Three Sample Research ProblemsNext, each of the three authors of this text describes the process that they experienced in formulating a research problem. When deciding on a dissertation topic, N. L. L. was interested in what makes a counselor "good" in the perception of clients:At the time, I was working as a supervisor of master's degree students who hoped to become counselors. There was anecdotal information available from clients about counselors in the field: some were considered excellent and received many referrals, other counselors were consid- ered marginal, and then a few were considered not good. How these differences between counselors were identified was not apparent. After looking into the literature, I found very little extant research on the topic; in fact, there was so little research that the problem would need to start with what is the definition of "good" when it comes to counselors?

After doing much reading, I realized that I was mainly interested in how potential clients chose a counselor. Once again, there was very little research in this area. I did find literature on willingness to seek counseling. The research in this area had been conducted mostly with undergraduate students. I thought it would be interesting to find out the level of willingness to seek counseling for counselors in training. There was an existing model in the literature that explained most of the variance of willingness to seek counseling for undergraduates; I wondered if this model would fit for counselors in training. Thus, my examinations into the research lead me to the research problem of Does Cramer's (1999) model of willingness to seek counseling fit for counselors in training?

Another example comes from J. A. G., who was working in the area of environmental physiology:

I was to give a talk at the Federation of American Societies for Experi- mental Biology. There were many well-known scientists at my talk, and I was nervous to say the least, especially because I felt that others in the audience knew more about my topic, regional distribution of blood flow during alcohol intoxication, than I. During the talk immediately preceding mine, a colleague sitting next to me asked how I felt. I answered that I felt fine but took my pulse and found my heart to be beating at a rate of about 110 beats per minute, considerably above my normal resting heart rate of 60 beats per minute but similar to my rate after moderate exercise. I wondered if this could be a healthy response. I first formulated my problem as could a high heart rate in the absence of exercise be normal?

Planning a Quantitative Research Project 25

Next, I found numerous studies that examined heart rate under conditions producing anxiety. The heart rate could get exceedingly high, much higher than mine had been. None of the previous studies examined the metabolic requirements (e.g., oxygen uptake and cardiac output) under these anxiety situations. On the other hand, several studies had examined metabolic requirements on heart rate during exercise. These studies considered elevated heart rate following exercise to be normal, since the heart must deliver an increased amount of oxygen to the tissues under higher metabolic demands. The problem now became a general question: If we measured the metabolic demands of a situation under anxiety would it be similar to a situation under exercise?

Now an obstacle became clear. How could we create two situations, one under anxiety and one under exercise that yielded similar heart rates? In order to remove this barrier, we decided to use a within-subjects design, where each participant took part in all conditions of the study. First we could determine the heart rate and metabolic requirements under an anxiety-provoking situation (e.g., prior to giving a talk). Next we could have the participant exercise on a treadmill at a workload high enough to give us a heart rate identical to that experienced under anxiety, and we could also measure metabolic requirements.

Now we could state our problem as how are heart rate and metabolic requirements related under conditions of anxiety? Our next step would be to change the problem statement into a prediction statement or hypothesis that could be directly tested.

The third example is a research problem faced by G. A. M. and his colleagues who were studying the motivation of infants to solve problems:

We had observed that infants who were born prematurely and also those who had been abused or neglected seemed to have lower motivation to master new skills and seemed to get less pleasure from trying. This clinical observation raised several issues. First, could the motivation of preverbal infants be measured? Achievement motivation in adults and older children had been assessed from stories they told in response to ambiguous pictures. Some other method would need to be developed for infants. Second, was it really the case that premature and abused or neglected infants were less motivated to master tasks? The second part of the research problem might be phrased as is there a relationship between prematurity, abuse, or neglect and mastery motivation?

Sources of Research Problems

The examples just discussed illustrate four common sources of research problems: the existing research literature, theory, personal experience, and clinical observation. The last two assume knowledge of the literature and theory in the field and the ability to relate it to the experiences or observations. Often experiences at work or school can be the source of a research problem, if you know what questions are unanswered at present and how to translate your unrest about incongruous phenomena into a testable research problem.

Clinical observation

In research, practice observation(s) that leads to a research problem or question.

26 Introductory Chapters

RESEARCH IN THE REAL WORLD

In sample study 1, Carnes-Holt and Bratton (2014) investigate a type of therapy with parents of adoptive children and present a research problem based on existing research literature. They state, "Barth et al. (2005) argued that the parent-child relationship is the primary reason that adoptive parents seek counseling" (p. 330). These authors are basing their study on the extant literature.

An important distinction that is sometimes confusing to students is that the word problem might convey the false impression that a research problem is the same as a personal or societal problem. These types of problems, however, may lead to research problems and questions or hypotheses that can be answered by collecting and analyzing data. For J. A. G. to worry that he would be nervous during his presentation is not a research problem. Likewise, for G. A. M. to be concerned about the apparent low mastery motivation of abused or neglected children is a societal but not a research problem.

One of the first steps in the research process is to read the research literature on and around the topic of interest so that you will be able to identify gaps in knowledge. We provide some advice about conducting the literature review later in this chapter.

Another source for research problems is theory. Kerlinger (1986) suggested that a theory explains natural phenomena, which is a goal of science, and he defined it as, "A theory is a set of interrelated constructs (concepts), definitions, and propositions that present a systematic view of phenomena by specifying relations among variables, with the purpose of explaining and predicting the phenomena" (p. 9). Most researchers suggest that a theoretical orientation should be presented at the beginning of an article as the basis for understanding the rest of the article, but we do not believe that a theory must be a part of the article; instead, explanation, rationale, or point of view could be substituted and satisfy, to some extent, the same purpose as theory. However, none of these concepts are as strong as a theory. A number of books deal extensively with the role of theory in research, but we have chosen to emphasize research design and how it influences data analysis and the interpretation of results.

The issue in contemporary social and health science research is not so much whether theory is important and how it should fit into an article but how important should theory be in designing research. We agree that theory is important and that the value of the results of a study depend, in part, on whether they support some theory.

Identifying Research Problems in the Literature

All published studies start with a research problem. Unfortunately, in many published research studies, the research problem is difficult to identify in one sentence or paragraph. It is common for authors to allude to the problem instead of outright stating the problem. Furthermore, due to the pressure to

Theory

A statement or group of statements that explains and predicts relationships among phenomena; a set of interrelated concepts, definitions, and postulations that present a systematic view of phenomena by specifying relations among variables.

Planning a Quantitative Research Project 27

integrate extant literature into introductions of research studies, many times it is impossible to discern how a research problem was identified; often, the research problem appears to have emerged from the existing literature based on how the information is presented.

Existing literature

That which has already been written about the topic or question to be studied; provides the basis for the literature review section.

RESEARCH IN THE REAL WORLD

An example of a research problem apparently stemming from personal experience is from Landrum and Mulcock (2007). These authors state their research problem as the following: "As the undergraduate psychology major continues to grow in popularity, the challenge to provide accurate advising information to large numbers of students also continues to grow" (p. 163).

Characteristics of a Good Research Problem

In addition to being grounded in the empirical (data-based) and theoretical literature and examining relationships between two or more variables, there are several other characteristics of a good problem. As indicated already, it should hold the promise of filling a gap in the literature or providing a test of a theory.

A good research problem should also be stated clearly and unambiguously, indicating the variables to be related. Often research problems start out too broadly or vaguely stated. Problems also should imply several research questions.

As just implied, a good research problem should be testable by empirical methods; it shouldn't be just a statement of your moral, ethical, or political position. You should be able to collect data that will answer the research questions.

Of course, the methods used must be ethical and consistent with the guidelines spelled out in Chapter 14. The problem also needs to be feasible, given your resources and abilities. Finally, it is desirable, especially for graduate students, to choose a problem that is of vital interest to you so that you can sustain the motivation to finish, often a difficult thing to do.

Another way of deciding on a good, appropriate research problem for a thesis or dissertation is to examine where it would lie on several dimensions. We provide three different dimensions on which to examine research problems.

Broad Versus Narrow

Previously, we stated that research problems often start out too broad. For example, "What factors cause low mastery motivation" is too broad and probably not a feasible problem. Remember that your time and resources are limited so practicality requires that you limit the scope of your problem. It is also important to realize that science progresses in small steps. Even big, well- funded research projects often raise more questions than they answer and usually address only a limited piece of a broad research problem. On the other

28 Introductory Chapters

hand, you don't want the problem to be too narrow. For example, usually it should deal with more than a single, limited research question and two variables.

Widespread Versus Limited Interest

It might seem that you would want as many people (scholars and the public) as possible to be interested in your research. Certainly you should be interested, and it is a good strategy to pick a problem that is of interest to your advisor. You will get more and better feedback from your advisor and committee if they have interest in and knowledge about the area. It is also desirable to choose a topic that is of widespread interest, but some topics become almost faddish and have so many studies about them that it is hard to make a contribution. If you choose a topic that is currently very popular, it is also important to find recent unpublished literature by attending conferences, searching the Web and ERIC documents, and writing researchers who have recently published in the area to see if they have something new. It is hard to find gaps in the literature of a currently popular topic because a lot of work still may be in progress. This point overlaps with the next.

Well-Researched Versus Unknown Territory

It is exciting to think that you might be the first one to explore an area. However, if that is the case, one might wonder why it is unexplored. Is the topic of very limited interest, as discussed already? Are there practical, ethical, or financial reasons? Is the topic too specialized or narrow? Of course, there are interesting and important topics that are relatively unexplored and are not faced with these objections, but they are not easy to identify. Quantitative researchers tend to place considerable emphasis on finding gaps in the literature so they tend to study relatively well-researched areas. Qualitative researchers, on the other hand, place less emphasis on finding literature ahead of time and tend to explore less well-researched topics, seeing where their observations lead them.

Frameworks for Stating Research Problems

A common definition of a research problem is that it is a statement that asks what relationship exists between two or more variables, but most research problems are more complex than this definition implies. The research problem should be a broad statement, perhaps using summary terms that stand for several variables, that covers several more specific research hypotheses or questions. Several ways to state the research problem are provided in the next section. We have used square brackets to indicate where one should fill in the appropriate name for the variable or group of variables.

Refers to the likely esoteric nature of the topic for a study, one which is of interest to a small and/or limited audience.

In this chapter, the authors each describe the process they went through to identify a research problem. For one of these three research problems, answer each of the following:

Is this a broad or narrow research problem? Explain.

Is this research problem of widespread or limited interest? Explain.

What topics in the literature might the researcher want to explore?

Explain.

Rate the stated research problem on the different characteristics of a

good research problemsupport your answer.