Partial Interval vs Whole Interval vs MTS: ABA Guide

In Applied Behavior Analysis (ABA), accurate data collection is the backbone of successful interventions. For BCBAs and RBTs, choosing the right measurement method can mean the difference between identifying progress and missing critical behavioral patterns. Discontinuous measurement procedures, such as partial interval vs whole interval vs MTS, offer efficient alternatives to continuous recording. However, they come with trade-offs in accuracy. This guide breaks down these methods to help you select and implement them confidently, supporting ethical, evidence-based practice as outlined in the BACB Ethics Code.

Here’s what this guide covers:

• Key differences between Partial Interval, Whole Interval, and Momentary Time Sampling.
• How measurement bias in ABA can affect your data and how to minimize it.
• A decision framework for choosing the right method for different behaviors and goals.
• Practical tips for RBTs to ensure high interobserver agreement (IOA).

What is Discontinuous Measurement in ABA?

Discontinuous measurement procedures sample behavior across predefined intervals rather than tracking every single occurrence. This approach is well-suited for busy clinical settings. These methods estimate a behavior's prevalence without requiring an observer's constant attention, which is crucial for high-frequency or ongoing behaviors. The main types are partial interval recording (PIR), whole interval recording (WIR), and momentary time sampling (MTS).

Unlike continuous methods, these approaches divide observation time into equal intervals and record data based on specific rules. This efficiency reduces observer fatigue but can introduce measurement bias. Understanding these trade-offs is key to making sound clinical decisions. For more details on different measurement types, see our guide on ABA procedures.

Partial Interval Recording (PIR): Overestimation and When to Use It

Partial Interval Recording (PIR) involves marking whether a target behavior occurs at any point during a short interval. The final data is reported as the percentage of intervals where the behavior was recorded.

PIR is efficient but has a known overestimation bias because even a brief instance counts for the whole interval. For this reason, PIR is best for behaviors you want to decrease. As confirmed in a review on discontinuous methods, its tendency to overestimate makes it very sensitive to any reduction, helping you see if an intervention is working for disruptive actions.

Whole Interval Recording (WIR): Underestimation and Its Purpose

Whole Interval Recording (WIR) is the opposite of PIR. It requires the target behavior to occur continuously throughout the entire interval to be scored. This strict approach is used to measure sustained engagement, such as "on-task behavior."

To implement WIR, you define the behavior and interval but only score "yes" if the behavior persists from the start to the end. WIR has an underestimation bias, as it misses any behavior that stops and starts within the interval. This makes it well-suited for goals focused on increasing a behavior's duration. For example, WIR is excellent for tracking progress on a goal to improve sustained attention, as supported by practitioner guidelines. For related continuous methods, see our guide on ABA latency, IRT, and duration.

Momentary Time Sampling (MTS): A Balance of Efficiency and Accuracy

Momentary Time Sampling (MTS) offers a practical compromise by sampling behavior only at the end of each interval. This "snapshot" method records if the behavior is occurring at that precise moment. To use MTS, you only need to observe and record at the exact instant the interval ends, making it the least labor-intensive of the three methods.

MTS is generally less biased than PIR or WIR but can miss behaviors that occur at other times in the interval. It is a good choice for tracking ongoing states like academic engagement in a busy classroom.

Side-by-Side Comparison: Partial Interval vs Whole Interval vs MTS

Choosing ABA data collection methods depends on the behavior, your goal, and available resources. Here’s a quick comparison.

Method	Bias	Best for Behavior Reduction	Best for Acquisition	Observer Effort
PIR	Overestimation	High (detects any occurrence)	Low (inflates baseline)	Medium
WIR	Underestimation	Low (misses partial occurrences)	High (confirms sustained skill)	High
MTS	Minimal/Mixed	Medium (snapshot reliability)	Medium (estimates ongoing state)	Low

As outlined in ABA teaching suggestions, selecting the right method is crucial for data integrity.

Summary of Pros and Cons

• PIR Pros & Cons: It is excellent for capturing any instance of a behavior you want to reduce. However, its overestimation bias can make a behavior appear more frequent than it is.
• WIR Pros & Cons: This method is effective for increasing sustained behaviors like attention. Its main drawback is the underestimation bias, which ignores behavior that doesn't last the full interval.
• MTS Pros & Cons: MTS is highly efficient and suitable for busy settings. The risk is that it can miss behaviors that happen just before or after the observation point.

BCBA Decision Tree: Choosing the Right Method

For BCBAs, a decision tree can simplify method selection. A model for selecting measurement procedures supports this approach:

1. Start with the behavior type. Is it brief and discrete (e.g., hitting)? PIR is effective. Is it continuous (e.g., on-task)? WIR is better for measuring duration.
2. Consider the goal. For behavior reduction, PIR's sensitivity helps detect small decreases. For behavior acquisition, WIR ensures the skill is truly maintained. For general tracking, MTS provides a good estimate with minimal effort.
3. Evaluate the setting. In group sessions or with multiple clients, MTS is often the most practical choice due to its efficiency.

Tips for RBTs: Ensuring High Interobserver Agreement (IOA)

RBTs are crucial for data fidelity. A key part of this is Interobserver Agreement (IOA), which measures consistency between two independent observers. Aim for 80% agreement or higher across at least 20% of sessions, per the RBT Handbook.

• For PIR and WIR, use interval-by-interval IOA. Compare your "yes/no" marks for each interval.
• For MTS, use exact-agreement IOA. You only need to agree on the behavior's status at the exact moment the interval ended.
• General Tips: Always use the same operational definition, observe independently, and calculate IOA using the formula: (agreements / total opportunities) × 100.

Frequently Asked Questions

What is the main difference between partial and whole interval recording?

Partial interval recording (PIR) scores a behavior if it occurs at any point during an interval, which can overestimate frequency. Whole interval recording (WIR) requires the behavior to occur throughout the entire interval, which often leads to underestimation. PIR is better for behaviors you want to reduce, while WIR is suited for skills you want to increase.

When should RBTs use momentary time sampling?

RBTs should use MTS when observation time is limited and behaviors are ongoing, like on-task engagement. It samples behavior only at the end of each interval, making it highly efficient for busy sessions with multiple clients.

How does measurement bias impact accuracy in ABA data?

Measurement bias, such as PIR's overestimation and WIR's underestimation, can distort an intervention's perceived effectiveness. This can lead to poor clinical decisions. According to practice guidelines, choosing the right method for the target behavior is critical to minimize this bias and ensure data is accurate.

How do I choose the best ABA data collection method?

To choose the best method, first define the behavior (is it brief or sustained?). Then, clarify your goal (are you trying to increase or decrease it?). Finally, consider your environment (can you observe continuously or do you need an efficient sampling method?). This framework helps align your measurement choice with the behavior and goal.

What are common pitfalls when choosing a data collection method?

A common pitfall is using WIR for brief behaviors you want to reduce, as it will likely miss them. Another is ignoring PIR's overestimation bias. As noted in ABA teaching suggestions, you should always align your method with the behavior's characteristics to avoid such errors.

How can BCBAs improve interobserver agreement (IOA)?

To improve IOA, BCBAs should ensure observers are well-trained on the recording rules and use clear, objective behavioral definitions. It's also vital to collect IOA data regularly and aim for at least 80% agreement, as required by the official BACB RBT Handbook.

Selecting the right discontinuous method empowers BCBAs and RBTs to gather trustworthy data. From PIR's sensitivity for behavior reduction to WIR's rigor for building skills and MTS's practicality, each method addresses unique clinical needs.

By understanding the inherent bias of each system and prioritizing interobserver agreement, you ensure your data reflects client behavior. This commitment to accuracy in ABA data not only improves intervention outcomes but also upholds ethical, evidence-based practice in the partial interval vs whole interval vs MTS debate.