Research Design Methods

The most feasible research design for Maritza's intervention is the A-B-A-B research design. The A-B-A-B design is an evaluation across settings and behaviors. It is very useful in evaluating interventions. In the A-B-A-B design there are two phases that alternate over time. The design measures a baseline (A), a treatment measurement (B), the withdrawal of treatment (A), and the re-introduction of treatment (B). The effects of the intervention are clear if performance improves during the first intervention phase, reverts to or approaches original baseline levels of performance when the intervention is withdrawn, improves when the intervention is reinstated in the second intervention phase, and again changes in the final intervention phase.

The A-B-A-B design has two parts; gathering of baseline information, and the application of treatment. It also measures the effects of treatment and the measurement of what happens when treatment is changed or removed and then applying the treatment and measuring the change. This treatment allows overlapping treatments to be measured on a single subject (Education.com, 2013).

There are also A-B-A-B designs that are available that include more than one treatment and more than four phases or that end with a new phase in which procedures are included to maintain the gains (Education.com, 2013). For example, suppose that the treatment (B1) does not change behavior after the baseline phase. The investigator would not continue the phase but would try another treatment (B2). This latter treatment would constitute a new phase and would probably be implemented later in the design. The design could be represented as an AB1 B2 AB2 design (Education.com, 2013).

I chose this research design because when changes in behavior occur during the intervention phase, it shows evidence that the changes in behavior is because of the intervention. Also with this design performance can be compared in different phases and Maritza can examine if the intervention will have an effect or not. The A-B-A-A design also suggests experimental control which allows for a functional analysis.

The ethical restraints in the evaluation are obtaining informed consent of the person that is participating. Also the unique conditions that are involved in the evaluation process must be discussed with any potential participants. All aspects of research and evaluation including the purpose, measurement, data collection and confidentiality are all information that must be included for informed consent.

Even when there are instances where informed consent is not required when using tools for practice evaluation practitioners should still practice ethics. According to the NASW Code of Ethics, practitioners should, as part of their everyday practice with clients, provide services to clients only in the context of a professional relationship based, when appropriate, on valid informed consent (NASW, 2012). Practitioners should use clear and understandable language to inform clients of the purpose of service, risks related to services, limits to services because of a third party payer, clients right to refuse or withdraw consent and the time frame covered by the consent (NASW,2012).

There can be many threats to validity concerning a program or evaluation of a program. One threat is the loss of data or participants. If there is not enough information collected about the participants or if too many participants drop out the results will be based on not enough data to be reliable. Another threat is participant's reactions to being studied alters their behavior and the study results. I will address these issues by identifying the sources of evidence such as documents, observations, people involved as well as appropriate methods for obtaining reliable and valid data.

One of the measurement tools that would yield the best data for Maritz's program is a questionnaire. A questionnaire is a way of eliciting the feelings, beliefs, experiences, perceptions, or attitudes of a sample of individuals. For use in data collection a questionnaire can be structured or unstructured (Research Design