Aligning research questions and methods

Good research questions give your reader information about the data being evaluated and a clue about what method will be used to do the evaluation. There are three types of quantitative research questions: descriptive, comparative, or relationship-based. In our last post, we talked about some possibilities of methods that can be used for each of those types of questions.

It’s important to understand what methods can answer what types of questions. What kind of outcome are you looking for? Do you need a significance test? Doz et al. (2023) is a great example of using multiple statistical methods to answer different quantitative research questions. Using multiple regression, the authors predict academic achievement using their key measurements. They use a path model to assess relationship mediation between their key measures, and a MANOVA to uncover differences between groups. These three statistical tools were masterfully chosen to complete very specific tasks from the research questions. Many more tools exist that can do similar work as the ones chosen, but the authors chose simple and interpretable models to best showcase their outcomes and measures.

Need help matching your research question to a method? Request a consultation with the UGA Quantitative Methods Consulting Center! Please visit our website or email us at qmcenter@uga.edu.

Reference: Doz, E., Cuder, A., Pellizzoni, S., Carretti, B., & Passolunghi, M. C. (2023). Arithmetic Word Problem-Solving and Math Anxiety: The Role of Perceived Difficulty and Gender. Journal of Cognition and Development, 24(4), 598–616.