Third in a series of short articles describing the ideas and methods of research in clear, understandable language. Each article will address a segment of research in a way that clinicians, prevention specialists, supervisors, and administrators will find useful and friendly. Collectively, the articles will provide a "refresher course" that can be referred to as needed.

In the last column, we learned that the best way to obtain really good, scientifically based results was to use the true experiment. This time we will work our way to a less precise method of obtaining decent results.

To satisfy cost, ethical, and practical concerns of treatment programs and still obtain solid outcomes, sometimes we must make some sacrifices in terms of validity. The vehicle for this is the "quasi-experimental" approach-the nonequivalent control group design or simply the comparison groups design experiment (Posavac & Carey, 1997; Heppner, Kivlinghan, & Wampold, 1998; National Institute on Drug Abuse, 1993).

Why is it called quasi, which does sound a little strange, like something out of a science fiction movie? Think of quasi as almost like the true experiment, or a pseudo-experiment. The reason for the "almost" is that we do not use an equivalent and/or a randomly determined control group. We still conduct our experiment in the same way, but the control group isn't as controlled.

Here are the basic steps in conducting a quasi-experiment:

• Form a comparison group and an experimental group.
• Do something to the experimental group (for example, give them a new treatment), but do not do it to the comparison group.
• Analyze the results to see if there is any difference between the groups.

This is essentially the same as the design of the true experiment described in the previous installment in this series.

When we resort to this form of experimentation, we open ourselves to the possibility of other explanations for our results (Posavac & Carey, 1997). (These types of alternative explanations result from "internal validity threats." More on those dilemmas in an upcoming column.) For now, we can still get pretty good results using the quasi.

Recall that with the true experiment, the process of randomization makes fairly sure that there will be no differences between groups before the experiment begins. This eliminates a lot of alternative explanations of which cause determined which effect, because the groups that start down the road of true experimentation are considered to be as similar as you can get.

For example, if you give a certain form of addiction treatment to one randomly determined group and not to another equally matched group, you are trying to ascertain if there will be any difference between the groups. If there is, you are onto something big. In this case, there can be very little argument at the end of an experiment as to what caused what.

It is important not to make claims about the effectiveness of treatment interventions without doing some sort of experiment. The best bet for making effectiveness claims is to conduct either the true or the quasi experiment. Watch for it in your reading.

Putting the Best Face on the Quasi

For the less stringent quasi-experiment, there are a number of ways to get some level of parity with the almighty true experiment. One way is called the interrupted time series with switching replications (Posavac & Carey, 1997). It sounds intimidating, but it isn't.

Let's say you have developed a new alcoholism treatment, and you want to see if it works. Using the quasi application, you would use a slightly refined version of the three-step process described above:

• Form two groups of patients, the experimental group and the comparison group.
• Pre-test both groups with an instrument that would provide a baseline for the groups, such as the Alcohol Dependence Scale.
• Apply the treatment to the experimental group and withhold it from the comparison group.
• Measure the experimental group many times (e.g., every two weeks) to see if it responded to your treatment. If it did, you would apply the same treatment to the control group, measure it many times, and see if you got the same results. If you did, you can assure yourself that this new treatment of yours is promising.

So the quasi uses basically same procedure as the true experiment, but without an equivalent control group (which spells trouble for the way we can interpret any results using this method). Instead, it uses a comparison group-a group that is as close as we can get to the random equivalent group, but isn't truly random.

Despite this drawback, we can modify the quasi and still get pretty good results. Hopefully, this information will encourage many of you to design a way to test a favored treatment approach using a quasi or true experiment.

Next: The least reliable forms of evaluation-the non-experimental post-test and the pre-test, post-test designs.

Michael J. Taleff, PhD, CAC, MAC, is assistant professor in the Counselor Education Department at Pennsylvania State University. He is also a member of the NAADAC Research Committee and welcomes comments on this series. His e-mail address is This e-mail address is being protected from spam bots, you need JavaScript enabled to view it

References
National Institute on Drug Abuse (1993). A guide to evaluation: How good is your drug abuse treatment program? (NIH Publication No. 93-3609). Washington, DC: U.S. Government Printing Office.

Heppner, P.P., Kivlinghan, D.M., & Wampold, B.E. (1998). Research design in counseling (2nd ed.). Pacific Grove, CA: Brooks/Cole.

Posavac, E.J., & Carey, R.G. (1997). Program evaluation: Methods and case studies (5th ed.). Belmont, CA: Prentice Hall.

---

Comments

Add New Search RSS

Write comment
Name:
Email:	do not notifynotify
Title:
UBBCode:	-color-aquablackbluefuchsiagraygreenlimemaroonnavyolivepurpleredsilvertealwhiteyellow -size-tinysmallmediumlargehuge
	Please input the anti-spam code that you can read in the image.

Powered by !JoomlaComment 3.25

3.25 Copyright (C) 2007 Alain Georgette / Copyright (C) 2006 Frantisek Hliva. All rights reserved."