| Course | PSY 803 Behaviorism |
|---|---|
| Topic | Topic 3 |
| Paper type | Single-subject design evaluation |
| Length | About 950 words, 2 pages |
| Format | APA 7 student paper |
| School | Grand Canyon University (GCU) |
| Program | PhD in Psychology |
| Updated | October 2026 |
Free sample paper for PSY 803 Topic 3
Control Shown in One Subject: Evaluating the Reversal Design, Its Costs and the Sampling Objection
[Author Name]
College of Doctoral Studies, Grand Canyon University
PSY-803: Behaviorism
Topic 3 Single-Subject Design Evaluation
[Instructor Name]
[Date]
The organization, setting and figures below are a composite written as a model document. No real employer, client, colleague or patient is described.
Control Shown in One Subject: Evaluating the Reversal Design, Its Costs and the Sampling Objection
Why a Few Subjects, Measured Repeatedly
The experimental analysis of behavior preferred to study a few organisms intensively, measuring each repeatedly under changing conditions, rather than averaging single measurements across many. The preference was not convenience. Sidman (1960) argued that an average across subjects can describe a curve no individual actually followed, and that confidence in a finding comes from replication: seeing the same effect again in the same subject when a condition is reintroduced, and again in other subjects. Statistical significance across a group, on this view, was a weaker basis for confidence than an effect produced and reproduced in individual organisms. This evaluation works through one design built on that logic, states its costs and answers the objection that a few subjects cannot support general conclusions.
A Reversal Design, Worked Through
A reversal design, often labeled ABAB, alternates between a baseline condition (A) and an intervention condition (B). A composite example shows how it demonstrates control. A third-grade student frequently leaves their seat during independent reading time. Observers record the percentage of ten-second intervals in which the student is out of seat across 20-minute sessions.
In the first baseline, over five sessions, the student is out of seat in about 45 to 55 percent of intervals, with no trend. In the first intervention phase, the teacher gives a token, exchangeable later for free time, for each five minutes spent in seat; over five sessions, out-of-seat intervals fall to about 10 to 15 percent. In the return to baseline, tokens are withdrawn, and out-of-seat behavior rises again to about 40 percent. In the second intervention phase, tokens are reinstated and the behavior falls again to about 10 percent.
Table 1
Composite Reversal Design Data
| Phase | Condition | Sessions | Out-of-seat intervals, range |
|---|---|---|---|
| A1 | Baseline | 1 to 5 | About 45 to 55 percent |
| B1 | Tokens for time in seat | 6 to 10 | About 10 to 15 percent |
| A2 | Tokens withdrawn | 11 to 15 | About 35 to 45 percent |
| B2 | Tokens reinstated | 16 to 20 | About 8 to 12 percent |
The behavior tracks the condition. Each time the intervention is introduced, the behavior changes; each time it is withdrawn, the behavior returns. An alternative explanation, such as the student maturing or the class settling down over the term, would have to rise and fall in step with the condition changes, which is unlikely. Demonstrating the effect, removing it and demonstrating it again within one subject is what establishes control. Kazdin (2011) described this as the core logic of single-case designs: prediction of what baseline would have looked like, verification by returning to it and replication by reintroducing the intervention.
The Multiple Baseline
Some behaviors cannot be reversed, and for them the reversal design is simply the wrong tool. A skill once learned, such as reading a set of words, does not disappear when teaching stops, and it may be unethical to withdraw an intervention that reduced dangerous behavior. For these cases, the multiple baseline design introduces the intervention at different times across several baselines: different behaviors in one person, the same behavior in different settings or different people. If each behavior changes only when the intervention reaches it, and not before, the staggered timing rules out explanations that would affect all baselines at once. The design gives up the within-condition replication of a reversal, but it keeps the essential logic: each new baseline is a fresh prediction that the behavior will change only when the intervention arrives, and each confirmed prediction adds to the case for control.
The Design's Costs
The design's costs should be stated honestly, because its advocates have sometimes understated them. A reversal requires withdrawing an effective intervention, which may be unacceptable when the target behavior is harmful. Effects must be large and stable enough to see in repeated measurements of one subject; small effects may be missed. And conclusions often rest on visual analysis of graphed data rather than statistical tests. DeProspero and Cohen (1979) asked experienced reviewers to judge the same graphs and found that their agreement about whether an effect had occurred was low, particularly when effects were moderate or the data variable. Later work has proposed structured criteria for visual analysis and statistical supplements, but the reliability of judging graphs by eye remains a known weakness.
The Sampling Objection
The sampling objection holds that findings from one student, or a few pigeons, cannot show anything about people or pigeons in general. A group study with a random sample, the objection runs, is needed to generalize.
The answer is that single-case research generalizes by a different route. Sidman (1960) distinguished direct replication, repeating the same procedure with the same or similar subjects to show reliability, from systematic replication, varying the subjects, settings, behaviors or procedural details to show how widely the effect holds. A token procedure that reduces out-of-seat behavior in one student, then in other students, in other classrooms and with other behaviors, builds generality one replication at a time. Each replication that succeeds widens the range of conditions under which the finding holds; each that fails marks a boundary.
That route supports a specific kind of claim: that a procedure reliably produces an effect in individuals of a given kind, under conditions of a given range. It does not support claims about the average effect in a defined population, which is what a group study with random sampling estimates. Policy decisions that depend on population averages, such as how much a program will reduce a problem across a school district, need group data. Decisions about whether a procedure will work for a particular individual, and how to adjust it, are often better served by single-case evidence.
References
DeProspero, A., & Cohen, S. (1979). Inconsistent visual analyses of intrasubject data. Journal of Applied Behavior Analysis, 12(4), 573-579. https://doi.org/10.1901/jaba.1979.12-573
Kazdin, A. E. (2011). Single-case research designs: Methods for clinical and applied settings (2nd ed.). Oxford University Press.
Sidman, M. (1960). Tactics of scientific research: Evaluating experimental data in psychology. Basic Books.
PSY 803 Topic 3 instructions, in plain terms
Topic 3 in PSY-803 asks doctoral students to evaluate single-subject research design. Strong evaluations explain why the experimental analysis of behavior preferred repeated measurement of a few subjects to group averages, work one design through with data showing how control is demonstrated within a subject, and present the multiple baseline for behavior that cannot be reversed. Students should state the design's costs honestly, including disputes about visual analysis, and give the sampling objection the final section, answering it with direct and systematic replication while saying plainly what kind of claim that route supports. A composite data set is acceptable for illustration, provided the paper labels it as constructed and describes the measurement exactly.
Inside the PSY 803 Topic 3 example
The finished evaluation reads as a methodological argument with one design at its center. It explains why the experimental analysis of behavior preferred repeated measurement of a few organisms to averages across many, citing Murray Sidman's account of replication as the source of confidence. A reversal design is worked through with a composite data description, showing how behavior tracking the introduction and removal of a condition demonstrates control within one subject. The multiple baseline appears as the answer for behavior that cannot be reversed. The paper then states the design's costs honestly, including disputes about the reliability of visual analysis. The sampling objection receives the final section, answered with direct and systematic replication as an alternative route to generality, and with a plain statement of what that route demands.
PSY 803 Topic 3 rubric: what Excellent looks like
Faculty reward a design argued, not described. An Excellent evaluation draws on Sidman's case for replication within and across subjects. Credit goes to a composite ABAB table in which out-of-seat behavior tracks the token condition, read through Kazdin's logic of prediction, verification and replication. The rubric values the multiple baseline for irreversible behavior and honest costs, including DeProspero and Cohen on inconsistent visual analysis. Strong evaluations answer the sampling objection with systematic replication. Top marks go to separating individual from population claims.
PSY 803 Topic 3 help before you submit
Dismissal is the typical way a paper loses here: single-subject research called unscientific because the samples are small, without explaining what the designs are built to establish. That verdict treats group statistics as the only route to a causal claim. The reverse error presents these designs as flawless and ignores genuine disputes about visual analysis. Papers describing a reversal without explaining why each phase transition matters have copied a diagram and missed its logic. Treating the experimental analysis of behavior as a mere collection of animal studies misses that its designs were built deliberately to answer a question about individuals. Evaluations that never take up the generality objection leave the strongest criticism unanswered, and doctoral criteria expect it to be met directly.
Get a PSY-803 Topic 3 example written to your instructions
Send us the PSY-803 Topic 3 instructions, the rubric in your classroom and any studies or designs your section named. We write a custom example to those criteria, with the design explained from its own rationale, a phase-by-phase worked case and the generality objection answered, returned in 24 to 48 hours. The first one is free.
More PSY 803 papers and PhD in Psychology samples
- PSY 803 Topic 1: Evidential Commitment Statement
- PSY 803 Topic 2: Respondent and Operant Analysis
- PSY 803 Topic 4: Radical and Methodological Comparison
- PSY 803 Topic 5: Program Boundary Dq Post
- PSY 803 Topic 6: Applied Outcomes Review
- PSY 803 Topic 7: Verbal Behavior Critique Reconstruction
- PSY 803 Topic 8: Behaviorist Legacy Assessment
- PSY 804 Topic 3: Ultimate Concerns Analysis
- PSY 802 Topic 3: Clinical Evidence Appraisal
- PSY 810 Topic 3: Problem Reconstruction Paper
- PSY 839 Topic 3: Goal Conditions Paper
PSY 803 Topic 3 questions, answered
How can one subject support a general claim?
Through replication rather than sampling. If an effect appears, disappears and reappears with the condition in one subject, and then does so again in further subjects and settings, confidence in its generality grows with each repetition. That route asks more of a research program over time, but it avoids reporting an average that may describe no individual in the sample.
Is visual analysis reliable?
It is contested, and saying so strengthens the paper. Large, immediate and consistent changes are read the same way by most analysts, while small or variable effects produce disagreement. Researchers have proposed statistical supplements for those cases. A strong evaluation acknowledges the dispute and notes that the design was built around effects large enough to see without help.
Why was the program so reliant on animal research?
Because control over the environment was essential to the method, and laboratory animals allowed conditions to be held steady for long periods. The designs themselves are not tied to animals, and applied researchers later used the same logic with people in schools, clinics and workplaces. The animal work was a choice made for control, not a limit of the method.
Where can I find a free PSY 803 Topic 3 sample paper?
The design evaluation is shown here complete: why few subjects, a reversal design worked through, the multiple baseline, the design's costs and the sampling objection. Send your PSY 803 Topic 3 prompt and a evaluation follows free.
How does a reversal design show control in PSY 803 Topic 3?
The behavior changes when the intervention starts, returns when it is withdrawn and changes again when it is reinstated. An outside cause would have to rise and fall in step with every switch.