Introduction
To start with, let me introduce the concept of a single system research design. According to Miller (n.d.), it “involves studying a single individual or system by taking repeated measurements of 1 or more dependent variables and systematically applying & sometimes, withdrawing or varying the independent variable” (p. 5). So, it is an assessment of the intervention’s efficiency made with the help of regular measurements. Single system research design can act as a relevant appliance towards the methodological arsenal of the business analyst.
Single system research designs are considered to be decreasing from the order of composure, differing from basic designs; it is used to classify the incidence of adjustment. The entity of single-system research designs is referred to as a system. The system can be represented as an applicant, a family, two spouses, or a minuscule team in clinical practice. In program evaluation, however, a whole company, a program, a department, or a chain of departments, even separate states can play a role in a system.
The most feasible research design for this intervention
As Vaus (2005) claims, “it is important to be clear about the role and purpose of research design” from the very beginning (p. 1). So, first of all, it is necessary to decide what behavior has to be changed, and what goals have to be achieved (in other words, we need to choose dependent variables). After that, the kinds of interventions have to be determined, and steps to evaluate the effectiveness of the process should be defined.
In our case, the primary goal of the intervention is to help the women from a group workshop to develop parenting and self-efficacy skills. I also want to find out whether individual sessions two times a week would be more efficient than group training. For this purpose, different activities, such as conversations, role-playing, skill practice, and others will be applied.
To evaluate the intervention, I am going to conduct regular tests (self-checklists, questionnaires, etc.) regarding a client’s parenting and self-efficacy skills and monitor any progress or regress. It is also useful to present the results as a graph (Royse, Thyer, & Padgett, 2010, p. 142). In such a case, it is easier to see whether a client is showing any improvement.
My reasoning for choosing the design
I believe that this kind of a single system research design will achieve success as it can provide regular assessments of the client’s progress. As Nestor and Schutt (2014) state, I will have “the repeated measurement of a dependent variable” where “a dependent variable” is parenting and self-efficacy skills (p. 231). That, in its turn, will help to adjust to the results. Firstly, I am not going to implement all kinds of activities at a time.
In that case, it will be easier for me to identify which one of them has become the cause of the changes in the client’s behavior. Considering indicators of progress, I will change a set of activities, adding or withdrawing some of them. That is how we can achieve better results. Besides the fact that this single system research design is efficient, it is also very easy to implement and will not take a lot of time.
Discussion of two practical or ethical constraints in my evaluation
Surely, my research design does have some constraints. First of all, even with the best physiological tests, it is hard to identify how good a person is developing certain skills. That is because although we call desired skills “a dependent variable,” this variable is not quantitative (Nestor & Schutt, 2014, p. 231).
The dependent variable itself can cause complications because of its limited dimension. According to Nestor and Schutt (2014), “in experimental design, you want research participants to show a sufficient range of scores on a dependent variable. The greater the range in scores a dependent variable shows, the greater the likelihood that these changes result from changes in the independent variable” (p. 189). The second constraint is a human factor. I will never know for sure that my client answers honestly, that she is in a good mood for training and taking tests, and so on.
Discussion of two threats to validity and the ways I can address these in my evaluation design
Ohlund and Yu (n.d.) determine several threats to validity, which they divide into two large groups: threats to internal validity and threats to external validity. According to Ohlund and Yu (n.d.), Internal validity “refers specifically to whether an experimental treatment/condition makes a difference or not, and whether there is sufficient evidence to support the claim” (para. 7). External validity depicts the result of treatments in general. Let us focus on two different threats, one from each group.
One of the main factors that jeopardize internal validity is statistical regression. This factor is created by the draft of material on the ground of acute accounts or characteristics. If a client sees that his or her statistics worsen a client either gives up on training at all or somehow improves future results of the purpose test. I can avoid this kind of outcome by not telling my client the test results if I know that this will upset her. One of the threats to external validity is multiple interferences. Ohlund and Yu (n.d.) claim that “as multiple treatments are given to the same subjects, it is difficult to control for the effects of prior treatments” (para. 13) In such a case, it is very difficult to control all of them. To prevent this, I just will not take too many tests.
A measurement tool that I believe would yield the best data for this type of research. Explanation of why I think this tool will be helpful
To yield the data for this type of research I would choose a method suggested by Royse et.al. (2010), and it is a graphic representation “with time on the horizontal axis, and the outcome measure on the vertical axis” (p. 142).
I am going to take the tests with the same maximum number of points and mark points that my client will earn on the same graph. I will disconnect points by using a marked vertical axis; then, every point will be marked with an understandable heading. In a case when the vertical line is not marked at all, I would heighten it a little to distinguish it from the horizontal line. So, the outcome measure here is the number of earned points. Then I will connect the points with the lines, and that will give me a complete picture of my client’s improvement. I believe that this measurement tool will be helpful since it presents information in an understandable form. Besides, it is easy to use.
References
Miller, B. (n.d.). Single Subject Research Design. Web.
Nestor, P. G., & Schutt, R. K. (2014). Research Methods in Psychology: Investigating Human Behavior (2nd ed.). Thousand Oaks, CA: SAGE Publications.
Ohlund, B., & Yu, C. (n.d.). Threats to validity of Research Design.
Royse, D., Thyer, B., & Padgett, D. (2010). Program Evaluation: An Introduction (5th ed.). Belmont, CA: Cengage Learning.
Vaus, D. (2005). Research Design in Social Research. London, Great Britain: SAGE Publications.