Continuous vs discontinuous measurement

Continuous measurement records every instance of a behavior during the whole observation, while discontinuous measurement samples the behavior in intervals and gives an estimate. Frequency, rate, duration, latency and interresponse time are continuous; partial interval, whole interval and momentary time sampling are discontinuous.

What is the difference between continuous and discontinuous measurement?

Continuous methods capture every occurrence from start to finish, so the data show what actually happened. Discontinuous methods divide the session into intervals and record only whether the behavior met a rule in each one, so the data are an estimate reported as a percentage of intervals. Continuous is more precise; discontinuous is easier to do while teaching.

Discontinuous methods exist because watching every instance is not always possible. A technician running a group activity cannot always count every hand flap, but can check at the end of each interval.

The trade-off is accuracy. Partial interval tends to overestimate, whole interval tends to underestimate, and momentary time sampling can go either way.

Continuous and discontinuous measurement side by side

ContinuousDiscontinuous
What is recordedEvery instanceA sample, interval by interval
MethodsFrequency, rate, duration, latency, IRTPartial interval, whole interval, momentary time sampling
ResultCounts, times, ratesPercentage of intervals scored
AccuracyExact if done correctlyAn estimate with known biases
EffortHigh, full attention neededLower, can be done while teaching
Good fitClear, countable or timeable behaviorHigh-rate or long behaviors, busy settings

How do you tell continuous and discontinuous measurement apart on the exam?

Look for intervals. If the session is split into equal blocks and each block is scored yes or no, the method is discontinuous. If the technician counts or times every instance across the whole session, it is continuous. Results given as a percentage of intervals point to discontinuous, while counts, seconds or responses per minute point to continuous.

  • Every instance captured: continuous.
  • Intervals scored yes or no: discontinuous.
  • Percentage of intervals: discontinuous.
  • Count, rate or seconds: continuous.

Worked scenarios

Scenario 1. A technician clicks a counter each time a learner requests help during a 30-minute session. Answer: continuous (frequency).

Scenario 2. A session is split into 15-second intervals, and the technician marks each one if the learner hums at any point. Answer: discontinuous (partial interval).

Scenario 3. A technician starts a stopwatch when crying begins and stops it when crying ends, for every episode. Answer: continuous (duration).

Scenario 4. Every 5 minutes, a timer vibrates and the technician notes whether the learner is in her seat at that instant. Answer: discontinuous (momentary time sampling).

What are the common traps with continuous and discontinuous measurement?

The common trap is assuming any method that uses a timer is discontinuous. Duration and latency use timers but record every instance, so they are continuous. Another trap is reporting discontinuous data as a count, such as saying the behavior happened twelve times when twelve intervals were scored. Intervals scored and occurrences are not the same thing.

  • Treating timed methods as discontinuous.
  • Reporting intervals scored as a count of behaviors.
  • Forgetting that permanent product recording is a separate method.

When would you see each in a session?

You will use continuous methods for countable targets like requests, correct responses or aggression, and for timed targets like tantrum length. You will use discontinuous methods for high-rate or ongoing behaviors, such as stereotypy or engagement, especially during group activities. The supervisor picks the method; you apply it exactly and keep interval lengths as written.

If a method becomes impractical, for example because the learner needs constant physical support, tell your supervisor instead of quietly switching methods. Switching changes what the data mean.

Practice questions

Question 1 · Data collection and graphing

Leo, a second grader, calls out answers without raising his hand. Circle time is always exactly 20 minutes. Which measure should the technician use to capture how often Leo calls out?

Question 2 · Data collection and graphing

At a day program, a vibrating timer goes off every 2 minutes. Each time, the technician glances at Grace and records whether she is engaged with her craft at that moment. Which method is this?

Question 3 · Data collection and graphing

A technician uses whole interval recording for Darius's independent play. He often plays for most of an interval but stops briefly near the end. How are these data most likely to compare with his actual play time?

Question 4 · Data collection and graphing

A learner's humming is recorded with 10-second intervals. The technician marks an interval if humming occurs at any point during it. Which recording method is this, and how does it tend to bias the data?

Questions people ask

Is permanent product recording continuous?

It is usually taught as its own method, measuring the outcome a behavior leaves behind rather than watching the behavior itself.

Is discontinuous measurement less valid?

It is less precise, but it is valid for many behaviors when the method matches the behavior and the plan.

How ready are you?

Ten questions, about 3 minutes, no account. Get your readiness score and your weakest domains.

Check my readiness