Dissertation Rule #534: Clean up your Research Questions
When frustrated Ed.D. and Ph.D. candidates come to see me with their dissertation woes, the first thing I do is look at their research questions. More often than not, their questions are ambiguous, or worse, impossible to build a study around. Then we get to work on revising the research questions. Research questions should be:
Once your questions exhibit these 5 qualities, the study opens up and becomes manageable.
Below is an example of 2 research questions and revisions. The following questions could be for either a qualitative or quantitative research design.
Research Question 1:
How do Latino students define success?
Woah! This is way too broad. First of all, what is success? Success is an abstraction and means different things to different people. This question is too big and not possible to measure in one dissertation. It does not meet number 5, specific.
Revision of Research Question 1:
How does membership in the Latino Student Union contribute to Latino students’ success in postsecondary education as measured by the XYZ assessment/survey?
The question is now clear, concise and measurable. The question now has a specific independent variable (membership in the student union) and dependent variable (results on the XYZ assessment/survey).
Research Question 2:
Does intervention program X work for students who are struggling to read on grade level?
The first problem with this question is that it is a yes or no question. People don’t write dissertations to answer yes or no questions. Your study should identify factors that affect certain outcomes. Another problem with this question is the word struggling. The term struggling is ambiguous and should be narrowed. This question is not specific enough to build a study around.
Revision of Research Question 2:
In what ways does intervention program X affect student performance on the state reading exam?
In the revision, the question becomes much more specific than before. In addition, rather than a yes/no question, it lends itself to having several answers. For example, the intervention may affect students’ scores in comprehension, vocabulary, fluency, etc. Finally, the revision presents a clear independent variable (intervention program X) and depended variable (performance on the state reading exam).
Things to remember when constructing questions:
You are not trying to save the world here or solve all problems in education. You are conducting a small study that can be used to inform decisions. The smaller the better. The #1 goal of the dissertation is to GET IT DONE. Keep your research questions tight, measurable and manageable. If it feels too broad, it probably is. Be specific as to what aspect you are trying to measure.