Condition label

A label names what was operating

Readers should be able to connect a label to a dated, versioned procedure. “Treatment” says too little. “Token board” may still omit the earning rule, exchange price, backup communication, and partner actions. The graph needs a compact label, while the protocol supplies the detail.

A series label identifies which data path a symbol or line represents, such as correct responses or partner implementation. A condition label identifies what arrangement operated during a segment. Keep the two functions distinct in the legend.

Use terms that distinguish genuinely different conditions. If only one component changes, name that component when practical. If several change together, use a package label and document every component elsewhere.

Keep labels factual and neutral

Avoid labels such as “good phase,” “noncompliant,” “failed treatment,” or “successful intervention.” They insert judgment or outcome into the condition name.

Prefer “Baseline,” “Choice menu,” “Video model,” “Schedule 2,” or another observable arrangement. Outcome belongs in the data and interpretation. Client identity, diagnosis, motivation, or character does not belong in a condition label.

Align the label with the phase-change line

Place a condition label over the data segment it describes. The phase-change line should fall between the last observation under the prior condition and the first observation under the new condition.

The effective date comes from the actual implementation record. A plan-signature date, training date, software entry date, and first implementation can differ. Use the date when the condition began for the graphed data and preserve the other dates separately.

Avoid placing a phase line through a data point. If a procedure changes mid-session, define how the observation is assigned or split before analysis.

Version labels when procedures change

“Visual schedule” can describe several versions. If icons, timing, choice options, prompting, or partner responses change materially, record a new version and decide whether the graph needs a new condition.

Minor corrections can remain under one label when the procedure and decision stay comparable, but the implementation log should still record them. Predeclare what counts as a material change when possible.

Bundled labels need component records

A label such as “FCT + schedule thinning” names a package. It cannot show which component produced change, whether both began together, or how parameters shifted. Use component analysis or another design only when clinically and ethically warranted.

When safety, medication, school schedule, staffing, or another concurrent event changes, annotate it without casually calling it the treatment condition. Attribution requires appropriate evidence.

A fictional communication-program audit

Rowan’s graph contains three condition labels: “Baseline,” “Visual schedule,” and “Visual schedule + modeling.” The reviewer checks each against the protocol version, first implementation date, trained implementers, and graph line.

“Baseline” and “Visual schedule v2” match their records. The third label omits that AAC availability and feedback also began on the same date. Label-to-record accuracy is 2 of 3.

The reviewer changes the third label to “Schedule v2 + modeling/feedback; AAC available” and adds a note linking the component record. The graph becomes accurate, while the bundled change still prevents attributing any effect to modeling alone.

Audit every label against evidence

For each condition, verify:

• exact wording and procedural definition
• version and owner
• first and last implementation dates
• phase-line placement
• participants, setting, and measurement rule
• implementer training and fidelity record
• concurrent changes and deviations
• link to the source protocol

Label accuracy can be reported as labels matching every required field divided by all labels due for review. Keep failed labels in the denominator and correct the graph and linked record together.

Labels support visual analysis

The BACB BCBA Test Content Outline, sixth edition includes graphing, visual analysis, experimental design, and data-based decision making. It is examination content rather than a condition-label style manual.

A precision-teaching synthesis describes charting, decision making, and within- and between-condition analysis. Its standardized-chart context reinforces the need to connect visual segments with actual instructional conditions.

Visual analysis still examines level, trend, variability, immediacy, overlap, consistency, and design logic as applicable. A precise label improves readability but cannot create experimental control.

Make labels accessible

Use readable type, direct placement, adequate spacing, and sufficient contrast. Avoid relying on color alone to distinguish conditions. Provide a text table listing condition, dates, and definition for screen-reader access and detailed review.

Exported images can lose hover text or interactive legends. Ensure the essential label appears in the static graph and the underlying data remain available.