:

Caregiver outcome:
Maria implemented the communication procedure independently in 9 of 10 opportunities.

Child outcome:
Eli independently communicated in 7 of 10 relevant opportunities.

Together, those numbers are much more informative than either one alone.

Measure whether family life is actually getting better

This is one of the places where ABA parent training can become too narrow.

Suppose the family originally asked for help because bedtime takes two hours and everyone is exhausted.

The BCBA teaches the parent a bedtime procedure. The parent implements it at 90% fidelity.

That sounds successful.

But bedtime still takes two hours.

Or it now takes 90 minutes, but the parent reports that following the procedure requires so much monitoring and preparation that evenings feel even more stressful.

Those outcomes belong in the clinical picture too.

Routine-specific measures can be very simple:

Bedtime

• duration of the routine;

• number of times the parent needs to return to the room;

• parent's rating of how manageable bedtime felt.

Morning routine

• time required to get out the door;

• number of caregiver prompts;

• whether the visual was actually practical to use.

Dinner

• whether the caregiver can prepare dinner without repeatedly stopping;

• whether the child can remain engaged independently for a meaningful period;

• parent's rating of how difficult the routine felt.

These are not less important because they do not look like traditional ABA acquisition graphs.

They may be much closer to the outcome the family originally wanted.

Consider measuring parenting stress

Parenting stress is an outcome BCBAs may not routinely measure, but there are good reasons to consider it.

Parent-training research has measured parenting stress alongside child outcomes and caregiver skill. In a randomized trial involving 180 children with autism and disruptive behavior, Iadarola and colleagues compared behavioral parent training with parent psychoeducation. They measured parenting stress, caregiver strain, and parenting competence. Parents receiving behavioral parent training showed greater improvement in parenting competence, caregiver strain, and the Difficult Child domain of the Parenting Stress Index than parents receiving psychoeducation (Iadarola et al., 2018).

At the same time, the broader evidence tells us not to assume that parent intervention automatically reduces stress. A 2019 meta-analysis found a favorable effect of behavioral parent interventions on parenting stress (Tarver et al., 2019), while a later meta-analysis of 37 parent-intervention studies found improvements primarily in parenting confidence and mental health, but not in parenting stress or caregiving burden (MacKenzie & Eack, 2022).

That inconsistency is one reason measurement matters.

If parent training is supposed to make daily life more manageable, it is useful to know whether stress is actually changing rather than assuming it is.

The Parental Stress Scale

The Parental Stress Scale (PSS) is one option.

Berry and Jones originally developed the PSS as a brief measure of both the rewarding and stressful aspects of parenting (Berry & Jones, 1995).

A 2024 systematic review of 20 PSS validation studies gave the measure an “A” recommendation using COSMIN criteria, although the authors noted that properties such as responsiveness to change have been studied less extensively than internal consistency and construct validity (Montoro-Pérez et al., 2024).

There is also particularly relevant evidence for families of children with developmental disabilities. Sneed and colleagues evaluated the PSS with 3,220 caregivers of children with intellectual and developmental disabilities; 91% of the children in the sample were autistic. The original 18-item version did not fit their data well, but a 15-item version showed good internal consistency, discriminant validity, and model fit (Sneed et al., 2026).

That does not mean every BCBA should automatically administer a PSS.

It does mean parenting stress is a legitimate outcome to consider when the family's goals include making parenting routines more manageable.

The Parenting Stress Index

The Parenting Stress Index, Fourth Edition (PSI-4) provides a more extensive assessment of stress in the parent-child system.

The full PSI-4 evaluates stress related to child characteristics, parent characteristics, and situational/life stress. The PSI-4 Short Form contains 36 items and produces scores for Parental Distress, Parent-Child Dysfunctional Interaction, Difficult Child, and Total Stress. The short form generally takes about 10 minutes to complete (Abidin, 2012).

Unlike a simple informal rating scale, the PSI-4 is a commercially published standardized assessment with administration, scoring, licensing, and qualification requirements. BCBAs considering it should make sure their training and use comply with the publisher's requirements.

For an interesting ABA-specific example, the PSI-4-SF is also used in TRICARE's Autism Care Demonstration for younger beneficiaries. That does not make it a universal ABA requirement, but it shows that caregiver stress is already being treated as a relevant outcome within at least one major ABA funding system.

Autism Parenting Stress Index

Another option is the Autism Parenting Stress Index (APSI).

The APSI was developed specifically around parenting stress associated with autism-related and co-occurring concerns. Its initial psychometric study included 274 young children and was designed to help identify areas where parents may need support and to examine changes in parenting stress following intervention (Silva & Schalock, 2012).

The important point is not that every ABA program needs one particular stress measure.

It is that caregiver stress is an outcome worth considering when it is relevant to what the family is hoping services will improve.

Fidelity and stress can tell you different things

Imagine this pattern:

Caregiver fidelity: 92%
Child behavior: improving
Parenting stress: increasing

That should prompt a different clinical conversation than:

Caregiver fidelity: 92%
Child behavior: improving
Parenting stress: decreasing

The first pattern may mean the intervention is technically effective but costly for the family to sustain.

Maybe the strategy takes too much time.

Maybe the parent is constantly monitoring the child.

Maybe implementation competes with caring for siblings.

Maybe the procedure works but requires more support than the caregiver currently has.

Fidelity tells you whether the caregiver is implementing the strategy.

Stress may tell you something about what implementing it is costing the caregiver.

Neither measure replaces the other.

Measure confidence and self-efficacy

Another outcome that often gets overlooked is whether the caregiver feels increasingly capable of handling challenges.

A parent may learn one specific protocol perfectly and still feel completely dependent on the BCBA every time something changes.

That is different from developing broader confidence and problem-solving ability.

The Parenting Sense of Competence measures have been used to examine parenting efficacy and satisfaction. Gilmore and Cuskelly recently revised the original scale and developed the 16-item Parenting Sense of Competence-Revised (PSOC-R), with separate efficacy and satisfaction dimensions and improved psychometric properties (Gilmore & Cuskelly, 2024).

A 2024 systematic review of parental self-efficacy among parents of autistic children also identified relationships between self-efficacy and outcomes including parenting stress, mental health, parenting practices, and parent-child relationships (Brennan et al., 2024).

For ABA parent training, one long-term question might be:

Is the caregiver becoming more able to respond to new situations without waiting for the BCBA to tell them exactly what to do?

That may be one of the most important outcomes of caregiver coaching.

Consider caregiver strain, not just stress

Stress and strain are related, but they are not identical.

Parenting stress asks about the experience of parenting.

Caregiver strain can capture the broader effect caregiving has on the person's life.

The Caregiver Strain Questionnaire (CGSQ) includes objective and subjective dimensions of strain. In a study of 304 caregivers of autistic children, the 21-item CGSQ demonstrated good reliability and validity and was related to measures including quality of life, family functioning, coping, social support, and the child's functional impairment and behavior concerns (Khanna et al., 2012).

This may matter when the intervention affects things like:

• work;

• relationships;

• other children;

• family activities;

• worry;

• guilt;

• emotional burden.

A caregiver can be very skilled at implementing an ABA procedure and still experience substantial strain.

Measure family quality of life when the goals are bigger than one routine

Sometimes the outcome of interest is broader still.

The Beach Center Family Quality of Life Scale was developed for families of people with disabilities and includes 25 items across five domains:

• family interaction;

• parenting;

• emotional well-being;

• physical/material well-being;

• disability-related support (Hoffman et al., 2006).

The scale has also been studied with families of autistic children. In one sample of 452 parents of young autistic children, satisfaction scores were negatively related to parenting stress and showed sensitivity to change (Rivard et al., 2017).

More recent research with 406 caregivers found that stronger family-professional partnerships and better contextual fit in parent training were associated with higher family quality of life (2025).

That connects directly to an important clinical question:

Is the plan helping this family live the life they are trying to live?

Social validity should be measured, not assumed

Behavior analysts talk about social validity frequently, but the caregiver-training literature suggests it is not always measured.

O'Neill and Koudys reviewed 34 studies of caregiver training for challenging behavior. Only 19 studies—56%—reported caregiver social-validity information. Among those that did, questionnaires and interviews were the most common approaches (O'Neill & Koudys, 2025).

Social validity also needs to mean more than:

“Did you like the training?”

You may want to know whether:

• the goal is still important to the family;

• the strategy is acceptable;

• the effort required feels reasonable;

• the parent believes it is helping;

• the family wants to continue;

• anything about the plan conflicts with their values;

• the strategy fits the routine where it is supposed to occur.

These questions can be asked formally or informally depending on the purpose.

A few simple repeated questions can provide valuable information:

“Is this still something you want to work on?”

“How manageable was this strategy this week?”

“Did the benefit feel worth the effort?”

“What made it harder to use?”

“Would you want to keep using this if ABA ended tomorrow?”

Those answers are data about Fit.

Individualized family outcomes may tell you more than a standardized scale

Standardized measures are useful when you want to compare a score to norms or track a validated construct such as parenting stress.

But not every meaningful family outcome fits neatly into a standardized instrument.

Goal Attainment Scaling (GAS) is one approach for tracking individualized outcomes. GAS allows people to identify personally meaningful goals and define different levels of progress rather than requiring every person to be measured against exactly the same outcome. It has been used in autism intervention research specifically because priorities can differ considerably across individuals and families (Lee et al., 2022).

Suppose a parent says:

“I just want bedtime to stop taking two hours.”

A simple individualized scale might look like this:

Or the outcome might not be duration at all.

Maybe what matters most to the parent is:

“I want to be able to make dinner without stopping every two minutes.”

That is a perfectly reasonable family outcome to measure if that is what the intervention is intended to improve.

Formal measures and everyday clinical data serve different purposes

Not every measure belongs in every session.

A standardized stress, strain, or quality-of-life measure may be useful at baseline and periodically throughout services.

Routine-specific data may be useful every week.

A caregiver fidelity checklist may be used while teaching a skill.

A self-efficacy measure may be useful when the broader goal is increasing caregiver confidence.

These data answer different questions.

Standardized tools should be selected based on their intended population, psychometric evidence, licensing requirements, qualifications for use, and whether the construct they measure actually relates to the family's goals.

More measures do not automatically make the assessment better.

Don't measure everything

A post like this can easily create the impression that every BCBA needs six questionnaires, three fidelity checklists, child data, parent data, and a family-quality-of-life measure.

That would create exactly the kind of burden parent training should avoid.

Start with the clinical question.

If the family wants help getting out of the house in the morning, you might need:

• one measurable caregiver skill;

• one relevant child outcome;

• a simple measure of whether mornings are becoming more manageable.

If parenting stress is a major concern, add an appropriate stress measure.

If the parent knows every procedure but still feels unable to handle anything new without the BCBA, consider measuring self-efficacy or independent problem solving.

If a technically effective intervention keeps getting abandoned, measure fit.

Collect the data that will change a clinical decision.

Look for patterns across outcomes

Some of the most useful information comes from seeing outcomes together.

Pattern 1

Caregiver skill: improving
Child outcome: improving
Stress: decreasing
Fit: high

The plan may be worth continuing and generalizing.

Pattern 2

Caregiver skill: high
Child outcome: improving
Stress: increasing
Fit: low

The intervention may work technically but need to be redesigned to reduce burden.

Pattern 3

Caregiver skill: low
Understanding: high
Fit: high

The caregiver may need more rehearsal, modeling, or feedback.

Pattern 4

Caregiver skill: high during sessions
Use at home: low
Stress: high
Fit: low

More training may not be the answer at all.

The plan may need to change.

That is why caregiver data should not be reduced to a single percentage.

Don't assume parent training automatically improves parent outcomes

One reason to measure parent outcomes is that the research does not support assuming they will improve simply because an intervention includes parents.

For example, Tarver and colleagues found a favorable overall effect of behavioral parent interventions on parenting stress in their 2019 meta-analysis. But MacKenzie and Eack's later meta-analysis found small benefits for parenting confidence and mental health while finding no significant improvement in stress, caregiving burden, family adjustment, or physical health (Tarver et al., 2019; MacKenzie & Eack, 2022).

Those findings reinforce an important point:

A parent outcome is not automatically improved just because the parent participated in treatment.

If reducing stress, increasing confidence, or making family life easier is important, measure it.

Connect the data to the next session

The purpose of parent-training data is not to fill another graph.

It should help decide what happens next.

If the caregiver cannot identify when to use the strategy, teach discrimination.

If they understand it but cannot perform it, practice.

If they perform it with the BCBA but not at home, find out why.

If the strategy is being implemented accurately but stress is increasing, look at burden and fit.

If caregiver skill is improving but the child outcome is not, reassess the intervention.

If the family's original routine is becoming easier, recognize that progress even if it is not captured by a traditional acquisition target.

The data should help the BCBA and caregiver decide together whether to continue, modify, generalize, or move on.

That is also what keeps parent training from becoming reactive. For a deeper look at maintaining continuity while still responding to what happened this week, see Why ABA Parent Training Becomes Reactive—and How to Give It More Structure.

Use the Session Companion to keep the information connected

The free ACHIEVE Session Companion can help organize the information you need from one parent-training session to the next.

You can identify the caregiver goal before the session, capture what happened between sessions, record caregiver performance and barriers, and decide what you need to follow up on next.

That same information can then support your documentation.

See What to Include in an ABA Parent Training Note for examples of turning parent-training data into a SOAP-based narrative note.

Download the Free ACHIEVE Session Companion

Parent-training data should tell you more than whether the caregiver followed the procedure correctly.

It should help you answer whether the caregiver is learning, whether the child is benefiting, whether the plan works in real life, and whether the family is actually moving toward the outcomes that matter to them.

References

Abidin, R. R. (2012). Parenting Stress Index (4th ed.). Psychological Assessment Resources.

Berry, J. O., & Jones, W. H. (1995). The Parental Stress Scale: Initial psychometric evidence. Journal of Social and Personal Relationships, 12(3), 463–472. https://doi.org/10.1177/0265407595123009

Brennan, J., Ward, O. F., Tomeny, T. S., & Davis, T. E., III. (2024). A systematic review of parental self-efficacy in parents of autistic children. Clinical Child and Family Psychology Review, 27(3), 878–905. https://doi.org/10.1007/s10567-024-00495-2

Gilmore, L., & Cuskelly, M. (2024). The Parenting Sense of Competence scale: Updating a classic. Child: Care, Health and Development, 50(1), e13173. https://doi.org/10.1111/cch.13173

Hoffman, L., Marquis, J., Poston, D., Summers, J. A., & Turnbull, A. (2006). Assessing family outcomes: Psychometric evaluation of the Beach Center Family Quality of Life Scale. Journal of Marriage and Family, 68(4), 1069–1083. https://doi.org/10.1111/j.1741-3737.2006.00314.x

Iadarola, S., Levato, L., Harrison, B., Smith, T., Lecavalier, L., Johnson, C., Swiezy, N., Bearss, K., & Scahill, L. (2018). Teaching parents behavioral strategies for autism spectrum disorder: Effects on stress, strain, and competence. Journal of Autism and Developmental Disorders, 48, 1031–1040. https://doi.org/10.1007/s10803-017-3339-2

Khanna, R., Madhavan, S. S., Smith, M. J., Tworek, C., Patrick, J. H., & Becker-Cottrill, B. (2012). Psychometric properties of the Caregiver Strain Questionnaire among caregivers of children with autism. Autism, 16(2), 179–199. https://doi.org/10.1177/1362361311406143

Lee, C. E., Shogren, K. A., Segal, J., Pezzimenti, F., Aleman-Tovar, J., Lounds Taylor, J., & others. (2022). Goal Attainment Scaling–Community-Based: A new method to incorporate personalized outcomes into intervention research with youth and adults on the autism spectrum. Autism, 26(1), 178–187. https://doi.org/10.1177/13623613211024492

MacKenzie, K. T., & Eack, S. M. (2022). Interventions to improve outcomes for parents of children with autism spectrum disorder: A meta-analysis. Journal of Autism and Developmental Disorders, 52, 2859–2883. https://doi.org/10.1007/s10803-021-05164-9

Montoro-Pérez, N., Alós-Maldonado, R., Montejano-Lozoya, R., Richart-Martínez, M., & Martínez-Alzamora, N. (2024). A systematic review and meta-analysis of the Parental Stress Scale's psychometric properties. Stress and Health, 40(6), e3481. https://doi.org/10.1002/smi.3481

O'Neill, P., & Koudys, J. (2025). Scoping review: Caregiver training to reduce challenging behaviors displayed by children on the autism spectrum. Behavior Analysis in Practice, 18(1), 56–73. https://doi.org/10.1007/s40617-024-00960-y

Silva, L. M. T., & Schalock, M. (2012). Autism Parenting Stress Index: Initial psychometric evidence. Journal of Autism and Developmental Disorders, 42(4), 566–574. https://doi.org/10.1007/s10803-011-1274-1

Sneed, L., Pfingston, B., Cook, I., Taylor, R., Samelson, D., & Fitchett, B. (2026). Psychometric properties of the Parental Stress Scale for parents of children with intellectual and developmental disabilities. Journal of Autism and Developmental Disorders, 56(5), 1836–1848. https://doi.org/10.1007/s10803-024-06700-z

Tarver, J., Palmer, M., Webb, S., Scott, S., Slonims, V., Simonoff, E., & Charman, T. (2019). Child and parent outcomes following parent interventions for child emotional and behavioral problems in autism spectrum disorders: A systematic review and meta-analysis. Autism, 23(7), 1630–1644. https://doi.org/10.1177/1362361319830042