13.2 Discontinuous Measurement Procedures (Partial, Whole, MTS, PLACHECK)

Discontinuous Measurement Procedures (Partial, Whole, MTS, PLACHECK)

Exam Tip: Discontinuous measurement (time sampling) questions on the QASP-S exam are among the most heavily tested and clinically critical items. You must memorize the exact directional measurement errors (artifacts) of each system: Partial-Interval Recording (PIR) systematically OVERESTIMATES duration and UNDERESTIMATES high-rate frequency, making it the conservative choice for behavior reduction; Whole-Interval Recording (WIR) systematically UNDERESTIMATES duration and frequency, making it the conservative choice for skill acquisition / behavior increase; and Momentary Time Sampling (MTS) does not systematically overestimate or underestimate duration across large samples, freeing observers from continuous tracking. For groups, know how to calculate PLACHECK.

While continuous measurement procedures capture every discrete behavioral occurrence, there are numerous clinical environments where continuous tracking is technically impossible, clinically counterproductive, or practically unfeasible. In busy inclusive classrooms, multi-client group homes, community job sites, or during complex parent coaching sessions, a single practitioner cannot stare at a stopwatch while simultaneously delivering instruction, prompting responses, delivering tokens, and managing crisis safety.

To address these real-world constraints, applied behavior analysts utilize discontinuous measurement procedures, universally known as time-sampling methods. In discontinuous measurement, the observation session is partitioned into discrete, equal time intervals, and the practitioner records whether the target behavior occurred according to specific, predetermined interval rules.

---

The Inherent Trade-Off: Feasibility vs. Measurement Artifact

Every discontinuous measurement procedure involves a scientific trade-off between practical clinical feasibility and measurement accuracy. Because discontinuous systems sample time rather than recording the entire behavioral stream, they inherently introduce measurement artifacts—data that show an apparent behavioral pattern that exists only because of the way the data were collected, rather than reflecting the true dimensional reality of the behavior.

Continuous Recording:    [=== Full 100% Observation Stream Captured ===]  ──► High Precision / Zero Artifact / High Effort
Discontinuous Sampling:  [ Int 1 ] [ Int 2 ] [ Int 3 ] [ Int 4 ] [ Int 5 ]  ──► Feasible / Directional Artifacts Introduced

The Impact of Interval Length on Sensitivity

The sensitivity and fidelity of time-sampling data are profoundly dictated by the length of the interval:

• Short Intervals (e.g., 5 to 15 seconds): Provide high measurement sensitivity and closely approximate continuous recording data, but impose heavy observational demands that require near-continuous attention.
• Long Intervals (e.g., 1 to 5 minutes): Substantially increase clinical feasibility, allowing staff to teach or manage groups, but introduce severe, distortive measurement artifacts:
  • In a 5-minute Partial-Interval schedule, a single 2-second vocal outburst in each interval is scored as 100% of intervals, creating the illusion of non-stop disruption!
  • In a 5-minute Whole-Interval schedule, an individual who attends to a task for 4 minutes and 58 seconds, but glances away for 2 seconds, is scored as 0% engagement, completely masking near-perfect performance!
• Clinical Best Practice Guideline: In direct 1:1 clinical sessions, intervals should rarely exceed 10 to 30 seconds. In classroom or group consultation settings where staff must multitask, Momentary Time Sampling with 1- to 2-minute intervals is preferred over excessively long PIR or WIR blocks.

---

1. Partial-Interval Recording (PIR)

Operational Recording Rule

In Partial-Interval Recording (PIR), the observation period is divided into equal, consecutive time intervals (e.g., 10 seconds). The observer records a positive score ($+$) if the target behavior occurs at any point, for any duration, at any millisecond during that interval.

• If the behavior occurs at second 2 of a 10-second interval, the technician immediately marks a plus ($+$). The technician does not need to observe the client for the remainder of that interval regarding that behavior.
• If the behavior does not occur at all during the entire 10 seconds, the technician marks a minus ($-$).

$\text{PIR Percentage} = \frac{\text{Number of Intervals Scored Positive } (+)}{\text{Total Number of Intervals Observed}} \times 100$

Directional Measurement Artifacts of PIR

1. Systematically OVERESTIMATES Overall Duration: Because emitting a 1-second response causes an entire 15-second interval to be scored positive, PIR inflates the apparent amount of time the behavior occupied. If a child screams for 1 second in each of ten 15-second intervals, the child screamed for only 10 total seconds out of 150 seconds (true duration = $6.7%$). However, PIR records 10 out of 10 intervals positive ($100%$), overestimating the duration by fifteen-fold!
2. Systematically UNDERESTIMATES High-Rate Frequency: If an individual emits 8 distinct motor tics within a single 10-second interval, PIR records only a single positive mark ($+$). Multiple responses within the same interval are collapsed into one score, completely masking high-density behavioral bursts.

Clinical Indication: Target Behaviors for REDUCTION

PIR is the evidence-based standard for behaviors targeted for deceleration or reduction (e.g., physical aggression, vocal disruption, stereotypic hand-flapping, elopement, self-injurious behavior).

• Why Overestimation is Clinically Desirable for Behavior Reduction: In clinical behavior reduction, overestimation is clinically conservative. If our measurement tool tends to overestimate problem behavior, the behavior must truly and substantially decrease in the physical world before our data graph will reflect a downward trend. We will never prematurely declare an intervention successful due to a measurement illusion.

---

2. Whole-Interval Recording (WIR)

Operational Recording Rule

In Whole-Interval Recording (WIR), the observation period is divided into equal intervals (e.g., 10 or 15 seconds). The observer records a positive score ($+$) if and only if the target behavior persists continuously throughout the ENTIRE duration of the interval from the first millisecond to the last.

• If the target behavior begins at second 1 and terminates at second 9 of a 10-second interval, the observer MUST score a minus ($-$).
• A positive score is entered only if the behavior was emitted without a single second of interruption across all 10 seconds.

$\text{WIR Percentage} = \frac{\text{Number of Intervals Scored Positive } (+)}{\text{Total Number of Intervals Observed}} \times 100$

Directional Measurement Artifacts of WIR

1. Systematically UNDERESTIMATES Overall Duration: Because an interval is scored minus if the behavior stops even momentarily, WIR fails to capture substantial portions of ongoing behavior. If a child engages in cooperative play for 14 out of 15 seconds in every interval, WIR scores 0% of intervals, falsely reporting that zero cooperative play occurred!
2. Systematically UNDERESTIMATES Frequency: WIR completely ignores brief, discrete behavioral responses that do not span the entire interval.

Clinical Indication: Target Behaviors for ACQUISITION / INCREASE

WIR is the clinical standard for continuous behaviors targeted for acceleration or increase (e.g., sustained academic engagement, on-task behavior, independent play, remaining seated, social cooperative play).

• Why Underestimation is Clinically Desirable for Behavior Increase: Underestimation is clinically conservative for skill acquisition. By underestimating performance, WIR ensures that the clinician does not give premature credit for inconsistent or fragmented responding. A client is only reflected as mastering on-task behavior when their engagement is truly continuous, robust, and durable.

---

3. Momentary Time Sampling (MTS)

Operational Recording Rule

In Momentary Time Sampling (MTS), the observation period is partitioned into intervals (e.g., 1 minute, 2 minutes, or 5 minutes). The observer records whether the target behavior is occurring at the PRECISE INSTANT the interval elapses.

• During the interval, the observer does NOT need to watch the client at all. The observer can teach, grade papers, assist other learners, or interact with parents.
• An auditory or tactile signaling device (e.g., a Motivaider vibrating timer, a smartphone interval chime) alerts the observer the exact moment the timer expires.
• The observer immediately looks up at the client at that precise split-second. If the behavior is actively occurring at that moment, the observer marks ($+$). If the behavior is not occurring at that moment, the observer marks ($-$), even if the behavior occurred 2 seconds earlier!

$\text{MTS Percentage} = \frac{\text{Number of Moments Scored Positive } (+)}{\text{Total Number of Momentary Checks Observed}} \times 100$

Measurement Properties of MTS

Unlike PIR (which systematically overestimates) and WIR (which systematically underestimates), Momentary Time Sampling does NOT produce systematic directional bias across large aggregate samples.

• While individual momentary checks may randomly miss brief behaviors or catch isolated ones, across a full session or multiple days, MTS data correlate exceptionally well with true continuous duration data.
• The Golden Clinical Advantage: MTS completely liberates the practitioner from continuous observation. It is the premier data collection system for general education teachers managing inclusive classrooms, paraprofessionals supervising multiple students, and supervisors conducting periodic spot-checks.

---

4. Planned Activity Check (PLACHECK)

Operational Definition & Group Mechanics

Developed by Doke and Risley (1972), the Planned Activity Check (PLACHECK) is a specialized group adaptation of Momentary Time Sampling designed to quantify the collective engagement of groups of individuals in structured environments (e.g., classrooms, vocational workshops, group home living areas).

Implementation Protocol:

1. A target activity or engagement repertoire is operationally defined (e.g., "actively manipulating assigned vocational sorting materials" or "eyes directed at instructional board with pencil on paper").
2. Time intervals are established (e.g., every 5 minutes).
3. At the exact instant the interval expires, the observer performs a rapid, standardized visual scan across the group (e.g., scanning from left to right across the room).
4. The observer tallies two discrete numbers at that exact moment:
   • Numerator: The number of individuals who are actively engaged in the target activity.
   • Denominator: The total number of individuals present in the designated area.

$\text{PLACHECK Percentage for Interval } i = \frac{\text{Number of Individuals Engaged at Moment } T}{\text{Total Number of Individuals Present at Moment } T} \times 100$

$\text{Mean Session PLACHECK \%} = \frac{\sum_{i=1}^k \text{PLACHECK } \%_i}{k} \quad (\text{where } k \text{ is total scans})$

Step-by-Step Clinical Calculation Vignette:

A QASP-S evaluates student engagement during a 30-minute vocational packaging routine in an autism transition classroom. The supervisor conducts a PLACHECK scan every 10 minutes (3 total scans):

• Scan 1 (Minute 10): 6 students are present; 4 are actively packaging items. $\text{PLACHECK}_1 = \frac{4}{6} \times 100 = 66.7\%$
• Scan 2 (Minute 20): 6 students are present; 5 are actively packaging items. $\text{PLACHECK}_2 = \frac{5}{6} \times 100 = 83.3\%$
• Scan 3 (Minute 30): 5 students are present (1 departed for speech therapy); 3 are actively packaging items. $\text{PLACHECK}_3 = \frac{3}{5} \times 100 = 60.0\%$
• Mean Session PLACHECK: $\text{Overall PLACHECK} = \frac{66.7\% + 83.3\% + 60.0\%}{3} = 70.0\%$
• Administrative Utility: PLACHECK enables supervisors to objectively evaluate whether environmental modifications (e.g., changes in staffing ratios, visual supports, antecedent task modifications) increase active engagement across entire groups without requiring 1:1 data collectors for each student.

---

Comparative Analysis Matrix of Discontinuous Measurement Systems

---

Discontinuous Measurement Decision Tree