Observing Behavior Using A-B-C Data
Contributed by Cathy Pratt, Ph.D. & Melissa Dubie, M.Ed.
All members of the student’s individualized education program (IEP) can observe behavior to learn about patterns and functions of behavior. Everyone who observes behavior probably looks for similar characteristics of autism spectrum disorders (e.g., communication challenges, social deficits, restricted area of interests, sensory needs, etc.) and the impact on behavior. How information is gathered may be different for each person collecting the data and depending on the complexity of the situation. One format involves directly observing and recording situational factors surrounding a problem behavior using an assessment tool called ABC data collection. An ABC data form is an assessment tool used to gather information that should evolve into a positive behavior support plan. ABC refers to:
- Antecedent- the events, action, or circumstances that occur before a behavior.
- Behavior- The behavior.
- Consequences- The action or response that follows the behavior.
The following is an example of ABC data collection. ABC is considered a direct observation format because you have to be directly observing the behavior when it occurs. Typically it is a format that is used when an external observer is available who has the time and ability to observe and document behaviors during specified periods of the day. It is time and personnel intensive. From this data, we can see that when Joe is asked to end an activity he is enjoying (we know that he enjoys playing computer games), he screams, refuses to leave, and ignores. We also can see that the response to Joe’s refusal consists mostly of empty threats. If we follow Joe throughout the day, we may find that he is asked repeatedly to follow diections. In addition, the data reveals that Joe’s family uses threats that are not followed through. Joe has learned that persistence, ignoring, and refusal will wear parents down.
|Parent asks Joe to stop playing on the computer.||Joe screams, "NO!" and refuses to leave the computer.||Parent tells Joe to leave the computer again.|
|Parent tells Joe to leave the computer.||Joe again refuses to leave.||Parent starts counting to 10 as a warning to get off the computer.|
|Parent starts counting to 10 as a warning to get off the computer.||Joe does not move from the computer station.||Parent finishes counting to 10 and again warns him to get off the computer.|
|Parent finishes counting to 10 and again warns him to get off the computer.||Joe stays at the computer and refuses to leave.||Parent threatens that Joe lose computer privileges in the future.|
|Parent threatens that the Joe will lose computer privileges in the future.||Joe ignores and continues working on the computer.||The parent count to 10 again and again threatens future computer use.|
|The parent counts to 10 again and again threatens future computer use||Joe ignores and continues computer use.||The parent becomes angry and leaves the room.|
While it is important to look at both the antecedents and the form of the behavior, the focus of this article is on the consequence portion of the data collection. Examine the consequence portion of the data collection form when identifying those responses that both increase and decrease problem behavior. For example, if attention seems to increase problem behavior, then it may be important to teach the individual to get attention in a more appropriate fashion or to use attention for positive behaviors. If escape from a difficult task seems to be a consistent theme in the consequence section, then it may be important to either change the task or to teach the child to ask for help. And we may choose to use downtime as a reinforcer. Our responses should always focus on strengthening desired behavior, promoting the use of the replacement behavior, and decreasing the occurrence of the problem behavior (Sugai, et. al., 2000). An important aspect of this prospect is understanding those responses or consequences that maintain, and either enhance or decrease behavior over time.
Assessment is the key to developing an effective program and tracking the progress of individuals. Yet there are barriers in collecting the data such as time, remembering in a crisis situation, and being consistent. We can overcome these barriers by planning ahead, matching collection strategies to the setting, and simplifying the data collection chart. Remember anyone (e.g., parents, educators, teachers, support personnel, administrators) can take the data when given clear direction and parameters. Here is an example taken from what Joe’s parents know about his situation at home using the ABC approach. Notice the responses have already been established on the form. These are the responses that are typically identified as motivating behavior. While this system may be more efficient, you will note that much of the richness of the narrative is missing.
|Parent asks Joe to stop playing on the computer.||Joe screams, "NO!" and refuses to leave the computer.|
|Parent tells Joe to leave the computer.||Joe again refuses to leave.|
|Parent starts counting to 10 as a warning to get off the computer.||Joe does not move from the computer station.|
|Parent finishes counting to 10 and again warns him to get off the computer.||Joe stays at the computer and refuses to leave.|
|Parent threatens that the Joe will lose computer privileges in the future.||Joe ignores and continues working on the computer.|
|The parent counts to 10 again and again threatens future computer use||Joe ignores and continues computer use.|
Sometimes the ABC data collection form is used to document a behavior incident. At the end of this article is an example of this type of form. Remember that this type of form will give you limited data and focuses heavily on negative behaviors. However, it is easier when someone is not available to do more indepth observing. In truth, the ABC data collection should not be used just to document behavior incidents. It is best used as a narrative during a specified time of the day. Equally important is to document those conditions that surround positive behaviors. By documenting these, professionals and family members can identify effective strategies that can be replicated.
Once accurate and sufficient data is collected; placements, planning, modifications, instruction, and feedback are easier, more valid, and effective (Morton & Lieberman, 2006). ABC data collection can be used for all individuals with behavior issues at home and in school, not just individuals on the autism spectrum.
Sugai, G., Horner, R.H., Dunlap, G., Hieneman, M., Nelson, C.M., Scott, T., Liaupsin, C., Sailor, W., Turnbull, A.P., Turnbull III, H.R.; Wickham, D., Wilcox, B., and Ruef, M. (2000). Applying positive behavior support and functional behavioral assessment in schools. Journal of Positive Behavior Interventions, 2(3), 131-143.
Morton & Lieberman, 2006. Strategies for collecting data in physical education. Teaching Elementary Physical Education, 17(4), 28-31.
Pratt, C., & Dubie, M. (2008). Observing behavior using a-b-c data. The Reporter, 14(1), 1-4.
IRCA | 1905 North Range Rd. | Bloomington, IN 47408 | 812-855-6508 | firstname.lastname@example.org
How can Ms. Rollison determine why Joseph behaves the way he does?
Page 7: Collect Data: Direct Observations
In addition to the information gathered from interviews and rating scales, data from direct observations can offer insight into when, where, and how often a behavior occurs, as well as how long it lasts. A direct observation occurs when someone actually sees the student in the classroom setting and gathers data on the problem behavior. Ideally, an objective observer (e.g., a behavior analyst, a member of the S-Team, another teacher) will collect the data. Direct observations can be used to:
- Conduct an ABC analysis
- Collect baseline data about the problem or target behavior
Conducting an ABC Analysis
Recall that the ABC model is used to identify the antecedents (A) that set the stage for the problem behavior (B) to occur and the consequences (C) that appear to be maintaining the problem behavior. An observer might collect data over several sessions before obtaining enough information for a clear ABC pattern to emerge. This usually requires eight to ten occurrences of the problem behavior (except in cases of extreme behaviors, such as fighting or self-injurious actions). In addition to recording the ABC events, the observer should note the setting and the time of day in which the behavior took place, as well as the persons involved. Once an ABC analysis has been completed, the team can develop a hypothesis about the function of the behavior. Click here to see David's A-B-C analysis results.
The video below depicts an interaction between a teacher and a student who refuses to do his work. During the video, Kathleen Lane conducts an ABC analysis, explaining each step and demonstrating how to fill out the recording form (time: 5:08).
View Transcript | View Transcript with Images | Click to view Cameron's ABC analysis form
Collecting Baseline Data
Once a clear ABC pattern of behavior has been identified, further observation can provide baseline data about the problem or target behavior. It is important to select a direct observation method that fits the behavior and is feasible in the classroom. Observers should consider one of four direct observation methods when collecting this type of data:
|}||if length of time is a measurable factor in the problem behavior|
|}||if the teacher needs to determine how often a behavior occurs|
Before implementing the intervention, baseline data should be collected over three to five observational periods to ensure a representative sampling of the behavior. The same data collection procedures should be repeated once the intervention is implemented. This allows the teacher to compare the intervention data to the baseline data to determine whether the intervention is effective. In David's case, the team wanted to know how much time he spent off-task, so they selected duration recording.
For Your Information
Data need not be collected throughout the entire day. If, for example, a behavior occurs primarily during independent reading time, data need only be gathered for a portion of that period.
Data may also be collected on replacement behaviors. It may be necessary to have one recording system for the problem behavior and a different system for the replacement behavior.
Ms. Rollison has identified the following problem behaviors for Joseph:
- Joseph makes sarcastic and teasing comments after other students answer questions in class.
- Joseph makes rude comments when called on or spoken to by the teacher.
Which data collection method could she use for his problem behaviors? (check all that apply)
Duration recording: Try again. Duration recording is not appropriate because Joseph’s problem behaviors are brief comments that do not last for any length of time.
Latency recording: Try again. Latency recording is not appropriate because Joseph does not delay doing something that Ms. Rollison has asked him to do.
Event recording: Good work! Event recording is appropriate, as each of Joseph’s problem behaviors (i.e., comments) can be counted.
Interval recording: Good work! Interval recording is appropriate, as an observer would be able to identify whether either behavior occurred during a specified period of time.
Ms. Rollison has identified the following replacement behaviors for Joseph:
- Joseph will respond in a positive and respectful manner when called on or spoken to by the teacher.
- Joseph will listen without commenting when other students answer questions in class.
Duration recording: Good work. Duration recording can be used to record the length of time that Joseph listens without making comments.
Latency recording: Try again. Latency is not appropriate because Joseph does not delay doing something that Ms. Rollison has asked him to do.
Event recording: Good work! Event recording is appropriate, as each of Joseph’s positive or respectful comments can be counted.
Interval recording: Good work! Interval recording is appropriate, as an observer would be able to identify whether either replacement behavior occurred during a specified period of time.
Click here to see how data were collected on David's off-task behavior.
David's teacher schedules ten minutes at the end of language arts class for students to work on independent writing assignments. Because the information gathered thus far indicate that this is when David is the most off-task (i.e., out-of-seat behavior), an observer—in this case, a member of the S-Team—takes this opportunity to collect duration data. Whenever David leaves his seat without permission, the observer starts his stopwatch. When David returns to his seat, he stops the stopwatch. He maintains this data collection procedure for the entire 10-minute period. At the end of the period, his stopwatch indicates that David was off-task for 7 minutes and 29 seconds. He divides the 7 minutes and 29 seconds by the entire 10-minute independent work period and determines that David was off-task 75% of the time.
The observer makes two more visits to David's classroom, collecting data during the independent seatwork period. A relatively stable pattern of out-of-seat behavior emerges:
Observation 1: 75%
Observation 2: 54%
Observation 3: 61%
The baseline data indicate that David is out of his seat an average of 63% of the independent seatwork time, confirming the teacher's initial concern that David's off-task behavior is a problem.
David’s Baseline Data graph: This double-line plot graph shows David’s Baseline Data. The x-axis is labeled “Observation”; observations 1 through 3 are labeled on the axis. The y-axis is labeled “percent;” 0 to 80 percent are labeled in 10 percent intervals. The first graph is yellow and labeled “Replacement (on-task)” in the key to the right of the graph. This graph has three plot points corresponding with the three observations. The points are at 25%, 45%, and 40%. The second graph is red and labeled “Problem behavior (off-task).” This graph has three plot points corresponding with the three observations. The points are at 75%, 52%, and 60%.