In qualitative research, there is no clear beginning, middle, and end to the research process. Data collection, analysis, and writing phases of research projects are often blurred and inherently iterative. This presents big challenges for doctoral students trying to complete qualitative dissertations. Each of the phases of research informs one another and introduces a messiness into the process that doctoral students undergoing their first major research project can find disorienting and confusing. While you will not necessarily “finish” data collection and analysis before moving forward with writing up the findings and results of your dissertation, the reality is that many students tend to move toward the significant writing of their findings following the competition of data collection and analysis. In today’s post, I want to provide some suggestions for how to know you are finished with qualitative data collection.
First, a disclaimer: I strongly encourage you to not close off the opportunity to gain additional insights and gather more evidence as you move into writing. Yet, the writing process itself can bring on a valuable perspective that cannot be replicated during collection and analysis.
The question that doctoral students often ask is when am I ready to write?
The answer is almost always before you feel ready to begin.
There is a tendency to want more information and know everything you could possibly want to know before beginning to write up your findings.
This desire is understandable as the ambiguity involved in qualitative research is uncomfortable and stands in stark contrast to the clarity often evident in qualitative dissertation students in terms of the steps needed to complete data analysis and report results.
That said, the way to know when you are ready to begin writing is when you hit what is called the data saturation point.
You have reached the data saturation point when you are no longer gaining additional information or that there is sufficient information to replicate the study.
In many ways, data saturation is problematic for doctoral students due to a lack of experience and the fact that in many ways you can only know saturation when you see it.
There are no formal rules to suggest after so many interviews or after coding so many documents that you will reach saturation.
Each study will be a little bit different on when you achieve saturation. Data saturation is simply not about numbers.
It is impossible to say that a large sample size will get you to saturation or a small sample size will preclude you from achieving this goal.
So how do you know that you have achieved saturation? There are a few questions that you can ask when thinking about your data:
- Are you starting to hear the same answers to your questions from multiple sources?
- Are you asking fewer follow up questions during interviews?
- As you work through coding, are you reusing the same ones over and over?
- As you put codes into themes and categories, are you adding more to existing themes and categories?
- Do you have data from all of the various perspectives that you anticipated when writing your proposal?
- When you ask participants to identify additional possible participants (snowballing), are you hearing the same names over and over? Have you already talked to these people (assuming you can get access)?
If you answer yes to these questions, the odds are you are probably getting close or have achieved data saturation. If you are unsure or worried you may not have enough data, a conversation with your dissertation chair can help provide clarity as their experience will give you confidence that you have met saturation or if you need to do more work.
After achieving data saturation and finishing much of your data analysis, you can begin moving more formally into the writing of findings stage of the dissertation.
It is natural to struggle with knowing when you have enough data to really change your focus to writing up your findings. I hope this helps you have a little better better understanding of how to know you are finished with qualitative data collection.