Print

How to Craft a Successful Research Question – SOBT

Count the number of concepts or phenomena that your research problem statement proposes to investigate. Do not include any concepts or phenomena that define the population, only those to be directly measured or examined or described. If your research problem statement contains only one concept or phenomenon to investigate, skip to Part 2 of the Activity. If it contains two or more concepts or phenomena, use Part 1. Follow the steps in your chosen part sequentially.

Part 1: Crafting Quantitative Research Questions

1. Start with your research problem (developed in Track I and no doubt revised in the interim, as your lit review developed). On a sheet of paper, write down the concepts and the target population in the problem statement. Identify the IVs, the DVs, and any sample variables. Next, write down all the sample variables and sample characteristics your research problem contains. If your problem statement does not identify the target population in enough detail to identify the characteristics or variables of the sample, go back to the literature and determine what they should be.
   
   

   Note: Learners who are proposing a qualitative (one-concept) problem, for the sake of practice, please re-tool it as a two-concept correlational problem.
   
   Practice example: The research problem is the impact of emotional intelligence, learning style, and personality type on job satisfaction in health care managers is not known. Your list of concepts would be:

   • Emotional intelligence (IV 1).
   • Learning style (IV 2).
   • Personality type (IV 3).
   • Job satisfaction (DV).
   • Health care managers (sample).
   
   

   Immediately, you notice that "health care managers" is a very broad category and does not help you determine specific sample characteristics and variables. A quick review of your literature suggests that you should study female managers working in specialty clinics. So to your list you add to "health care managers":

   • Female (sample variable).
   • Health care managers (sample characteristic).
   • In specialty medical clinics (sample variable).
     
     

   Apply: Do Step 1 with your own research problem.

2. Next, write down exactly the form of measurement that your problem requires. The main forms include:

   • Differences between IV-based groups on a DV score (used for causal and predictive questions).
   • Correlations between two or more variables (used for correlational questions).
   • Measures of individual variables (used for descriptive questions, usually subquestions).
     
     

   Practice Example: In the example, what is not known is the impact of the three IVs on the DV. Therefore, the kind of relationship we are seeking will be a causal relationship, based on that word impact. The word effect can be substituted, making even clearer that the relationship we are after is a causal one.

Apply: Do Step 2 with your own research problem.


3. Now, craft a conceptual version of your question using these ingredients. After you have a draft, examine it carefully. Does your question:

   • Clearly name the variables you are going to measure?
   • Clearly identify which are the IV and which are the DV (if applicable)?
   • Clearly indicate what relationship between or among the variables will be investigated (by words such as relationship, correlation, cause, affect, effect, influence, difference, predict, impact, and so on)?
   • The characteristics and sample variables of the sample?
   • End in a question mark, and have the characteristics of a good English sentence?
     
     

   If it misses on any point, correct it.

Practice Example: A first draft of the conceptual version of the example might be:

What is the impact of personality type on the job satisfaction of female health care managers in specialty medical clinics?

If we’re actually interested in whether this IV predicts the DV—prediction being a type of impact—another version might be:

Is personality type a predictor of job satisfaction in female health care managers in specialty medical clinics?




Apply: Do Step 3, the conceptual version, for your own research question.

4. Write an operational version of the research question.
   
   Practice Example: To determine effects (or impacts), one must do an experimental or a quasi-experimental design. This means that two or more groups are formed and the different effects on the IV are compared in those groups. To get to that, we look at the IV (or individual IVs if there is more than one). We ask ourselves what levels of the variable we are interested in. In reviewing the literature, we found that the so-called Big Five personality traits (Costa & McCrae, 1992) are widely used. This shows us five levels of personality type. But perhaps for some reason, we are most interested in the impact of extroversion and conscientiousness for the variable personality type.

Therefore, we might have an operational version of the research question that looks like this:


Are there statistically significant differences in levels of job satisfaction for female health care managers in specialty medical clinics who are highly extroverted compared with those who are highly conscientious.


Apply: Do Step 4 for your own research question.

5. The next step is to examine your research question and write any necessary subquestions. Use your operational version for this step. The basic approach is to write a descriptive question for each of the variables being investigated, as well as for any sample variable that must be measured.
   
   

   Practice Example: Our question has three variables to be measured, and therefore we have three subquestions to write:

   • Level of job satisfaction (DV).
     

     What are the levels of job satisfaction in female health care managers in specialty medical clinics?

   • Level of extroversion (IV), needed to assign participants to the "high extroversion" group.
     

     What are the levels of extroversion in in female health care managers in specialty medical clinics?

   • Level of conscientiousness (IV 2), needed to assign participants to the "high conscientiousness" group.
     

     What are the levels of conscientiousness in in female health care managers in specialty medical clinics?
     

6. Finally, write a set of hypotheses for each of the main questions and any subquestions that require at least a correlational analysis. If there are more than one question or subquestion that asked for at least a correlation (for example, if a complex main question requires a series of subquestions about correlations), we would make hypotheses for any question that requires a correlational analysis or stronger. We do not make hypotheses about descriptive questions, only correlational or difference questions.
   
   

   Practice Example: Because we have only one main question, we will need two hypotheses, a null and an alternate.

   • Null hypothesis (H₀): There will be no statistically significant differences in levels of job satisfaction for female health care managers in specialty medical clinics who are highly extroverted compared with those who are highly conscientious?
   • Alternate hypothesis (H₁): There will be statistically significant differences in levels of job satisfaction for female health care managers in specialty medical clinics who are highly extroverted compared with those who are highly conscientious?
     
     
   Prepare to Revise and Rewrite Your Question as Your Lit Review Deepens
   
   

   Your search of the literature about your topic and research question will continue until you complete your dissertation. When you find new research that changes the nature or focus of your problem statement, your question will need to be revised accordingly. This is quite common. So plan to keep revisiting your research problem and question regularly, up to the point where they have been finally approved by your dissertation committee and the IRB and you are ready to start your participant recruitment. That question and only that question will be your final edition.

Part 2: Crafting Qualitative Research Questions

1. State your research problem. Clearly explain the concept or the phenomenon to be investigated.
   
   

   Next, describe the qualitative meta theory and methodology you intend to use to guide the research design. In the School of Business and Technology, qualitative researchers may use Case Study, Ethnography, Phenomenology, Delphi or Exploratory methodologies.
   
   Now, identify the research setting and participants and explain how they align with the stated qualitative meta theory.

   Write down the characteristics of your research population, the sample frame and sample. What criteria might you use to define specifically the participants to include in the study?

   Review your draft problem statement. Did you spell out the methodology? Did you identify the target population? If not, you may need to revise your statement. Refer to the literature, as well as to methodology resources on your selected meta-theory.

   Practice example: If the research problem is how managers of a global virtual team assimilate and accommodate new team members from cultures not represented on the team. Your concept or phenomenon statement would be:

   • The experience of assimilating and accommodating new team members from cultures not represented on the team.
     
     

   Sample characteristics identified in the problem statement would be:

   • Managers of a global virtual team.

   
   On review it is apparent that while the phenomenon is stated clearly, the sample characteristics are stated broadly. "Managers of a global virtual team" does define the sample characteristics. A scan of the literature and reflection on your personal interests might suggest that you could study managers of teams with operations in the United States and India.

   Before you determine the criteria for selecting participants, you need to know what sampling approach you will use. Are you looking for a homogenous or heterogeneous group of participants? Depending on your decisions, you decide to add criteria such as:

   • Gender or age range.
   • Level of professional experience.
   • Type of industry or company.
   • Managers from one or from different companies.
     
     

   After reworking the statement, it might read:
   

   How experienced (more than two years) managers of virtual teams within one manufacturing firm with operations the United States and India assimilate and accommodate new team members from cultures not represented on the team.

   Apply: Do Step 1 with your own research problem.

2. The research problem as stated does not make reference to the meta theoretical methodology being used for the study. This one problem could generate very different types of studies, depending on the methodology. For example:

   • Case Study: you could look deeply at the "case" of one or more teams.
   • Ethnography: you could study the culture of the organization.
   • Phenomenology: you could look at the perceptions of team managers and/or members.
   • Delphi: you could look at how current team practices predict future behaviors.
   • Exploratory: you could explore selected aspects of the research problem.
     
     

   Apply: Do Step 2 with your own research problem, and record the methodology you intend to use for the research and what it allows you to study.

3. Next, write down exactly the form of data collection that your problem requires. Qualitative researchers collect data from:

   • Interviews: The researcher collects data based on participants' responses to questions.
   • Observations: The researcher collects data by watching how participants act or interact, or by watching the phenomenon as it occurs.
   • Archives and Documents: Researchers analyze the content of texts such as written documents, reports or company data, websites, blogs or other documents relevant to developing an understanding of the research problem).
     
     

   Practice example: In the example, the research problem is: how experienced (more than two years) managers of virtual teams within one manufacturing firm with operations the United States and India assimilate and accommodate new team members from cultures not represented on the team.
   
   As a researcher you could choose to collect data using any of the approaches used by qualitative researchers or a combination approaches:

   • Interviews: You could interview managers, or you could interview managers, existing and new team members.
   • Observations: You could observe team meetings in Adobe Connect, Web Ex etc., making note of how managers, experienced and new members participate and communicate.
   • Archives and Documents: You could analyze company documents to track productivity, records of timely team achievement of projects, complaints to human resources or company ombudsmen.
     
     

   Apply: Do Step 3 and record what data collection approach you intend to use, given your own research problem, methodology, and sample.

4. Now, craft a conceptual version of your research question using these steps. After you have a draft question, examine it carefully. Does your question:

   • Clearly name the phenomenon you are going to investigate?
   • Clearly define and support your decision of a qualitative meta theoretical methodology?
   • Clearly indicate the characteristics of the sample?
   • Clearly identify the kind of data collection you will use?
     
     

   Make changes as needed to encompass all required elements.

Practice Example: Taking into account all these ingredients, a first draft of the conceptual version of the example might be:

Do the perceptions of experienced (more than two years) managers of virtual teams within one manufacturing firm with operations the United States and India align with the perceptions of team members when describing successful tactics for assimilating and accommodating new team members from cultures not represented on the team?

This version indicates that data will be collected from interviews with managers, experienced and new team members. The term "perceive" suggests that this is a phenomenological study.

Apply: Do Step 4, the conceptual version, for your own research question.

5. Rewrite the research question as often as it takes to make it crystal clear, succinct, and yet complete. Once you have the content in place, review for proper sentence structure and proofread for grammatical errors.
   
   

6. Prepare to Revise and Rewrite Your Question as Your Lit Review Deepens
   
   Your search of the literature about your topic and research question will continue until you complete your dissertation. When you find new research that changes the nature or focus of your problem statement, your question will need to be revised accordingly. This is quite common. So plan to keep revisiting your research problem and question regularly, up to the point where they have been finally approved by your dissertation committee and the IRB and you are ready to start your participant recruitment. That question and only that question will be your final edition.

Reference

Costa, P. T., & McCrae, R. R. (1992). Revised NEO personality inventory (NEO-PI-R) and NEO five-factor inventory (NEO-FFI) professional manual. Odessa, FL: Psychological Assessment Resources, Inc.

---

Doc. reference: phd_t2_sobt_u04s4_h01_craftque.html