Multielement design

Rapid alternation creates repeated contrasts

Conditions may alternate across sessions, activities, time blocks, or other defined opportunities. A three-condition sequence might appear as A, C, B, B, A, C, with order randomized or counterbalanced within blocks. The key is repeated exposure to each condition close enough in time to compare their data paths.

The BACB BCBA Test Content Outline, 6th edition lists multielement among single-case experimental designs and covers functional assessment, dependent and independent variables, procedural integrity, and data interpretation. It is examination content rather than a clinical protocol.

“Multielement design” and “alternating-treatments design” are often used as synonyms. The original Barlow and Hayes paper proposed the alternating-treatments label for a design that had also been called multielement baseline, multiple schedule, and randomization design. Authors sometimes use “multielement” especially for functional-analysis conditions and “alternating treatments” for intervention comparisons. Read the actual arrangement instead of inferring it from the label.

Functional analysis is a common use

In a traditional functional analysis, defined test conditions and a control condition may alternate rapidly. The analyst asks whether the measured response differentiates across programmed antecedent and consequence arrangements. A clinical review of functional-analysis methodology identifies the multielement design as an efficient way to make several comparisons and notes that reversal or pairwise arrangements can help when rapid alternation produces poor discrimination.

The design and the assessment are separate concepts. A multielement design can compare treatment conditions, teaching supports, or other variables. A functional analysis can use multielement, pairwise, reversal, or another justified arrangement. The design name supplies neither the condition procedures nor permission to implement them.

Functional-analysis conditions can involve serious behavior and clinically significant risk. They require individualized assessment, qualified oversight, medical and interdisciplinary input when indicated, explicit stop criteria, trained personnel, a safe setting, and applicable consent and review. A glossary example cannot serve as an FA recipe.

Condition cues support discrimination

Rapid alternation works best when the participant can discriminate which condition is active. A distinct material, schedule label, partner statement, room feature, or other accessible cue may signal the current contingency. Explain the cues in a communication mode the person can use.

Cues become part of the experimental condition. If one procedure always uses a tablet and another always uses paper, the result concerns each procedure-and-material package. Different staff or rooms can produce the same interpretive problem.

The 2025 functional-analysis decision guide discusses condition-correlated stimuli, fixed and randomized sequences, and the tradeoff between discrimination and sequence effects. A fixed order can manage some establishing-operation concerns while creating predictable order. Random order can spread order effects while increasing other variability.

Carryover and alternation can blur the paths

Carryover occurs when one condition changes responding in the next. Learning, fatigue, medication, satiation, reinforcement, distress, or an emotional response may persist. An alternation effect occurs when repeated switching changes behavior. Sequential confounding occurs when condition position or the preceding condition systematically influences the result.

Choose outcomes expected to respond quickly and recover before the next contrast. Consider spacing, washout, condition order, separate stimulus sets, or another design. A washout period needs a defined rationale and safe acceptance criteria. It cannot erase durable learning.

For skill acquisition, presenting two teaching methods to the same targets can make attribution impossible after learning occurs. An adapted alternating-treatments design assigns matched target sets to separate conditions. The assessment-based instruction tutorial explains this distinction and emphasizes balanced session exposure. Target-set difficulty then becomes a threat that requires matching, assignment, and cautious interpretation.

Build each comparison before collecting data

For every condition, specify:

• the independent-variable steps and accessible condition cue
• the dependent variable, opportunity, session length, and scoring window
• condition order, blocking, exposure, and stopping rules
• ordinary supports and contextual features held constant
• procedural-fidelity and observer-agreement procedures
• carryover, sequence, and discrimination checks
• client preference, assent-withdrawal, adverse-effect, and safety rules

Graph every observation by condition in chronological order. Review level, trend, variability, overlap, immediacy, and consistency. Means and percentages can summarize data, while the time sequence reveals drift, order, or emerging carryover.

Measure implementation separately in each condition. Equal session counts do not create equal exposure when session lengths, opportunities, prompts, or doses differ. Keep every scheduled condition in the record and report exclusions with reasons.

A fictional multielement example

Sam is a fictional adult who chooses to compare three acceptable supports for a five-step community-work checklist: paper, digital, and digital with a self-advancing highlight. The conditions occur in six randomized blocks, once per block. Materials, tasks, partner, session length, AAC access, and ordinary safety supports stay constant.

With paper, Sam independently completes 19 of 30 steps. With digital, the result is 21 of 30. Digital plus highlight produces 27 of 30. Each condition has six observations, and the assigned procedure is delivered correctly in 18 of 18 observations. Sam also selects digital plus highlight as the easiest format to use.

The separated data paths support the highlighted package for this routine if raw session data show consistent differentiation without meaningful carryover or order effects. The comparison does not isolate the highlight from every other feature of that package or establish superiority for another person. Preference, everyday use, maintenance, and burden remain separate outcomes.

Clinical safeguards limit condition exposure

Preserve AAC, food, water, bathroom access, mobility, prescribed care, pain care, rest, emergency help, and effective safety protections throughout the comparison. Stop when consent is withdrawn, assent is withdrawn when applicable, risk changes, an adverse effect appears, or a condition becomes unacceptable. An efficient design is still unsuitable when rapid alternation creates confusion, distress, or unsafe carryover.

The current BACB Ethics Code applies to BCBA and BCaBA certificants and people who completed an application. It addresses competence, client involvement, informed consent and assent when applicable, assessment, medical needs, risk, data, and evaluation. Experimental control remains subordinate to those responsibilities.