A partial-interval recording percentage equals valid intervals with any occurrence divided by all valid intervals observed, multiplied by 100. Report valid intervals divided by planned intervals as a separate coverage measure. A missed or invalid interval needs its own state. The percentage samples whether behavior occurred at least once in each interval and gives no exact count or duration.

Define a positive interval

Specify the response boundary, interval clock, any-occurrence rule, prompts, access, and visibility.

Build the valid denominator

Include only intervals with adequate observation under the predeclared rule.

Calculate coverage separately

Show valid divided by planned intervals before interpreting occurrence percentage.

Keep sampled meaning clear

Use continuous count or duration when exact frequency or time occupied drives the decision.

Decide whether partial-interval recording fits the question

A partial-interval recording percentage is most useful when the team needs a feasible sample of whether a response appeared during repeated intervals. It can help show how broadly responding was distributed across an observation. It cannot tell the team how many times the response occurred or how long it lasted. A two-second response and a response that continues for most of the interval both produce one positive interval.

Write the decision question before selecting the procedure. If the question concerns exact rate, total duration, latency, or time between responses, choose the direct measure that matches it when feasible. If sampling is necessary, document why the chosen interval length and observation schedule can represent the settings, people, activities, and times relevant to the decision.

Prepare a scorable observation sheet

Before observation, define the response with examples and nonexamples. State the interval length, observation start and stop rules, planned number of intervals, pause rules, and what makes an interval invalid. Use separate marks for positive, negative, missed, invalid, and outside-exposure intervals. A blank cell should never carry an unstated meaning.

Test the form with a short practice sample. Two observers should be able to start the clock at the same point, apply the any-occurrence rule consistently, and distinguish a valid negative interval from one they could not observe. Record changes to the response definition or interval procedure as a new version.

Calculate the percentage in a reproducible order

First count planned intervals. Next remove only the intervals that meet the predeclared missed or invalid rule. The remaining intervals form the valid denominator. Count the valid intervals with at least one occurrence, divide that number by valid intervals, and multiply by 100. Calculate observation coverage separately by dividing valid intervals by planned intervals.

Keep the counts beside both percentages. Reporting “50%” alone conceals whether the result came from 9 of 18 intervals or 1 of 2. Apply the stated rounding rule only after the division. If two people calculate the worksheet independently, they should reach the same numerator, denominator, coverage, and result.

Build the calculation around a clear decision

The partial-interval worksheet for Felix identifies the response definition, unit, interval or event boundary, planned exposure, valid denominator, missing-state rule, calculation, responsible qualified role, and review date before data collection.

Worked example: 20 planned intervals

Felix has 20 planned intervals. Two are missed, leaving 18 valid intervals; nine contain at least one defined occurrence. Partial-interval percentage is 9 divided by 18, or 50%. Coverage is 18 divided by 20, or 90%. The two missed intervals stay outside both positive and negative scores.

Report coverage with the result

For Felix, report planned, observed, valid, positive, negative, missed, invalid, and outside-exposure units as applicable. Preserve raw records, timestamps, interval length, procedure version, and rounding. A sampled percentage retains the limits of the selected procedure.

Keep access available during measurement

During the transparent partial-interval calculations review for Felix, 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 whether the outcome matters.

Use evidence within its stated scope

For Felix, the BACB ethics hub and CASP public summary provide professional context. The BCBA Test Content Outline covers continuous and discontinuous procedures, temporal dimensions, product measures, validity, reliability, and representative measurement as examination content.

For the transparent partial-interval calculations question, Pritchard and colleagues and Wirth and colleagues examine accuracy and decision implications of discontinuous measurement procedures. Lobo and colleagues supply single-case measurement and analysis context. ASHA says AAC users should always have access to their communication tools or devices. These sources support careful method selection without creating a universal interval length, correction factor, or clinical threshold.

Document the clinical review

Review the partial-interval worksheet with Felix and the responsible clinician. Record the procedure, raw data, coverage, calculation, limitations, selected action, owner, next sample, and review date. Reopen the method when the response pattern, access, setting, interval, exposure, system, or decision changes.

Clinician review checklist

  • Does the response definition support reliable scoring during every sampled activity?
  • Are planned, valid, positive, negative, missed, and invalid intervals all visible?
  • Is the positive percentage calculated only from valid intervals?
  • Is coverage reported with its own numerator and denominator?
  • Does the report explain that the result samples occurrence across intervals?
  • Are interval length, setting, people, access conditions, and procedure version recorded?
  • Can the client review the outcome and its practical meaning in an accessible form?

Related resources

Sources