In applied behavior analysis, the graph is where decisions get made. Behavior analysts rely on graphed data, not statistics alone, to see whether behavior is changing and whether an intervention is responsible. This guide covers the types of graphs used in ABA, the parts of a well-built line graph, how to read a graph using visual analysis, and the graphing mistakes that cost points on the BCBA exam.
Key takeaways
- Line graphs are the workhorse of ABA: they show behavior over time and across conditions.
- Other graph types: bar graphs, cumulative records, scatterplots, and the Standard Celeration Chart each answer different questions.
- A good line graph has seven parts, and data points are never connected across a condition change line.
- Visual analysis looks at six features: level, trend, variability, immediacy of effect, overlap, and consistency.
- On the exam, expect questions on choosing a graph type, reading a graph, and spotting graphing errors.
In this guide
Why graphs matter in ABA
Graphs let behavior analysts see patterns that are hard to spot in a table of numbers: whether behavior is stable, improving, or getting worse, and whether those changes line up with changes in the intervention. Because decisions are made session by session, graphs are used formatively, to adjust treatment as it happens, not just at the end.
On the BCBA exam, graphing falls under Domain C, Measurement, Data Display, and Interpretation (21 scored questions), especially tasks C.10 (graphing data) and C.11 (interpreting graphed data).
Types of graphs used in ABA
| Graph | What it shows | Best used for |
|---|---|---|
| Line graph | Behavior across time (sessions, days) with data points connected by a data path | Almost everything: tracking progress and comparing conditions |
| Bar graph | Summary values, such as averages, as bars | Comparing conditions, people, or behaviors at a glance |
| Cumulative record | The running total of responses over time; slope shows rate | Showing response rate and local changes in rate |
| Scatterplot | Individual data points plotted against two variables, often time of day | Finding when or under what conditions behavior happens |
| Standard Celeration Chart | Rate on a semilogarithmic (multiply/divide) scale | Precision teaching; showing proportional change (celeration) |
Line graphs, bar graphs, cumulative records, and scatterplots are equal-interval graphs: the same distance on an axis always represents the same amount. The Standard Celeration Chart is semilogarithmic: equal distances on the vertical axis represent equal proportional changes, so a change from 2 to 4 looks the same as a change from 20 to 40.
The parts of an ABA line graph
| # | Part | What to know |
|---|---|---|
| 1 | Vertical (y) axis and label | The dependent variable (the behavior measure), usually starting at 0 |
| 2 | Horizontal (x) axis and label | Time, usually sessions or days |
| 3 | Condition change line | A vertical line, often dashed, marking a change in conditions |
| 4 | Condition labels | Brief names for each condition, such as Baseline and Intervention |
| 5 | Data points | Each point is one measurement |
| 6 | Data path | The line connecting data points within a condition |
| 7 | Figure caption | A short description of what the graph shows |
Graphing conventions to follow
- Do not connect data across a condition change line. Each condition gets its own data path.
- Do not connect data across missing sessions or long breaks (for example, a holiday). Leave a gap.
- Start the vertical axis at zero in most cases, so changes are not exaggerated.
- Keep a sensible proportion. A common guideline is a vertical axis about two-thirds the length of the horizontal axis, so slopes are not distorted.
- Label everything: both axes, every condition, and the figure.
- Use different symbols (filled and open circles, triangles) when plotting more than one data path.
How to read a graph: visual analysis
Visual analysis is how behavior analysts decide, from the graph, whether behavior changed and whether the intervention caused it. Start within each condition, then compare between conditions.
| Feature | What to ask |
|---|---|
| Level | Where do the data sit on the vertical axis? Did the level change between conditions? |
| Trend | Are the data going up, down, or staying flat? Did the direction change? |
| Variability | How much do the data bounce around? High variability makes effects harder to see. |
| Immediacy of effect | How quickly did behavior change after the condition changed? Compare the last few points of one condition with the first few of the next. |
| Overlap | How many data points in one condition fall in the same range as the previous condition? Less overlap means a clearer effect. |
| Consistency | Do similar conditions (for example, both baselines in an ABAB design) show similar patterns? |
One clear change between two conditions is a basic effect. To show a functional relation, you need that effect replicated, as in an ABAB reversal or a multiple baseline design.
Before you change conditions: check stability
Wait for steady state responding before changing conditions: data with little variability and no strong trend, especially no trend in the direction you expect the intervention to produce. Otherwise, you will not be able to tell whether a later change came from the intervention or was already happening.
How to read a cumulative record
- Steeper slope means a higher response rate.
- A flat, horizontal line means no responses occurred.
- The line never goes down, because each response is added to the total.
- Overall response rate is the average rate across the whole record; local response rate is the rate over a shorter segment.
Common graphing mistakes
- Connecting data across a condition change line or across missing sessions.
- Truncating the vertical axis so small changes look dramatic.
- Stretching or squashing the axes, which distorts slopes.
- Missing labels on axes, conditions, or the figure.
- Changing conditions before the data are stable.
- Using a bar graph when you need to show change over time, since bars hide trend and variability.
Practice question
A BCBA is reviewing a cumulative record of a learner’s requests. For the last 10 minutes of the session, the line is flat and horizontal. What does this indicate?
- The learner’s rate of requesting was very high
- The learner made no requests during that time
- Data collection stopped
- The learner’s requests decreased to a lower, steady rate
Show the answer
B. The learner made no requests during that time. On a cumulative record, each response adds to the running total, so a flat line means no new responses. A steep slope would mean a high rate, and a gentler slope would mean a lower but steady rate. A flat line does not by itself show that data collection stopped.
Master Domain C, one task at a time
The free BCBA Study Checklist lists all 12 Measurement tasks, including graphing and interpreting data, so you can track exactly what you have mastered.
ABA graphs FAQ
What is the most common graph in ABA?
The equal-interval line graph is the most common graph in applied behavior analysis. It shows how a behavior changes over time, across conditions such as baseline and intervention.
What are the main types of graphs used in ABA?
Line graphs, bar graphs, cumulative records, scatterplots, and semilogarithmic charts such as the Standard Celeration Chart.
What is visual analysis in ABA?
Visual analysis is the systematic inspection of graphed data to decide whether behavior changed and whether the change was caused by the intervention. Analysts look at level, trend, and variability within and across conditions, along with immediacy of effect, overlap, and consistency across similar conditions.
Why are data points not connected across a condition change line?
Connecting data across a condition change line would suggest the data are continuous across two different conditions. Leaving a break makes it easy to compare the conditions and see any change at the moment the condition changed.
How do you read a cumulative record?
The steeper the slope, the higher the response rate. A flat, horizontal line means no responses occurred. Because each response is added to the running total, a cumulative record never goes down.
What does the Standard Celeration Chart show?
The Standard Celeration Chart is a semilogarithmic chart that shows celeration, the factor by which a behavior’s rate multiplies or divides over time. Equal distances on the vertical axis represent equal proportional changes.
Related reading: Types of IOA in Various Data Collection Methods · Understanding Multiple Baseline Design · 20 Free BCBA Practice Questions
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.), Chapter 6: Constructing and interpreting graphic displays of behavioral data. Pearson.
- Kratochwill, T. R., Hitchcock, J., Horner, R. H., Levin, J. R., Odom, S. L., Rindskopf, D. M., & Shadish, W. R. (2010). Single-case designs technical documentation. What Works Clearinghouse.
- Lane, J. D., & Gast, D. L. (2014). Visual analysis in single case experimental design studies: Brief review and guidelines. Neuropsychological Rehabilitation, 24(3–4), 445–463.
- Ledford, J. R., Lane, J. D., & Severini, K. E. (2018). Systematic use of visual analysis for assessing outcomes in single case design studies. Brain Impairment, 19(1), 4–17.