The Week 4 assignment calls for you to identify the research design you plan to use in your evaluation of the program reflected by the scenario you selected at the beginning of the course.

The chapter in the course text includes a discussion of longitudinal evaluation designs (trend studies, cohort studies, and panel studies). Longitudinal studies conduct observations over an extended period of time, usually measured in years. For example, children with a parent diagnosed with schizophrenia and those who do not might be followed and compared over many years to see if they become afflicted with schizophrenia. If the incidence of schizophrenia among the children with parents with schizophrenia is significantly higher than it is among the other children, then having a parent with schizophrenia would be deemed a risk factor for developing the disorder.

That being said, longitudinal evaluation studies are not appropriate for the outcome evaluation required for this course.

The instructions also ask if your evaluation design will include one or two groups. I posted a Word document titled Research Designs to be used as a supplement to the information contained in chapter 10 of the course text.

In that document, one-group or pre-experimental designs include those shown in the table below.
Where:
X = the intervention
Oi = one or more “observations” of the dependent or outcome variable, which could be an actual
observation by the researcher or a collateral, a self-report by the client, or the results of
some instrument (or all three)

Design Name Features Problems with Use
One-shot Case Study
( X O) • No randomization of subjects
• No pretest measure(s)
• No control or comparison
group • Threats to internal validity:
history, maturation,
selection bias, experimental
mortality
One-group Pretest-Posttest Design
(O1 X O2) • No randomization of subjects
• No control or comparison
group • Threats to internal validity:
history, selection bias,
maturation, testing, regression
to the mean
Simple Interrupted Time Series Design

(O1 O2 O3 … X O1 O2 O3 …) • No randomization of subjects
• Similar to One-group Pre-test-
Posttest Design with the
exception that multiple
measures of the dependent
variable are taken before and
after the experimental treatment
is administered or introduced • Must rule out history and
maturation as rival explanations
for results

You will also note in the document I posted that all two-group true experimental designs require both randomization of subjects and experimental and control groups. While these kinds of designs are required for studies such as those related to some experimental drug regimen or in the development of some evidence-based practice, they raise serious questions about the ethics of withholding the intervention for the control group.

Thus, the use of a true experimental design is also inappropriate for the outcome evaluation required for this course.

The only two-group evaluation designs that are appropriate for the outcome evaluation required for this course are shown in the following table.

Non-equivalent Comparison Group Design

(O1 X O2) experimental
(O1 O2) comparison • No randomization or matching
of subjects
• In the absence of
randomization the second
group is called a comparison
group and not a control group • Selection bias may be
operating given that there is no
randomization of subjects
• Rival explanations for findings
must be ruled out
• Given that no randomization of
subjects has occurred, there is
really no well to tell if the
experimental and comparison
groups are really equivalent

Complex Interrupted Times Series Design

(O1 O2 O3 … X O1 O2 O3 …) experimental
(O1 O2 O3 … O1 O2 O3 …) comparison • No randomization of subjects • Given that no randomization of
subjects has occurred, there is
really no well to tell if the
experimental and comparison
groups are really equivalent

You are to select one of the three one-group designs or one of the two two-group designs for your program evaluation study.

NOTES
• You will need to use the same design for each of the interventions (e.g., support group, therapy, stress management) represented by the program being evaluated.

• The dependent or y-variable is the desired outcome of the intervention. The independent or x-variable is the intervention being used.

• In all cases, you will be using a non-probability convenience sample since the subjects of your study will be those who walked through the door or were referred by another agency.

 

For a custom paper on the above topic or any other topic, place your order now!

What Awaits you:

• On-time delivery guarantee

• Masters and PhD-level writers

• Automatic plagiarism check

• 100% Privacy and Confidentiality

• High Quality custom-written papers