The following paper presents the use of an ABA single subject research design in evaluating a 16 year old foster youth’s behavior, and the use of an intervention on decreasing negative behavior and increasing positive behavior. Use of the ABA design, factors to be evaluated, treatment variables, implementation of the research design, data gathering, and data analysis are all discussed. Limitations of single subject research designs are also presented. Finally, graphed data is attached in the appendix. Single Subject Research Design
Factors to be Evaluated Halle is a 16-year-old female who currently resides in foster care. Over the past few years, she has struggled with a number of behavioral concerns including cutting, verbal aggression and physical aggression. Since Halle was placed in foster care, her verbal and physical aggression has decreased and seems to be provoked by contact with her biological mother. Although these behaviors have decreased, Halle continues to cut herself with various objects on a regular basis as a way to cope with stress.
Using an ABA single subject research design, the researcher will evaluate the effectiveness of weekly individual outpatient counseling on Halle’s cutting. The foster mother was able to document the number of times Halle cut over a four week period while she was on the waiting list for counseling. This period provided baseline data on Halle’s behavior, and documented that Halle cut herself 43 times in four weeks. Halle began the counseling program after the four-week waiting period, and will meet with the counselor weekly for 10 weeks.
The ABA design was chosen to evaluate the effectiveness of the current intervention for three main reasons. First, the foster mother was able to collect baseline data. Second, there is one planned intervention (individual counseling) and one main problem being treated (cutting). Third, the intervention is time limited, there are no additional planned interventions at this time, and the foster mother can continue tracking the cutting during and after the intervention (Bloom, Fischer & Orme, 2006).
The treatment variables measured include cutting and use of an alternative method to cope with anxiety. First, Halle will eliminate cutting as a way to cope with her anxiety by the end of the counseling program. Second, Halle will increase the number of times she goes for a walk when she is angry to vent her feelings. The frequency of these behaviors would be measured by the foster mother recording when Halle cuts, how many cuts occur, and how many times Halle goes for a walk when she becomes angry. Implementation of the Single Subject Design
It is important to hear from Halle about her current situation and behaviors, and to involve her in the decision regarding what intervention will be implemented and evaluated (Bloom, Fischer & Orme, 2006). Because the behavior identified was cutting, and Halle has a tendency to hide this from others, the foster mother was identified as the data collector. At 10pm each night, the foster mother would check Halle for cuts and tally the number of cuts on a daily calendar. After four weeks, Halle began her counseling, which started the treatment phase, and the foster mother continued to use the same method to collect the data.
The target of the individual counseling sessions is for Halle to develop an understanding of her cutting behavior and aggressive reactions to anger. In addition, Halle will learn techniques to constructively cope with her feelings. The goal is for Halle to decrease her aggression and eliminate her cutting by replacing these behaviors with going for a walk to vent her feelings. Once the counseling began, the foster mother also began tallying the number of times Halle became angry and the number of times when angry that Halle went for a walk.
For Halle, angry is defined as yelling, throwing objects and/or physical aggression. In addition to the foster mother tracking the frequency of Halle’s behavior, Halle also completed the Adolescent Anger Rating Scale (AARS) at each weeks counseling session, which provided an additional form of data to evaluate Halle’s progress in counseling (Burney, 2001). After the 10 week counseling program was completed, the foster mother continued to collect data on cutting, episodes of anger and going for a walk when angry for four weeks.
This served as the follow-up or second A period. Data Analysis If behavior change occurred, the frequency of Halle’s cutting would be highest during the baseline phase, should decrease in the intervention phase, and the decrease in the behavior should continue or be maintained throughout the follow-up phase (Bloom, Fischer & Orme, 2006). The frequency of the walks should increase in the intervention phase and increase or be maintained during the follow-up phase, if the intervention was successful regarding this treatment variable (Bloom, Fischer & Orme).
To analyze the data, first enter the data into an Excel spreadsheet. Baseline followed by treatment data would be entered in one column on the spreadsheet in order by week. Using the Excel functions, calculate the mean and standard deviation of the baseline. For cutting, the mean was 10. 75, and the standard deviation 1. 5. For walking, the mean was . 75, and the standard deviation . 95. Next, calculate the upper boundary by adding the mean and standard deviation, and the lower boundary by subtracting the standard deviation from the mean. For cutting, the upper boundary was 12. 5 and the lower boundary was 9. 25. For walking, the upper boundary was 1. 71 and the lower boundary was -. 21. Next, a line graph for cutting and for walking would be generated in Excel (see Appendix A). By conducting a visual of the data, the social worker can determine if it appears that behavior change occurred; in other words, did Halle’s negative behavior decrease and positive behavior increase during the intervention phase? It is clear that Halle’s walking did increase and cutting decreased, both target behaviors of the intervention.
To determine if the intervention is statistically significant, the social worker would refer to the binomial table to determine the number of observations during the intervention phase versus the baseline phase that are needed to determine if the intervention was significant. In other words, how many days must the number of cuts fall below the lower boundary, and the number of walks fall above the upper boundary during the intervention phase. The intervention was significant at the . 05 level, based on this analysis that shows that Halle’s cutting falls below the lower boundary 12 out of 14 weeks, and her walking falls above the pper boundary 13 out of 14 weeks during the intervention. Additionally, in the follow-up phase, Halle’s behavior changes appear to be maintained. Limitations of a Single Subject Research Design (SSRD) A major limitation of SSRD’s is the possible lack of external validity. It may be difficult to generalize the findings of one SSRD to the general population and replication of the SSRD would be needed to do so (Jordan & Franklin, 2005). Another limitation is the ability of SSRD’s to determine cause and effect of interventions on client problems.
Certain types of SSRD’s, such as AB designs, cannot prove with relative certainty that an intervention caused a change in behavior. Other types of designs, such as ABA can say this with relative certainty; however, in order to do so other variables must be ruled out (Class PPT Ch 11, 2009). It may be challenging to identify and rule out other variables, as clients and their situations are often changing. Additional limitations of SSRD’s include: social workers may have limited time to conduct evaluations or limited contact with the client.
Social workers may work as a team to serve clients, in which their contact with the client varies, or work with clients in a group setting (Bloom, Fischer & Orme, 2006). Both situations may make conducting SSRD’s difficult. In addition, client mortality and social worker turnover may impede evaluations using SSRD’s. A final limitation with any type of evaluation and research involving clients is that problems change, get worse or new problems develop. These factors can complicate SSRD’s and social workers will need to be flexible, adaptive and consider the effect of these changes on interventions and outcomes.
References
Bloom, M., Fischer, J. & Orme, J. G. (2006). Evaluating Practice: Guidelines for the Accountable Professional. New York: Pearson. Burney, D. M. (2001). Initial development and score validation of the adolescent anger rating scale. Educational and Psychological Measurement, 61. Retrieved October 25, 2009 from the Academic Search Premier database. Class Powerpoint, Chapter 11. (2009). Retrieved from Blackboard October 25, 2009. Jordan, C. & Franklin, C. (2005). Clinical Assessment for Social Workers: Quantitative and Qualitative Methods. Chicago, IL: Lyceum Books.
