Evaluator's Toolbox: Quasi-Experimental Designs.
Quasi-experimental designs are excellent research designs to use in evaluative research such as program evaluation. The director of training at GlobalEd would like to complete a study using a quasi-experimental design and needs your help. His hypothesis is as follows: Multicultural training will increase instructor's understanding of how culture can contribute to an appearance of a behavior problem. Your Acme Corporation Behavioral Health supervisor thought that a pretest/post test quasi-experimental design would be ideal to test his hypothesis. In this discussion, we will talk about the process the training director should use to complete his study.
What are some of the steps for completing this quasi-experimental study?
When responding to other Acme Corporation Behavioral Health interns comment on the order of the steps as well as if you agree or disagree. By the end of the discussion, there should be a clear picture of what the director should do, step by step.© BrainMass Inc. brainmass.com October 25, 2018, 7:18 am ad1c9bdddf
Thanks for using BMass! Please see the attached. Hope this helps. We hope to see you soon. Good luck in your studies. Take care!
Quasi Experimental Designs
A quasi-experimental study is a type of evaluation which tries to determine whether a course, program or research intervention has had the original and the intended effect /outcome envisioned on the participants in the study. Such studies can be in different styles or forms, and are generally regarded as not having some major or key aspects of a true experiment, thus the term Quasi-experimental. For example, 'true experiment' will usually include: a pre-posttest design, a treatment group as well as a control group, and in addition, a randomly assigned group of participants.
Quasi-experimental studies; however, is usually missing or lacking one or more of those elements. So to reiterate, the two forms of quasi-experimental studies can be either: a pre-posttest design study without a control group or, a pre-posttest design with a control group.
That said, the most common type of quasi-experimental studies usually includes a pre-posttest design with two groups: the treatment group and the control group. Care must always be taken to ensure that the treatment and control group do not differ at the start of the experiment, (for example, they all have never taken or been exposed to the cultural course content in any way /time prior to the intervention-so one group will not have an 'extra edge' or advantage over the other). Researchers conducting such studies (QE) will usually do a number of things to prevent this. Such methods include but are not limited to: matching the treatment groups to similar control groups; and they also control for any differences in their final data analyses.
The purposes of your research then would be: 1) to determine whether multicultural training will increase instructor's understanding of how culture can contribute to behavioral problems; 2) the effectiveness of the training in instructors' understanding of cultures and contributory factors to behaviors (if any); 3) to identify the factors that influence cultural awareness, personal factors; and behaviors and; 4) to determine, upon analysis of the data, what aspect(s) of the training instructors reported as being attributed to perceptions towards cultural impressions and impact on behaviors before and after the training.
Random selection of employees or (students/employees not sure) at GlobalEd
Pre-test, post-test control group quasi-experimental ...
Quasi Experiment Design: Non Equivalent Groups
Assume that I wanted to test some treatment on A university Program Evaluation students in the School of Public Policy and Management. I decided to give this treatment to groups 1 and 2, and use groups 3 and 4 as a control/comparison group. Remember that you were divided into groups on the basis of your last name: alphabetically the first student went into group one, the second into group two and so on with the fifth student going into group one.
(1) Is this an experimental or quasi-experimental design? Why?
(2) What possible differences might there be between the two groups? (Hint: remember how I originally divided you into groups)
(3) What threats to internal validity would not be controlled by this design?
(4) Do you consider this a good design? Why?
Suppose I gave then decided to evaluate how much students learn about program evaluation in PPM 804N.
(5) What would be a reasonable comparison group? Why?
(6) Delineate your design using X's and O's, and label it with the appropriate name.
(7) Would you use a pre- and post- test? If so, generally define what form that test(s) will take (2-3 sentences - I don't need exact questions!) Will you use the same test pre and post? Why?View Full Posting Details