7.99 Epilogue: Analyzing Your Data Using Therapy-Science

Give a boy a hammer and everything he meets has to be pounded.
Abraham Kaplan (in Horowitz, 1962)

You now have most of the tools that you need to structure and analyze any intervention for your client. You have a vocabulary for describing the level, trend, and variability within and between phases in a single-case design, as well as quantitative tools for evaluating these measures. You’ve also been provided with a suite of powerful visual, quantitative, and statistical tools so that you can to validate your assumptions about whether your client’s performance is changing over the course of the intervention. The last section of this chapter deals with perspectives on accessing clinical change. This should provide you with a set of heuristics for figuring out some the critical decisions that you need to make in developing and implementing your rehabilitation program.

Kaplan’s quote above (Law of the Instrument) warns us that just because you now have a new set of robust methods to visualize and analyze your client’s performance, and that you have some training in how to use them, you have to be careful to use them appropriately. It’s easy to “pound” everything with a favorite technique – or software program. Throughout the text, I have advocated for using a critical thinking perspective when researching, developing, and conducting your intervention program. The same holds true for analyzing your data. You have an ethical responsibility to make the best decisions you can, based on the data, for the benefit of your client. As you go about your data analysis, please pay particular attention to Section 7.3 and pick the right hammer for the specific nail you want to pound.

Thought Questions and Exercises

  1. Use one of your clients with which you are currently doing therapy for the following case study. Take the two approaches discussed in 7.3 and apply them to your client’s case, as if you could start all over again. What would you do differently this time? What prevented you from implementing these changes when you started out with your client? Can you make the changes now? Why or why not?
  2. Using the same client, take the first three data points that you have collected from your client and subject them to the First Few Data Points topic in Section 7.3.
  3. In a research project conducted by a doctoral student at UB, parents were taught a set of strategies to use with their children-with-autism to enhance their literacy development. Below, you will find a data set representing parental strategy use during a baseline (Phase A) and intervention (Phase B). Your task is to analyze the data in the graph using visual, descriptive, and statistical analysis techniques contained in the single-case graph software.
  4. Here is the data. Use the Therapy-Science graph program to analyze these data and answer the questions below the graph. You can use the Notes tool to write down your responses, However, your data will not be saved if you update or close the page.

    • Use your visual analysis skills to describe each phase, then compare the baseline to the intervention. Has a demonstrable change occurred? What is your reasoning?
    • For each of the graphs, use tools provided by Therapy-Science to describe the relationship between baseline and trend regarding level, trend, and stability. With your visually enhanced approach, does your conclusion differ from using your visual skills alone?
    • Exclude the Session 3 data point – the most extreme baseline score by ghosting it. Now compare this new ghosted baseline phase to the same baseline without ghosting it. Has ghosting the data influenced your thinking about the impact the intervention on your client’s performance?
  5. Now use the three statistical tests – PND, CDC, and Tau-U to test for statistical differences between the baseline and treatment phases of the study. Perform each test, first with the Session 3 data point un-ghosted, then with it ghosted.
    • Provide a one to two sentence analysis describing the results of each of the types of tests. Make sure you know the specific assumptions that each test makes about how it evaluates the data.
    • Were there differences between the PND, CDC, and Tau-U? Were there differences between the ghosted and un-ghosted data? How do you account for these differences?
    • In selecting among the tests, you probably want to pick the one that matches your clinic question the best. What test was it? Why did it match your clinic question best?

References

Barlow, D. H., Hersen, M., & Nock, M. K. (2009). single-case experimental designs (3rd ed.). New York: Pearson.

Brossart, D. F., Vannest, K. J., Davis, J. L., & Patience, M. A. (2014). Incorporating nonoverlap indices with visual analysis for quantifying intervention effectiveness in single-case experimental designs. Neuropsychological Rehabilitation, 24 (3/4), 464–491.

Gast, D., & Ledford, J. (Eds.). (2014). single-case Research Methodology Applications in Special Education and Behavioral Sciences, 2nd Edition. Routledge.

Hayes, S. C., Barlow, D. H., & Nelson-Gray, R. O. (1999). The scientist practitioner: Research and accountability in the age of managed care (2nd ed.). Needham Heights, MA, US: Allyn & Bacon.

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.

Lenz, A. S. (2012). Calculating effect size in single-case research: A comparison of nonoverlap methods. Measurement and Evaluation in Counseling and Development, 46.1 , 64-73.

Parker, R. I., & Vannest, K. (2009). An improved effect size for single-case research: Nonoverlap of all pairs. Behavior Therapy, 40(4), 357–367.

Scruggs, T. E., & Mastropieri, M. A. (2013). PND at 25 Past, Present, and Future Trends in Summarizing Single-Subject Research. Remedial and Special Education, 34(1), 9–19.