An ABC sequential analysis time window specifies how far before or after an anchor event another event may occur and still enter the sequence calculation. Define the anchor, direction, duration, timestamp precision, overlap rule, and treatment of truncated windows before analysis. A sensitivity check can show how results change across defensible windows, while the primary window remains the predeclared decision rule.

Anchor the sequence

Name the exact onset or offset that starts the clock. Use the same anchor across records and record uncertainty when timestamps lack the required precision.

Choose direction and duration

Antecedent windows look backward or start from an exposure; consequence windows look forward from the response. State the number of seconds or intervals.

Predefine overlapping-window rules

When anchor events occur close together, decide whether windows can overlap, reset, or receive separate sequence IDs. Retain the underlying timestamps.

Use sensitivity analysis carefully

Recalculate with another defensible window to assess fragility. Keep the primary result identifiable and avoid selecting the most dramatic window after reviewing outcomes.

Choose a window the timestamps can support

An ABC sequential analysis time window should match the review question, plausible event timing, and precision of the source record. Second-level timestamps can support a time-based rule. Coarse interval marks may support only interval-based sequencing. A ten-second window calculated from timestamps rounded to the nearest minute creates false precision.

Define whether the window begins at response onset, response offset, event onset, or event offset. State whether an event exactly on the boundary is included. Apply the same rule to every record in the primary analysis and preserve the original timestamps for audit.

Handle overlaps and observation boundaries

Closely spaced anchor events can create overlapping windows. Decide whether one event can be mapped to more than one anchor, whether the clock resets, or whether each anchor retains its own sequence ID. Show the choice in the codebook and test it with examples before calculation.

An anchor near the start or end of observation may lack a complete look-back or look-forward period. Mark that sequence as truncated under the predeclared rule. Excluding it from the complete-window denominator is different from scoring no event.

Build a transparent sensitivity table

Show the primary window first with its numerator, denominator, probability, and truncated count. Add one or more defensible comparison windows only when they answer a stated robustness question. Keep the event definitions and eligible anchor cohort constant across rows.

For Talia, the 10-second row is 3 of 8, or 0.375, and the 20-second row is 5 of 8, or 0.625. The difference shows that two additional attention onsets fall between 10 and 20 seconds. It does not identify which window proves a behavioral function.

Frame one clear sequence question

The sequential-window sensitivity table for Talia names the observable events, sequence direction, time or interval window, planned exposure, valid denominator, missing-state rule, raw cells, calculation owner, and review date before percentages are interpreted.

Worked example: compare 10 and 20 seconds

Talia has 8 valid response events. Peer attention begins within 10 seconds after 3, giving 3 of 8, or 0.375. Within 20 seconds it begins after 5, giving 5 of 8, or 0.625. The 20-second window contains more unrelated activity, so both labeled results and the prespecified primary window remain visible.

Audit anchors and complete windows

Talia's timestamps use the same response anchor and attention-onset definition. A response in the last seconds of observation is marked truncated if the complete consequence window is unavailable.

Interpret window sensitivity

Window choice changes which events qualify. Short windows can miss delayed sequences, while long windows admit more coincidental events. Clinical relevance, event timing, observation precision, and existing evidence should guide the rule.

Protect access during observation

During Talia's predeclared ABC sequence windows review, keep AAC, interpreters, mobility, food, water, bathroom use, prescribed care, health support, rest, relationships, and emergency help available. Record access or health changes and use accessible communication to ask about relevant events, comfort, assent, dissent, and priorities.

Use evidence within its stated scope

The BACB ethics hub and CASP public summary provide professional context. The BCBA Test Content Outline includes direct, indirect, and product measures; validity and reliability; representative measurement; and indirect, descriptive, and experimental functional assessment as examination content.

For Talia's predeclared ABC sequence windows question, primary descriptive-assessment research compares conditional with background probabilities and shows why event direction, sequence windows, cell counts, and context matter. Antecedent versus consequent predictors, contingency space analysis, and comparisons of descriptive methods document limited correspondence between descriptive and experimental findings. Research on reducing functional-assessment ambiguity distinguishes observed association from experimental demonstration. ASHA says AAC users should always have access to their tools or devices.

Close the descriptive-data review

Review the sequential-window sensitivity table with Talia and the responsible clinician. Record raw observations, definitions, windows, coverage, calculations, limitations, the selected next assessment step, its owner, and a review date. Reopen the method when the response, context, access, health, observer, software, or decision changes.

Clinician review checklist

  • Is the anchor event and onset or offset rule explicit?
  • Can the timestamp precision support the chosen duration?
  • Are boundary, overlap, reset, and truncation rules written in advance?
  • Does each sensitivity row use the same eligible anchor cohort?
  • Are numerator, denominator, probability, and incomplete windows visible?
  • Is the primary window identifiable after alternate windows are added?
  • Does the interpretation stay descriptive and preserve client access and safety?

Related resources

Sources