The dissertation methodology chapter serves as the keystone to your dissertation, mainly because of this truism: A doable methodology ensures you have a feasible study. Equating your methodology chapter to a specific recipe in a cookbook gives you another way of thinking about it: It requires you to describe all of the ingredients, outline how you prepare and cook everything in the proper steps, such that others can replicate the same results as you describe in the end. When researchers forget about the true function of this key chapter, the scientific world ends up with a flurry of vetted, yet non-replicable, studies. (In fact, a spate of these types of studies have made the news recently.) As such, when writing your dissertation methodology, here are a few things to keep in mind.
Research questions: The point cannot be stressed enough: clearly stated specific research questions is key to successfully writing a methodology chapter and executing the study. When you conduct your research using this methodology, make sure you’re going to be able to clearly state support or not support of the research questions or hypotheses. This clear statement presumes that you have succinct research questions with operationalized constructs.
Participants: While this may sound obvious, you must fully describe your study participants’ characteristics so you can then target gaining access to this specific population. While very simple, there are lots of potential reasons why researchers fail to get access (e.g., the organization changed their mind, IRB said no, etc.). Therefore, pick a population, or data sources, that you definitely can gain access to in a timely manner.
You also want to properly describe study participants for generalizability purposes. Back in the old days, much of the research was conducted using males only and researchers could not then generalize the study’s findings to females. An analogous lack of generalizability applies to different ages, cultures, educational status, etc.
Sampling method: You should clearly describe how you will recruit a sample from your population. Most doctoral students use a convenience sample or snowball sampling. Strategy–this is absolutely fine. While dissertations have to address the topic of generalizability, don’t worry too much about the extent to which your self-funded project actually generalizes.
Materials or Instruments: Most importantly, researchers must use reliable and valid instruments in their studies. Thankfully, past research has already generated a multitude of instruments for which reliability and validity have already been assessed via tests like Chronbach’s alpha and construct validity. Be sure to use these metrics in your instrument section. Furthermore, we strongly suggest not creating your own survey because your committee or IRB will likely ask how you plan to validate your survey, and they may even ask you to pilot test your instrument, which extends your time to completion.
Procedure: The procedure you use has to be so clear that your next door neighbor can replicate your study, assumedly recreating the same findings. As such, take care to specify as much as necessary to enable this replicability process. To borrow from the recipe analogy earlier, if your stew meat requires cubing, state this fact; if the cubes are ¼ inch or 1 inch thick, include this information. If you feel someone needs to know something in order to replicate your study, write it down.
Data analysis plan: This piece to your methodology explains how you plan to analyze your data. For quantitative analyses, simply walk through the steps. In general, these steps will likely include something akin to conducting descriptive statistics on all of your data before then stating, for each hypothesis, how you will assess any and all assumptions for each analysis, conduct and interpret said analysis, and then create tables and figures to illustrate your findings.
For qualitative analyses, most studies seek to thematize data. (For reference, see Lincoln and Guba’s evaluative criteria .) Take care to address the issues of credibility, transferability, dependability, and confirmability, as well. With regard to the number of researchers, a general rule of thumb is that the more researchers involved in the process the better. Lastly, don’t forget to speak to the interrater reliability which you can assess with a kappa coefficient.
We work with graduate students every day and know what it takes to get your research approved.