Preference assessment

Preference is specific to person, options, and context

A preference assessment asks a relative question: among stated options and conditions, what does the person choose or engage with? Results may change with time, setting, communication access, recent access, health, sensory conditions, people present, and presentation.

Ask the person directly in an accessible form and let them nominate, add, reject, or remove options. Caregiver or staff input may add context but should not overwrite the person's communication. Record who suggested each option. Offer optional exploration of unfamiliar options before treating nonselection as low preference.

The Association for Behavior Analysis International basic-principles page lists a preference-assessment reading under “Schedules of Reinforcement and Choice”; it is not an administration standard. The current BACB BCBA Test Content Outline, 6th edition includes task F.4, “Design and evaluate preference assessments.” Both are education or examination content, not client protocols or authority to practice.

Common methods answer the question differently

Based on 65 peer-reviewed research articles, the Stimulus Preference Assessment Decision-Making System considers stimulus dimensions, client characteristics, administration time, and agreement between potentially effective context-specific procedures. Its authors say SPADS assumes prerequisite knowledge and does not replace appropriate training or direct supervision. It does not validate one universal method.

A 1996 seven-participant study by DeLeon and Iwata compared paired stimulus (PS), multiple stimulus with replacement (MSW), and MSWO. MSWO and PS produced moderate-to-high rank-order correlations and identified similar numbers of reinforcers; MSWO time was comparable with MSW and less than half the PS time. Subsequent tests found that some stimuli selected in PS and MSWO, but not MSW, reinforced arbitrary responses. The authors called the results preliminary, not universal benchmarks.

Choose the format around access and burden

Before selecting a method, examine:

• how the person communicates a selection, refusal, pause, or change
• visual, hearing, motor, attention, and scanning access
• whether pictures or symbols accurately represent the actual option
• the number, type, familiarity, and safe availability of options
• whether brief removal could create distress or risk
• allergies, swallowing, medication, health, sensory, and dietary needs
• the time and setting needed for a credible sample

In Kang and colleagues, two children completed each format five times. Both showed problem behavior in PS and MSWO and none in free-operant sessions. Across PS and MSWO, 98% of recorded problem behavior occurred when a chosen item was removed (participant range 90%–100%), 2% followed an instruction to choose, and none occurred while the child had the item. Functional analyses implicated tangible access; the study did not evaluate behavior maintained by other functions. This supports individualized selection, not a universal free-operant rule.

Protect choice, communication, and basic access

Explain any planned removal and establish accessible ways to select, reject all options, pause, and stop. Stop and reassess upon withdrawal, refusal, distress, or loss of a reliable response; an immediate safety response does not authorize continuing the assessment. Silence or compliance is not assent. Do not require speech, eye contact, or a specific motor action when a reliable alternative exists.

ASHA's AAC practice portal says AAC users should always have their tools or devices. Keep AAC and ordinary access to water, adequate food, the bathroom, movement and mobility, supportive relationships, prescribed health care, pain care, and safety available. Never withhold, ration, or make these needs contingent on a selection or another response. Food is not the default; first address choice, consent or assent, allergies, swallowing, and dietary, medical, cultural, and sensory factors.

The BACB Ethics Code applies to BCBA and BCaBA certificants and people who completed an application for either credential; BACB has no separate jurisdiction over organizations or corporations. It directs covered behavior analysts to work within competence and defined roles, involve clients and relevant stakeholders, obtain required consent and assent, and fit assessments to client needs, context, and resources while maximizing benefits and minimizing harm. A trained team member may administer assigned steps within role and supervision; an appropriately qualified professional authorized under applicable law, credential, and role selects the method, interprets results, and decides clinical use. Other requirements apply.

Report observable results and their limits

Define responses before collecting data. Report raw counts with percentages: selections out of presentations, engagement seconds out of observed seconds, refusals, no responses, prompts, access failures, and explained exclusions. A no response may reflect preference, unfamiliarity, inaccessible display, insufficient time, discomfort, or lack of understanding; preserve those possibilities.

A preference assessment identifies candidates, not reinforcement. A stimulus or event is a demonstrated reinforcer only when contingent delivery increases or maintains a defined response against an appropriate comparison under stated conditions. Reassess when selections change, an option loses value, new options appear, or context changes. A preferred activity may also be a meaningful end rather than payment for another response.

A fictional assessment example

Aaliyah is a fictional eleven-year-old who uses speech and AAC. She selects four activities for a ten-minute free-operant observation because she dislikes removal. All remain accessible for 600 seconds. The observer codes one primary activity or no engagement each second: music 240 seconds (40%), puzzle 160 (26.7%), gardening 140 (23.3%), quiet rest 60 (10%), and no engagement 0. These mutually exclusive durations total 600 seconds and 100%.

These data rank one sample; they establish neither music as a reinforcer nor quiet rest as lacking value. Aaliyah says the community garden matters most outside the session. The team records her report, repeats observation in a relevant setting, and retains all options. Her goal and context remain distinct from the duration ranking.