An ABA data range is the maximum observed value minus the minimum within a clearly defined, compatible sample. Report the minimum, maximum, range, unit, ordered values, dates, and condition. Range gives the full observed span. One extreme value or a larger sample can change it sharply, so pair the result with the raw data and a location summary.

Define the question and compatible sample

Range answers a narrow question: how far apart were the lowest and highest compatible observations in this sample? Before calculating, state the target, operational definition, measurement dimension, unit, phase, setting, observation window, and eligible dates. Keep counts, rates, percentages, durations, and latencies in separate samples. A five-minute observation and a full-session total also need a common exposure basis before their values can be compared.

Vera's worksheet uses one duration per eligible observation in the same phase, recorded in minutes. The analysis unit is the observation, and the denominator is seven observed durations. The worksheet keeps the source record ID and date for every value. Any scheduled observation without a valid duration would appear as missing or unavailable, never as zero.

Check the source data before calculating

Trace the measurement definition and each value to the original record. Confirm that the unit has not changed, the session belongs to the declared phase, the duration is plausible under the observation rule, and corrections are resolved. Look for duplicate imports, transposed digits, default zeros, mixed time units, truncated sessions, and values copied from a graph instead of the underlying record.

Sort a working copy while preserving the original order and source identifiers. Sorting should not detach a value from its date or context. If a value changes after correction, retain the earlier value, correction reason, author, time, and recalculated result according to the record process.

Calculate Vera's range step by step

The seven ordered durations are 3, 5, 6, 8, 9, 11, and 18 minutes.

  1. Minimum = 3 minutes.
  2. Maximum = 18 minutes.
  3. Range = maximum minus minimum.
  4. Range = 18 minus 3 = 15 minutes.

The answer is expressed in minutes, the same unit as the observations. Report all three pieces: minimum 3, maximum 18, and range 15. Writing only "range = 15" hides where the span begins and ends. No rounding is needed here. If endpoints had decimals, keep full precision through subtraction and apply the declared reporting rule once at the end.

Investigate endpoints, outliers, and missingness

Range depends entirely on two observations. Verify the 3-minute and 18-minute records before interpretation. The final 18-minute value could reflect real variation, a context change, an interrupted routine, a different observer, a recording error, or another meaningful event. NIST's outlier guidance supports investigation before deletion. Remove or correct a value only through a stated, source-based rule.

Missing observations can narrow the observed span. State how many records were expected, observed, excluded, and missing, with reasons. A one-value sample has minimum equal to maximum and a computed range of zero, yet it supplies no evidence about variability across observations. Repeated identical values also yield zero range and still require a source-data and measurement check.

Put the range on a graph

Keep the raw time-series graph primary. Show each observed duration in date order, phase lines, relevant condition changes, and ordinary supports. A table can add sample size, minimum, maximum, and range for the selected phase. Avoid drawing a shaded 3-to-18-minute band that implies expected future limits or a probability interval. The observed range describes this sample only.

Review level, trend, variability, immediacy, overlap, and consistency through an appropriate visual-analysis process. Lobo and colleagues and Wolfe, Barton, and Meadan provide single-case and visual-analysis context. A range can support that review, but it cannot replace the graph.

Use precise report language

Across seven compatible observations in the current phase, Vera's durations ranged from 3 to 18 minutes. The observed range was 15 minutes. The 18-minute endpoint was verified against its source record and remains visible for contextual review.

Add the dates, setting, measurement definition, missing-state result, and any material context. Pair the range with a location summary such as the median when useful. A larger sample has more opportunities to include extreme values, so compare ranges across phases only with sample size, exposure, definitions, and observation conditions visible.

Keep clinical meaning and access in view

The summary should be reviewed with Vera through accessible communication and with the responsible clinician. Keep AAC, interpreters, mobility supports, food, water, bathroom access, prescribed care, rest, relationships, and emergency help available during measurement and review. ASHA's AAC guidance supports continuous access to AAC tools or devices.

Ask whether the duration represents a meaningful outcome, whether the observation burden or context changed, and whether a high or low endpoint matters to safety, comfort, participation, or daily life. Clinical action belongs with the qualified clinician and should incorporate the graph, implementation evidence, client feedback, and other relevant sources.

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, display, and data-based decisions as examination content rather than a case protocol. NIST distinguishes distribution location, spread, and shape, and its box-plot page supplies related spread concepts. NIST's moving-average page describes smoothing, which answers a different question from range.

Clinician review checklist

Confirm the clinical question, target definition, unit, phase, eligible cohort, expected and observed records, exclusions, missing states, source lineage, ordered values, endpoints, arithmetic, rounding, graph, context, access, client perspective, implementation evidence, interpretation, decision owner, next evidence, and review date. Recalculate when a source value, definition, unit, phase, sample, or software rule changes.

Limits of range

Range does not describe typical performance, distribution shape, trend, treatment integrity, measurement validity, client benefit, future limits, statistical uncertainty, or cause. Two samples can share the same range while having very different patterns. A single extreme can dominate the result, and missing extremes can shrink it. Use range as one transparent description of a declared sample. Do not use it alone to start, stop, intensify, reduce, authorize, or judge treatment.

Related resources

Sources