BA420 Unit 9 Multiple Baseline Paper Example

Three Students, Staggered Starts and a Tier That Waited Its Turn: A BA420 Unit 9 Multiple Baseline and Multiple Probe Paper

[Student Name]

Purdue University Global

BA420: Methods of Behavior Analysis

Unit 9 Multiple Baseline and Multiple Probe Paper

[Instructor Name]

[Date]

The writer, the school, the students and every data point are composite and illustrative; the articles and book cited are real.

Why Not Withdraw

The middle school where I work in Wichita adopted self-monitoring this spring for students who lose focus during independent work. A student marks a sheet at regular vibrations of a timer, indicating whether they were working, and compares marks with the teacher's at the end of the period. Self-monitoring is meant to teach a lasting skill, so withdrawing it to show its effect would be both awkward and unlikely to work: if students learn to monitor themselves, removing the sheet may not return them to baseline. The district behavior analyst chose a multiple baseline design across students instead.

The Multiple Baseline Across Students

In a multiple baseline design, baseline data are collected on several tiers at once, such as different students, behaviors or settings, and the intervention is introduced to one tier at a time. Experimental control is shown when each tier's behavior changes only after the intervention reaches it, while tiers still in baseline remain stable. Kazdin (2011) describes the design's logic as replication across tiers, with the staggered start ruling out explanations, such as a schoolwide event, that would affect all tiers at once.

Three students, with their families' consent, participated: Jaylen in sixth-grade math, Tyrell in sixth-grade science and Sofia in seventh-grade English. I measured on-task behavior with ten-second momentary time sampling during independent work, with a second observer scoring about a third of sessions for agreement. All three began baseline in the same week of February. Jaylen's baseline averaged 48 percent on task; Tyrell's, 55 percent; Sofia's, 51 percent.

Self-monitoring began for Jaylen after five baseline sessions. His on-task percentage rose to the mid-80s within three sessions, while Tyrell's and Sofia's stayed near their baselines. After Jaylen's change was clear, self-monitoring began for Tyrell, whose on-task behavior rose to about 85 percent while Sofia's still did not change. Finally Sofia began, and her on-task behavior rose to about 88 percent, the highest of the three, which her teacher attributed to Sofia's enjoying the comparison with the teacher's marks.

Sofia's flat line during the weeks the other two improved was as important to the argument as any line that went up.

Because each student's behavior changed only when self-monitoring began, and the students still in baseline did not change, the design produced three demonstrations, each at a different point in the semester.

The Multiple Probe Variation

A second project used a variation. Sofia was also learning a four-step note-taking routine: write the date and topic, list three key words during the lesson, write a one-sentence summary at the end and circle one question. The team wanted to teach it in English, then science, then social studies. Measuring her note-taking every day in all three subjects during baseline would have meant weeks of repeated testing in subjects where she had not been taught the routine, which could be frustrating and could itself change her behavior.

Horner and Baer (1978) described the multiple probe technique as a variation of the multiple baseline in which baseline data are collected intermittently, through probes, rather than continuously. Probes are taken at the start, just before the intervention begins on each tier, and at other points, enough to show that untreated tiers remain stable without measuring them constantly.

Sofia's notes were scored after three probe lessons in each subject before teaching began in English. During English teaching, single probes in science and social studies showed no change. Probes taken just before teaching in science, and later in social studies, confirmed stable baselines, and each subject's scores rose only after the routine was taught there.

Concurrent and Nonconcurrent Designs

Both of these designs were fully concurrent: all tiers were measured during the same calendar weeks. Some multiple baseline studies are nonconcurrent, with tiers starting at different calendar times, for example because participants enter a program months apart. Carr (2005) recommended that researchers report multiple baseline designs across participants clearly enough that readers can see whether tiers were concurrent, since nonconcurrent designs offer weaker protection against history effects, events that happen at a particular time and could explain changes. Our team now notes start dates for every tier on the graph and in each progress report, so that anyone reading the data can tell at a glance whether the tiers ran at the same time.

Choosing Tiers and Timing

Two decisions shaped the design before any data were collected. The first was the choice of tiers. The district behavior analyst chose three students in three different classrooms, so that self-monitoring for one student was unlikely to spill over to the others; three students in the same class would have risked one student's sheet prompting a neighbor to work harder. The second was the rule for moving to the next tier. The team agreed in advance that a new student would begin self-monitoring only after the previous student showed a clear change for at least three consecutive sessions and the next student's baseline was stable. That rule meant Sofia waited about two weeks longer than Jaylen, but it protected the logic of the design. In practice, I also checked each baseline graph with the analyst before each start, since a rising baseline in a waiting tier would have made the next demonstration hard to interpret.

Limits

Multiple baseline designs require that tiers be independent: if self-monitoring for Jaylen had changed how the teacher treated the whole class, Tyrell's or Sofia's behavior might have changed early. They also require that some participants wait for the intervention, which can be hard to justify when it seems to work. The team kept waits short, adding each new student as soon as the previous one's change was clear, and told each family when their child's turn would come and why.

Conclusion

The multiple baseline across Jaylen, Tyrell and Sofia showed that self-monitoring raised on-task behavior, with each student changing only when the intervention began and the waiting students staying stable. The multiple probe variation taught Sofia a note-taking routine across three subjects without weeks of repeated testing. Kazdin's account of the design's logic, Horner and Baer's probe technique and Carr's reporting recommendations together describe how to show control without withdrawing a skill students are meant to keep.

References

Carr, J. E. (2005). Recommendations for reporting multiple-baseline designs across participants. Behavioral Interventions, 20(3), 219-224. https://doi.org/10.1002/bin.191

Horner, R. D., & Baer, D. M. (1978). Multiple-probe technique: A variation of the multiple baseline. Journal of Applied Behavior Analysis, 11(1), 189-196. https://doi.org/10.1901/jaba.1978.11-189

Kazdin, A. E. (2011). Single-case research designs: Methods for clinical and applied settings (2nd ed.). Oxford University Press.