Sixth 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.

We are getting close to wrapping up the research-design portion of this series before we move on to other things. This article will cover a residue of research that didn't fit into the previous columns. This series is an effort to review key points and make what appears to be unintelligible — easy to understand. If you want to push deeper into the realms of research, there are scores of good books available. (See References following each installment in this series.)

More of the Fundamentals

All research falls into two very broad categories: basic and applied. The basic type tries to find general laws that influence certain behaviors. This type of research is difficult to apply to the real world (Bordens and Abbott, 1996). An example might be in the field of biology. Here the researcher may be trying to establish what human neurotransmitters respond to heroin. That information in and of itself has little use to the substance use disorder (SUD) counselor working in the trenches.

Applied research, on the other hand, tries to investigate problems that have some practical value (Bordens and Abbott, 1996). For instance, after establishing which neurotransmitters respond to heroin, a team of researchers may want to find a way to neutralize heroin by administering a drug to block its effects.

Applied research has three sub-types. All are evaluation procedures:

• Effort (or process) research. This generally tries to answer these questions about treatment programs: What is the program doing? Do the services match what was originally planned? Do the resources support the program's mission? The focus is on how the program is actually running, and what resources (e.g., staff, time, and money) are being funneled into the various program components that are being appraised (National Institute on Drug Abuse, 1993).
• Outcome (or impact) research. This form of applied research attempts to determine whether a treatment program is really helping clients, or whether something else is making the program look good (National Institute on Drug Abuse, 1993). The true and quasi experiments (discussed in Parts 2 and 3 of this series) are one way to measure the effectiveness of programs.
• Cost-effectiveness analysis. The third type of applied research is an attempt to determine whether the monetary costs of a program are outweighed by its benefits (Polit and Hungler, 1995). In other words, is the program paying for itself? More efficiency is being required of human services these day, and cost-effectiveness is more important than ever. However, the principles of this analysis in human services are difficult to express in dollars. The general idea is to offer services that will justify the investment of funds (Posavac and Carey, 1997). Thus, a recap of the costs incurred by treatment providers (and analyzed in this form of applied research) is in order.

Posavac and Carey (1997) discuss several categories of costs. What will be the fixed versus the variable costs of a program? (Fixed costs are those needed to simply open the facility to clients. Variable costs are the supplies and maintenance that vary with the number of clients.) Incremental costs cover daily operations. Sunk costs have been expended already. Hidden costs include such things as adding staff, while utilities are an example of recurring costs.

What are the direct versus indirect costs? Direct costs are those that support the agency and make it possible to provide services, while indirect costs, or overhead, cover such expenses as the cleaning of buildings.

Some people might confuse all this evaluation with quality assurance, but they are not quite the same. Quality assurance tries to identify instances of substandard care, while evaluations do not involve comparisons to standards (National Institute on Drug Abuse, 1993).

Graphically, what we have so far is:

Basic Research Applied Research

Effort, Outcome, Cost-effectiveness

Experimental, Non-experimental

As you can see, applied research can have an experimental or a non-experimental flavor. In previous articles, we reviewed the experimental (true and quasi), and to some extent, the non-experimental (survey, correlation, and case histories). All we need to do now is briefly cover a few more designs in the non-experimental realm.

The two major designs in this category are observational studies and descriptive studies. The definitions are very similar depending on what book you read, but have a few subtle differences:

Observational studies. Observation does precisely that-it observes. The idea is to have one or several trained observers systematically watch and record a behavior of interest (Bordens and Abbott, 1996; Utts, 1999). There is no manipulation of variables or direct use of survey questions.

Generally, the observer uses a checklist of some sort, and indicates any behaviors observed that are on the checklist. The natural observation is a prime example of this type of research because one observes the subjects in their natural environment without attempting to control or manipulate any variables (Bordens and Abbott, 1996).

As simple as this sounds, it can yield very useful information. For example, I can observe a series of people with substance-use disorders, note how much time they take off from work, and compare that data to the work habits of a sample of people who do not have drug or alcohol problems. I haven't done anything to the sample I just observed. To draw some conclusions from this raw data, I can use a correlation statistic, which will give meaning to the behaviors I have observed. (For example, there is a higher correlation of people who drink excessively and miss work, versus a group that does not.) This process is an example of a case-control study; I compare subjects who have a certain problem with controls (Polit and Hungler, 1995). This kind of research definitely has a place in the addictions field.

The down side is that the information you obtain may not be particularly valid. With no controls or matched groups, results can be explained by other variables that the observer never sees.

Descriptive studies. In these approaches, you summarize the condition or status of phenomena as they currently exist. These can be specific variables, the relationship of those variables to each other, or a possible underlying dimension to those variables (Heppner, Kivlinghan, and Wampold, 1998). For instance, most epidemiological research is descriptive. The example of the rates of adolescent marijuana use over the past several years from Part 5 of this series is an example of epidemiological descriptive research. Many types of surveys fall into this research mold.

Another type of research that falls into this general mold is factor analysis (Heppner, Kivlinghan, and Wampold, 1998), a method to summarize or reduce a large amount of data into a common set of patterns or factors. As an illustration, let's take a case in which a treatment program wants to know why many of its clients relapse following treatment. The program conducts follow-up contacts and tests that indicate there are 25 clearly stated explanations why clients relapse, ranging from cravings and urges to peer pressure and boredom. It is all well and good that the survey discovered 25 reasons for relapse, but these reasons may not tell you much unless you can find a pattern in the information.

One way out of this dilemma might be to separate internal relapse triggers from external triggers. This makes the information more manageable; in fact, you might begin to notice that certain types of clients seem to fall into a pattern (factor) of relapse. Once that is established, you can build a treatment plan to offset that contingency.

The last descriptive design we will discuss is the ex post facto ("after the fact") (Heppner, Kivlinghan, and Wampold, 1998). In this case, the research takes place after the conditions or groups have been formed. Sometimes this happens in the natural course of events, and sometimes it happens following a condition that was not studied.

Think of the ex post facto as akin to a post-test only design (discussed in Part 4 of this series). For example, say you have a file room filled with old client records that were never used in formal research or any follow-up; many programs have this very situation. With an ex post facto, you could go back and relate possible effects of certain counselors or treatments to outcome.

The major problem with the ex post facto design is that it does not control for chance variables; thus, the results can lead to mistaken conclusions. However, this type of design is useful early in the investigation process, because it can suggest possible relationships to test under more stringent conditions.

The Location Dimension

Research can be conducted in a variety of places. We will discuss two of them (Shontz, 1986):

• Field research simply means that you go to the place where you want to observe whatever interests you. This is a naturalistic form of research because it describes events in a real-life setting. Observation of a DUI arrest or drinking at a family reunion would be considered field research.
• Laboratory research is the opposite of natural research. Here the experimenter wants to control the many conditions that might contaminate what seems to be the cause of an event under observation. Labs can also be used to collect survey data (e.g., subjects are asked about their drinking habits and perhaps given a test).

The Time Dimension

Just as there are location dimensions to consider when doing research, there are also time elements. Two types of research involve time:

• Cross-sectional research. In this case, the researcher collects data at one point in time (Polit and Hungler, 1995). This form of research is suitable for describing the status of phenomena, such as the recovery rate of clients from a program in June 2000. It can also describe relationships among variables at some fixed period in time, like a change in attitude about taking drugs following a prevention workshop. Cross-sectional research is practical and economical.
• Longitudinal research. Here research is collected over many points in time. For example, the famous Harvard Medical School study of adult development followed 204 men who were tested in college and for 40 years thereafter. Now that is a longitudinal study!

Getting the Big Picture

You know that important information can be obtained from a single well-constructed piece of research. Imagine what you could find out if you combined many research findings into one large finding. That is the aim of meta-analysis (Utts, 1999).

Once all the separate pieces of research are combined, you can often get a more precise estimate of a relationship, or find patterns across studies. In the case of outcome research, this generally measures the relationship between a treatment and an outcome (Sexton, Whiston, Bleuer, and Walz, 1997). As with all research there are problems with a meta-analysis. For example, you have to make sure you are comparing the right variables, and you could be comparing flawed or biased original studies. These problems would certainly throw off any ultimate finding.

I hope you now feel you have a pretty good understanding of the basic types of quantitative research used by most SUD researchers.

Michael J. Taleff, PhD, CAC, MAC, is assistant professor in the Counselor Education Department and project director of Chemical Dependency Programs 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
Bordens, K.S., & Abbott, B.B. (1996). Research design and methods: A process approach (3rd ed.). Mountain View, CA: Mayfield Publishing Co.

Heppner, P.P., Kivlinghan, D.M., & Wampold, B.E. (1998). Research design in counseling, (2nd ed.). Belmont, CA: Wadsworth.

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.

Polit, D.F., & Hungler, B.P. (1995). Nursing research: Principles and methods (5th ed.). Philadelphia: J.B. Lippincott.

Posavac, R.J., & Carey, R.G. (1997). Program evaluation: Methods and case studies (5th ed.). Upper Saddle River, NJ: Prentice-Hall.

Sexton, T.L., Whiston, S.C., Bleuer, J. C., & Walz, G.R. (1997). Integrating outcome research into counseling practice and training. Alexandria, VA: American Counseling Association.

Shontz, F.C. (1986). Fundamentals of research in the behavioral sciences: Principles and practice. Washington, DC: American Psychiatric Press.

Utts, J.M. (1999). Seeing through statistics (2nd ed.). Pacific Grove, CA: Duxbury Press.

---

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."