Knowing that an intervention works is only the first step. Behavior analysts also want to know why it works, how much of it is needed, and whether something else might work better. Component, parametric, and comparative analyses answer those three questions. This guide explains each type, how it is done, and how to tell them apart on the BCBA exam.
Key takeaways
- Component analysis: which parts of an intervention are responsible for the effect?
- Parametric analysis: which value or amount of the same intervention works best?
- Comparative analysis: which of two or more different interventions works better?
- Component analyses can be drop-out or add-in.
- Watch for sequence effects and multiple treatment interference.
In this guide
Why analyze an intervention further?
A basic experimental design shows a functional relation between an independent variable and behavior. But many interventions have several parts, can be delivered at different intensities, and compete with other treatment options. Analyzing the intervention further helps you make treatment more efficient (cut parts that do not help), more effective (use the best amount), and better matched to the learner (choose the better option).
Component analysis
A component analysis identifies which parts of a treatment package are responsible for behavior change. For example, a package might include prompting, praise, and tokens. A component analysis tells you whether all three are needed, or whether one is doing most of the work.
Drop-out vs. add-in
| Approach | How it works | Example |
|---|---|---|
| Drop-out analysis | Start with the full package, then remove components one at a time | Remove tokens, then praise, and watch whether performance drops |
| Add-in analysis | Start with one component, then add components one at a time | Begin with prompting alone, then add praise, then tokens |
Watch out for: components that only work in combination, and sequence effects, where the order in which components are tested affects the results.
Parametric analysis
A parametric analysis compares different values of the same independent variable. It answers questions about “how much” or “how often.”
- Magnitude of reinforcement (1 token vs. 4 tokens)
- Schedule of reinforcement (FR 2 vs. FR 5 vs. FR 10)
- Delay to reinforcement (immediate vs. 30 seconds)
- Session length or practice trials per day
Parametric analyses often show that more is not always better. In the example above, performance levels off after 4 tokens, so 8 tokens would cost more without adding much benefit.
Comparative analysis
A comparative analysis compares two or more different interventions to see which is more effective for the learner. For example, you might compare picture exchange with sign language for teaching requests, or compare two prompting procedures.
- Alternating treatments design: rapidly alternate the interventions and compare their data paths. This is efficient and does not require a long baseline.
- Multiple-treatment reversal design: introduce interventions in sequential phases, such as A-B-A-C-A-B.
- Concern: multiple treatment interference and carryover effects, because one learner experiences several interventions.
Comparing the three analyses
| Component | Parametric | Comparative | |
|---|---|---|---|
| Question | Which parts work? | Which value works best? | Which intervention works better? |
| What changes | The parts of one package | The amount of one variable | The type of intervention |
| Example | Prompting vs. prompting + praise + tokens | 1 vs. 2 vs. 4 tokens | Picture exchange vs. sign language |
| Common designs | Reversal, multiple baseline | Reversal, alternating treatments | Alternating treatments, multiple-treatment reversal |
| Main risk | Sequence effects; components that only work together | Too few values tested to see the full pattern | Multiple treatment interference |
Exam tip: if the options differ only in amount, it is parametric. If they are different procedures, it is comparative. If they are parts of one package, it is a component analysis.
Test yourself
Practice question 1
A BCBA compares the effects of 30-second, 2-minute, and 5-minute breaks on a learner’s task completion. This is an example of:
- A component analysis
- A parametric analysis
- A comparative analysis
- A functional analysis
Show the answer
B. A parametric analysis. The BCBA is comparing different values (break lengths) of the same independent variable. A comparative analysis would compare different interventions, and a component analysis would compare parts of a package.
Practice question 2
An intervention for on-task behavior includes self-monitoring, a visual timer, and teacher praise. The team removes the visual timer, then the praise, to see whether on-task behavior changes. This is:
- An add-in component analysis
- A drop-out component analysis
- A parametric analysis
- A comparative analysis
Show the answer
B. A drop-out component analysis. The team starts with the full package and removes components one at a time to see which ones are responsible for the effect.
Master experimental design, one task at a time
The free BCBA Study Checklist covers every 6th edition task, including experimental design and evaluating interventions, so you can track exactly what you have mastered.
Component, parametric, and comparative analysis FAQ
What is a component analysis in ABA?
A component analysis is an experiment that identifies which parts of a multi-component intervention are responsible for behavior change. It can be done by removing components one at a time from the full package (drop-out) or by adding components one at a time (add-in).
What is a parametric analysis?
A parametric analysis compares different values of the same independent variable, such as different amounts of reinforcement, session lengths, or reinforcement delays, to find out how the size of the variable affects behavior.
What is a comparative analysis?
A comparative analysis compares two or more different interventions to see which one is more effective for a learner. Alternating treatments and multiple-treatment reversal designs are commonly used.
What is the difference between parametric and comparative analysis?
A parametric analysis compares different values of the same intervention (for example, 1 token vs. 4 tokens). A comparative analysis compares different interventions (for example, picture exchange vs. sign language).
What is the difference between drop-out and add-in component analysis?
In a drop-out analysis, you start with the full package and remove components one at a time. In an add-in analysis, you start with one component and add others one at a time. Both can be affected by sequence effects.
What is multiple treatment interference?
Multiple treatment interference happens when the effects of one intervention are influenced by exposure to another, which can occur when several conditions are given to the same learner. It is a key concern in comparative and component analyses.
Related reading: Understanding Independent Variables, Dependent Variables, and Experimental Control · What Are Experimental Designs? · Understanding Reversal Designs
References
- Behavior Analyst Certification Board. (2022). BCBA test content outline (6th ed.).
- Cooper, J. O., Heron, T. E., & Heward, W. L. (2020). Applied behavior analysis (3rd ed.), Chapters 8 and 10. Pearson.
- Kazdin, A. E. (2011). Single-case research designs: Methods for clinical and applied settings (2nd ed.). Oxford University Press.
- Ward-Horner, J., & Sturmey, P. (2010). Component analyses using single-subject experimental designs: A review. Journal of Applied Behavior Analysis, 43(4), 685–704.