Paired Choice Preference Assessment: How to Run and Score It

A reinforcer is anything that follows a behavior and makes that behavior more likely to happen again. A Registered Behavior Technician (RBT) might reach for the same toy every time because it worked once, while a reinforcer the learner would choose first sits untouched.

A paired choice preference assessment gives Board Certified Behavior Analysts (BCBAs) a structured way to check which items a learner prefers using a series of two-item choices.

Below, we walk through how to set up the trials and score each item. We also cover the setup errors that skew a ranking and when a faster format fits better.

What is a paired choice preference assessment?

A paired choice preference assessment is a method for finding out which items or activities a learner prefers by presenting two options at a time and recording which one they choose. It's also known as a forced-choice preference assessment or a paired-stimulus preference assessment.

Because the learner can take only one item per trial, every choice shows which of two items they rank higher. Those near the top of the ranking tend to be stronger reinforcers than those near the bottom, a pattern a 2026 study found when the reinforcers were different kinds of social interaction.

Once you have that preference hierarchy (the reinforcers ranked from most to least chosen), you can pick reinforcers for a treatment plan, a token economy, or contingent reinforcement during teaching.

How to conduct a paired choice preference assessment in ABA

To conduct a paired choice preference assessment in Applied Behavior Analysis (ABA), present a learner with two items or options at a time, let them choose one, and repeat until every choice has been paired with every other choice at least once.

This round-robin format was first described in a 1992 study. Previous research had found that some learners were drawn to nearly every reinforcement offered, making stronger reinforcers difficult to distinguish.

Follow these steps to run the assessment:

1. Pick a small set of items: Choose toys, snacks and drinks, or activities you suspect the learner enjoys. A bigger set means a longer assessment.
2. Let the learner sample each item first: Give them brief access to every item before you start scoring, so the choice isn't driven by novelty alone.
3. Present two items at a time, spaced evenly apart: Ask, "Which one do you want?" or use a gesture prompt if the learner is non-vocal.
4. Record the selection, then remove both items: If the learner reaches for both, block the attempt and re-present the pair. If they don't respond within a few seconds, mark it "no response" and move to the next pairing.
5. Repeat until every item has been paired with every other item once: With six items, that's 15 total trials (6 × 5 ÷ 2). With eight items, it's 28.
6. Rotate item position left and right: This keeps a side bias from skewing the data.
7. Spread the assessment across sessions if needed: A learner who's tired or agitated halfway through gives you unreliable data, so finish the remaining pairs during the next visit.

Keep a data sheet on hand for every trial. Here's how the first five trials of a four-item toy set could look:

List every pairing your item set requires before the session starts, so the RBT running it has a complete sheet ready to go and doesn't have to track from memory.

How to score a paired choice preference assessment

To score a paired choice preference assessment, divide the number of times each item was selected by the number of trials it appeared in, then multiply by 100.

Because every item shows up in the same number of trials (one fewer than the total item count), this gives you a clean percentage you can rank from highest to lowest.

For example, if you assessed six items, each one was presented five times. An item selected in four of those five trials scores 80%. Line up all six percentages, and you have your preference hierarchy, with the items at the top as your strongest reinforcer candidates.

A low score means the item lost more often when something else was on offer, but it may still work as a reinforcer when it's offered alone.

When to use a paired choice assessment over other formats

Use a paired choice preference assessment when you need a precise, differentiated ranking and the learner can reliably attend to two items at once.

This format also helps when a token system stops earning engagement, since it shows which items still work as rewards and which no longer do.

That precision takes time. A multiple stimulus without replacement (MSWO) format, where the learner picks from the whole array, and each chosen item is removed, took less than half the time to run in an early comparison and produced similar rankings.

If your team needs a quick reinforcer check between sessions, an MSWO or a free-operant observation may be a better fit. In a free-operant observation, the learner moves freely between several items for a set period, and you record how long they engage with each one.

 Save the full paired choice procedure for initial treatment planning or a program that needs a full reset, where a precise ranking is worth the extra trials.

Common mistakes that affect paired choice results

Small setup errors can skew paired choice preference assessment results. Watch for these five:

1. Letting a learner briefly access the item they didn't choose: If both items end up in the learner's hands, the choice stops deciding anything. Remove the unselected item from view completely before presenting the next pair.
2. Skipping the sample phase: An item the learner has never touched can land at the bottom of the ranking because it's unfamiliar. If a new item scores low, note that it was new before you rule it out.
3. Running every trial in one sitting: Late trials can reflect tiredness more than preference. Watch for slower responses or a run of "no response" trials near the end, and stop there.
4. Leaving a side bias unchecked: Check your data sheet after each session. If most selections came from the same side, the ranking reflects position more than preference, so re-run those pairs with the sides swapped.
5. Pairing items that aren't easy to compare: A 30-second snack next to a 10-minute tablet turn makes it hard to give every item the same amount of access time. Run edibles in a separate assessment from toys and activities, so each comparison stays fair.

When every trial runs the same way, any change in the ranking points back to the learner.

Put paired choice preference assessment results to work with Passage Health

Once a paired choice preference assessment gives you a ranking, the next step is building those reinforcers into daily sessions and tracking how the learner responds.

At Passage Health, we keep that clinical data in one platform with your schedule and claims, so your BCBAs and RBTs work from the same record.

Here's how our platform supports the work that follows the assessment:

• Mobile data collection: RBTs record session data on the targets your BCBAs set up, and the session note fills in with that data in real time. Duration and rate targets run on start-and-stop timers, which you can set to start automatically in the phases you choose.
• Graphs and treatment reports: Graphs build automatically from the first data point, and you can add event lines to mark changes like a new set of reinforcers. Customizable treatment report templates pull those graphs into progress reports you can export as PDFs.
• Scheduling on web and mobile: Color-coded daily and weekly calendar views show who's booked where, and you can reassign or rebook a session from its edit screen. Team members with the right permissions can also create and edit events in the mobile app.
• Electronic billing and reporting: Sessions run in Passage Health can be turned into claims and sent through our clearinghouse partner, and claim scrubbing flags issues before submission. Reports cover sessions, signatures, and billing, and the schedule shows authorization utilization.
• Client Intake: Track leads from your intake form through conversion, and send forms through the caregiver portal.
• Roles, signatures, and funding: Custom permissions control what each role can do in scheduling, session notes, and intake. Signature questions record each signer's role with an automatic date and time, and each client profile can carry more than one funding source.

Book a demo to see how Passage Health connects your paired choice preference assessment results to the sessions, graphs, and claims that follow.

Frequently asked questions

What's the difference between forced-choice and paired-stimulus preference assessment?

There's no difference between forced-choice and paired-stimulus preference assessment: both names describe the same procedure. You may see any of these terms in research, reports, or treatment plans.

How many items should a paired choice preference assessment include?

A paired choice preference assessment has no fixed item count, so the right number depends on how many trials the learner can sit through. The pairs grow faster than the items: five items need 10 trials, while 10 items need 45.

How long does a paired choice preference assessment take?

A paired choice preference assessment's length depends on the number of trials and how fast the learner responds. Timing your first run gives you a baseline for planning repeats.

How often should you repeat a paired choice preference assessment?

Repeat a paired choice preference assessment any time a previously strong reinforcer stops holding a learner's attention. It's also worth re-running the assessment before you write a new treatment plan, so the reinforcers in it reflect the learner's current choices.

Can a paired choice preference assessment be used with older learners?

A paired choice preference assessment works with older learners, as long as they can reliably choose between two items or activities presented together. One study used it, alongside an MSWO assessment, to identify preferred job tasks for adolescents with autism.

References

DeLeon, I. G., & Iwata, B. A. (1996). Evaluation of a multiple-stimulus presentation format for assessing reinforcer preferences. Journal of Applied Behavior Analysis, 29(4), 519–533. https://doi.org/10.1901/jaba.1996.29-519 

Fisher, W., Piazza, C. C., Bowman, L. G., et al. (1992). A comparison of two approaches for identifying reinforcers for persons with severe and profound disabilities. Journal of Applied Behavior Analysis, 25(2), 491–498. https://doi.org/10.1901/jaba.1992.25-491 

Idaho Training Clearinghouse. (2025, January). Preference assessment quick guide. https://idahotc.com/Portals/0/ResourceFiles/1270/Preference_Assessment_Quick_Guide.pdf

Newman, Z. A. D., Roscoe, E. M., & Schaefer, A. C. (2026). Generality of a paired-stimulus preference assessment for identifying reinforcing forms of social interaction. Behavioral Sciences, 16(4), 543. https://doi.org/10.3390/bs16040543

St. Peter, C., Shuler, N. J., Toegel, C., et al. (2022). Using preference assessments to identify preferred job tasks for adolescents with autism. Education and Treatment of Children, 45(1), 17–32. https://doi.org/10.1007/s43494-021-00061-3