Absolute versus relative ABA change answers two different questions. Absolute change subtracts baseline from current in the original unit or percentage points. Relative change divides that difference by baseline. Report both with the underlying numerator, denominator, and therapeutic direction. Relative change becomes unstable near zero, while absolute change can conceal how large the difference is compared with baseline.

Define the two questions

Absolute change asks how many original units separate baseline and current values. Relative change asks how large that difference is compared with baseline. For a percentage measure, absolute change is expressed in percentage points. Relative change is expressed as a percentage of the baseline value. State both questions, the therapeutic direction, and which magnitude is most useful before calculating.

Zane's worksheet compares correct responses across ten opportunities in baseline and ten opportunities in the current sample. The target definition, opportunity rule, setting, phase, and observation method are compatible. Each phase keeps its numerator and denominator visible. That source detail is essential because percentages can look comparable while their opportunity definitions differ.

Verify underlying counts and exposures

Reconcile expected, observed, missing, excluded, duplicate, and corrected records in both phases. Confirm that an opportunity means the same thing, that ordinary communication and access supports were available, and that the observation windows belong to the declared phases. Do not turn a missing response into an incorrect response unless the measurement definition explicitly supports that state.

When denominators differ, report the raw counts and opportunities and consider whether the proportions remain compatible. Two of 10 and 20 of 100 are both 20%, yet the second estimate comes from more opportunities. Unequal exposure, clustered observations, or a changed opportunity type can affect interpretation even when the percentage arithmetic is correct.

Calculate Zane's absolute change

Baseline performance is 2 of 10 opportunities:

2 / 10 × 100 = 20%

Current performance is 4 of 10 opportunities:

4 / 10 × 100 = 40%

Absolute percentage-point change is current percentage minus baseline percentage:

40% minus 20% = plus 20 percentage points

The raw numerator changed from two to four correct responses, an increase of two responses within equal ten-opportunity samples. Calling the change "20%" at this step would be ambiguous. The original-unit result is 20 percentage points.

Calculate the relative change

Use the baseline percentage as the denominator:

Relative change = ((40 minus 20) / 20) × 100 = 100%

Zane's current percentage is 100% higher than the baseline percentage. That does not mean performance is 100%; the current value is 40%. It also does not mean all opportunities improved. The ratio compares a 20-point difference with a 20% baseline.

Keep the signed result and the intended direction. For a goal where higher correct responding is desired, the report can describe a 100% relative increase. For a reduction target, the same positive sign could indicate movement away from the goal.

Handle zero, small baselines, and ceilings

Relative change is undefined when baseline equals zero. Report current value and absolute change without inventing an infinite percentage. A small baseline can create a large relative change from a modest original-unit difference. For example, an increase from 1% to 2% is plus 1 percentage point and plus 100% relative to baseline.

Ceiling and floor constraints also matter. Moving from 80% to 100% is plus 20 points but only a 25% relative increase because the baseline is larger. The same 20-point absolute change can therefore produce very different relative percentages. Report both and explain which scale answers the clinical question.

Put both magnitudes beside the graph

Plot the raw, time-ordered observation data with phase lines, opportunity denominators, context changes, and ordinary supports. A companion table can show baseline numerator and denominator, current numerator and denominator, phase percentages, percentage-point difference, and relative change. Avoid using a large "100% improvement" label that obscures the current value of 40% and the two-response raw difference.

Review level, trend, variability, overlap, immediacy, and consistency through an appropriate visual-analysis process. Lobo and colleagues and Wolfe, Barton, and Meadan provide relevant context. NIST's moving-average guidance concerns smoothing and should be labeled if added to the display.

Use clear report language

Zane's observed correct responding changed from 2 of 10 opportunities, or 20%, in baseline to 4 of 10, or 40%, in the current sample. The absolute change was plus 2 responses and plus 20 percentage points. Relative to the 20% baseline, the change was plus 100%.

Add phase dates, missingness, rounding, opportunity definition, setting, treatment-integrity evidence, and client feedback. Use increase or decrease only after stating the target direction. Avoid the standalone phrase "doubled" when readers could mistake it for current mastery, broad improvement, or proof of effect.

Review clinical meaning and access

Discuss the result with Zane through accessible communication and with the responsible clinician. Maintain AAC, interpreters, mobility supports, food, water, bathroom access, prescribed care, rest, relationships, and emergency help during measurement and review. ASHA supports continuous AAC access.

Ask whether the target, opportunity, and change matter to Zane, whether burden or context shifted, and whether the pattern appears in relevant everyday settings. The qualified clinician should integrate the calculation with the graph, measurement quality, implementation evidence, risks, benefits, and the person's perspective.

Keep the source set in scope

The BACB ethics hub and CASP public summary provide professional and clinical context. The BCBA Test Content Outline includes measurement and data-based decisions as examination content. NIST's distribution page, outlier guidance, and box-plot page support broader descriptive review. These sources do not establish a universal change threshold.

Clinician review checklist

For an absolute versus relative ABA change review, confirm the question, target direction, operational definition, compatible phases, raw numerators, denominators, opportunity rules, missingness, phase percentages, absolute original-unit difference, percentage-point difference, nonzero baseline, relative-change arithmetic, rounding, graph, report wording, access, client perspective, implementation evidence, decision owner, and review date. Preserve source values and recalculate when either phase changes.

Limits of absolute and relative change

Neither measure establishes measurement validity, treatment integrity, clinical importance, benefit, generalization, future performance, statistical uncertainty, or cause. Absolute change can hide baseline scale, while relative change becomes undefined at zero and unstable near zero. Equal percentages can arise from different numerators and denominators. Report both measures with raw data and visual analysis, and avoid using either as a stand-alone rule for treatment decisions.

Related resources

Sources