Traditional experimental research dictates that there must be one or more experimental groups (groups receiving a treatment or intervention) and a control group (which does not receive the intervention). Participants are assigned randomly to the experimental and control groups. Although experimental designs may be preferable to other designs, even in action research, they are not always feasible for practitioners for many reasons. There may be ethical issues regarding withholding an intervention from a client group; or time, financial, or logistical constraints may make it impossible to include random assignment into groups.
In such cases, the use of time-series research can provide a meaningful alternative to true experimental research. Time-series designs are quasi-experimental, meaning that they still seek to evaluate the impact of a treatment or intervention on a target group (or individual) but do not include random assignment to treatment and control groups. Researchers still have control over admittance criteria to the treatment group (e.g., mothers under the age of 21). Times series then allows researchers to take measurements of treatment effectiveness at various intervals over time. Taken together, these data points provide information about the intervention under consideration.
In this Assignment, you apply time-series research techniques to a case study.
Imagine that you are a human and social services professional working with a child, Kaya, who is having some behavioral problems at her elementary school. As a scholar practitioner, you approach your work with her from an empirical standpoint and want to measure the effectiveness of your recommended services. To date, you have provided information about resources to the family and coordinated services for the child and family.
As such, you pose the following research question: “What impact do the services the child is currently receiving have on reducing off-task behavior, physical aggression, and verbal aggression for the identified child?”
For the purposes of this research, the target behaviors are as follows:
- Off-task: failure to maintain eye contact with task at hand for more than 3 consecutive seconds
- Physical Aggression: hitting/pushing/kicking peers/teachers, throwing objects
- Verbal Aggression: threatening, yelling
Consider the following observation report:
During the first observation period, which was conducted on Monday (in math class) between 11:20 and 11:30, 10-minute, 10-second, partial interval recording was used to assess the frequency of off-task behaviors, physical aggression, and verbal aggression. Kaya was observed to be off-task during 12% of the intervals observed, engaged in physical aggression 0% of the intervals observed, and engaged in verbal aggression 2% of the intervals observed.
During the second observation (during recess), which was conducted on Wednesday from 9:00 to 9:30, Kaya was observed to be off-task during 15% of the intervals observed, engaged in physical aggression 20% of the intervals observed, and engaged in verbal aggression 3% of the intervals observed.
Kaya was also observed on Friday from 10:30 to 10:40 during a group activity. Kaya was observed to be off-task during 20% of the intervals observed, was observed to engage in physical aggression 7% of the intervals observed, and was observed to engage in verbal aggression 10% of the intervals observed.
Off-task Physical Aggression Verbal Aggression
12% 0% 2%
15% 20% 3%
20% 7% 10%
The Assignment (1–2 pages)
Based on the data, you determine that you must design a behavioral intervention for Kaya and determine if it is successful. As you begin the process of investigating your research question:
- Briefly describe a time-series design that you would use to investigate the impact of your intervention, and explain how you would collect and record the additional data. Be specific.
- Select a method for graphically representing the provided data. Use the method to create a graph. (You can use Microsoft Excel or Word).
- Analyze the data and provide a narrative description of your analysis.