Module 1: Preliminary Research Steps
This course is a brief overview about research design that is intended to cover the basics of designing and implementing a scientific study. Although this course will address every step of the research process, it is by no means exhaustive and is no substitute for a college level course in research methodology, nor is it a substitute for an experienced research advisor. For further study in this area, see Trochim’s “Research Methods Knowledge Base.”(1) It is one of the main sources used for this guide and provides descriptions about the various parts of the process, including diagrams and examples to simplify complex concepts.
Choosing a Topic
For a researcher to choose a topic for a project, it is important to consider a broad area of inquiry and interest. This may be as broad as “global eye health” or “personality psychology,” but it should be an area that is of interest to the researcher. However, a broad area is useful only at the beginning of a research plan. Within a broader topic of inquiry, each researcher must begin narrowing the field into a few subtopics that are of greater specificity and detail. For example, a researcher may be interested in “global eye health,” but could focus more specifically on “proper eye care and how it affects individuals.” Although this topic is still too broad for a research project, it is more focused and can be further specified into a coherent project.
Oftentimes, students as well as professional researchers discover their topics in a variety of conventional and unconventional ways. Many researchers find that their personal interests and experiences help to narrow their topic. For students, previous classes and course material are often the source of research ideas. Furthermore, current events in politics as well as in academia can inspire topics for research. Academic journals such as Health Affairs, Health Economics, and the American Journal of Bioethics can provide good material for new studies and E-resources such as Pubmed, Google Scholar and Philosopher’s Index are also good starting places. Lastly, many research ideas are generated through dialogue—by talking with professors, fellow students and family.
One essential task when undertaking a research study is to review the existing literature on the topic and use it to inform the construction of your own study. The literature review should be conducted early in the research process, directly after you choose a topic. A literature review can bring clarity and focus to your research problem and broaden your knowledge base in your research area. In addition, past studies can improve your methodology and help you to contextualize your findings. The literature review is crucial because an important responsibility in research is to add to a body of knowledge and to compare your findings with others. The procedure is simple: search the literature in your area of interest, review the selected studies, and develop a theoretical framework for your own study. For those pursuing research about community eye health, Unite For Sight's Journal Article Database can be used as a starting point.
What makes a good research question?
Not all research questions are good ones—in other words, not all questions can be answered through qualitative and quantitative research methodology. A good research question needs to:
- “Make sense”: In other words, you must clearly define your terms using known definitions outlined in the literature. For example, a poor research question would be: How do people’s lives improve after surgery? Not only does this research question fail to specify the study population, it contains the vague term “improve”. The researcher must specify what he/she means by this term—does it involve a physical improvement or rather an improvement in mental state? The more specific your research question, the better.
- Address an important and relevant issue: Scientific research is done to increase knowledge, not simply for a single researcher’s personal satisfaction. Whatever question the researcher sets out to solve must have some beneficial implications. With this in mind, the researcher may continue narrowing the study focus to an area that can be addressed as a single question. For example, now that the researcher has chosen “proper eye care and how it affects individuals,” the topic can be further focused to be about “basic eye care and how it affects individual work productivity.” A good research question will also always have relevance to the time, place, and population of the study. For example, a study of Vitamin A deficiency in Southern India would be a poor choice as this is not a particularly significant problem in the area.
- Not already have been done: A good research study will be novel. This means that there will be some new aspect of the study that has never before been examined. However, this does not mean that you should avoid replicating past research. In fact, not only is replication a good way to get a research methodology, it is how science is supposed to advance knowledge. When replicating a pervious study, it is best to add or change one or two things to increase the novelty of the research.
- Be “operationalizable”: Oftentimes, beginning researchers pose questions that cannot be operationalized, or assessed methodologically with research instruments. From the example above, the idea of life improvement could be operationalized by a Quality of Life survey—a well known and validated research tool. In general, the more abstract the idea, the harder it is to operationalize.
- Be within a reasonable scope: A good research project will be manageable in depth and breadth. The scope will depend on the amount of time and the availability of resources you have for your study. In general, the more focused the research question the more likely it will be a successful project. For example, a study that seeks to identify the prevalence eye disease in a specific village is more likely to succeed than a comparable study that seeks to identify eye disease prevalence in the world population.
Qualitative and Quantitative Studies
Not all research projects require study measures. Some research simply involves observing the results of events in the field and drawing conclusions based on a theoretical framework. Others may involve analyzing data from clinics or other institutions, using statistics and reasoning to find patterns that may have important implications. However, many projects involve direct contact with participants, using an operationalized definition of a phenomenon. These projects require well-designed measures in order to be considered valid. There are two broad categories of research: quantitative and qualitative.
A study is classified as qualitative if the purpose is primarily to describe a situation, phenomenon, problem or event; the information is gathered through the use of variables or measured on qualitative measurement scales, and if analysis is done to establish the variation in the situation or problem without quantifying it. Qualitative studies tend to be more “in-depth”, focusing on a smaller population but probing deeper into a given problem. This research is often associated with focus groups, interviews or surveys and seeks to answer open-ended questions. Thematic and content analysis are two methods used to analyze qualitative data. Disciplines such as anthropology, history, and sociology are more inclined towards a qualitative approach.
On the other hand, quantitative studies often use standardized measures, numerical values, have larger sample sizes, and analyze data using statistical programs. A study is classified as quantitative if the researcher seeks to quantify the variation in a phenomenon and if information is gathered using quantitative variables. Both qualitative and quantitative approaches have their strengths and weaknesses, and advantages and disadvantages. Disciplines such as epidemiology, economics and public health are more inclined towards quantitative research.
A hypothesis is a suggested explanation for an observed relationship or a causal prediction about a relationship among several variables. Every research project is based on a hypothesis, which generally begins with a specific question. For example, “If people are provided with basic eye care services, will they be more economically productive on an individual basis?” This question is specific enough to be addressed by a research project, however it is not yet a hypothesis. Next, the researcher must operationalize the terms being used. Operationalization refers to defining otherwise abstract concepts or terms in a measurable way. For example, “economically productive” can be operationalized as “dollars earned per day,” “hours worked in a week,” or “number of objects successfully produced at work.” As we can see, a researcher must be careful to operationalize the measures in such a way that they reflect exactly what the researcher is trying to measure. Depending on how terms are operationalized, the results of a study can vary widely, so it is critical that a researcher carefully consider how each of the measures are to be operationalized before forming a hypothesis and beginning a study.
A hypothesis takes the operationalized definition of the factors to produce a clear prediction of the causal relationship between the independent variable and the dependent variable in the statement. The independent variable is a factor that the researcher can control or manipulate (whether or not a person receives basic eye care services), and a dependent variable is a factor that the researcher cannot manipulate, but instead varies in relation to the independent variable (the economic productivity of the individual). For example, a hypothesis might be “We predict that if nearsighted participants are provided with corrective lenses that bring their vision to 20/20, they will earn more money per week on average over the course of three months than nearsighted participants who did not receive corrective lenses.” This statement is a viable hypothesis because it clearly operationalizes what the researcher termed “basic eye care” and “economically productive” such that they can be measured and analyzed in an objective way. (2),(3)
When formulating a hypothesis, it is important not to try to “prove” that the hypothesis is true. Instead, one should seek to find evidence that it is not true. In other words, one can never accept a hypothesis; instead one fails to reject the null (posited) hypothesis. This is especially important when using statistics such as t-tests and p-values to determine significance.