When to Use Theoretical Sampling
This is part of our Practical Guide to Grounded Theory | Start a Free Trial of Delve
What is theoretical sampling in grounded theory?
Theoretical Sampling Definition
Theoretical sampling in grounded theory, as defined by Glaser and Strauss (1967) is a way of collecting data, and deciding what data to collect based on the theory and categories that emerge from your data.
Theoretical Sampling Method
There is no pre-set notion of who to recruit, or any predetermined groups of people to compare. Instead, you start somewhere with data collection, analyze the data, and then determine from your learnings where to collect data next.
When utilizing theoretical sampling, the process of collecting data, coding it, and analyzing it, happens simultaneously and recursively, and not as discrete steps that lead into one another.
Theoretical sampling is part of grounded theory, which you can learn more about in our Practical Guide to Grounded Theory.
When would you use theoretical sampling?
You should do theoretical sampling if you’re looking to determine a new theory based on data, such as when practicing a grounded theory method of research. You should also make sure to have a degree of flexibility in how you recruit, and the timeline that you’re working on.
You should not do theoretical sampling if you’re conducting research to verify or validate existing hypotheses, or conducting research that is intended to be descriptive. You should not do theoretical sampling if you have a strict criteria for recruitment or a strict timeline for your research.
How do you do theoretical sampling?
Begin data collection by starting somewhere
For the first set of data you collect, collect it based on existing domain knowledge or a partial framework, even if you don’t know yet whether or not these constructs will be relevant to your theory in the end. For example, if you’re studying a hospital, you may start by interviewing nurses, doctors, and patients, even if these ‘roles’ aren’t what end up being relevant in your final theory.
Be theoretically sensitive and have an open mind
As you collect data, you should have an open mind about various theories and categories that can emerge from your data. You should constantly be working to discover relationships between categories you derive from the data. In the example of doing research on a hospital, don’t be solely focused on the differences between nurse and patient experiences. Be open to new possibilities and theories. As you analyze data you might find other distinctions that are more relevant to your emerging theories.
Do not plan your data collection in advance
Some research studies, especially studies that are intended for hypothesis verification or description, involve a detailed process of defining recruitment criteria and methods for collecting data, before going out to collect or analyze data. In theoretical sampling, you do not do that. Instead, you take the initial set of data you collect, analyze it, determine what some emergent categories are, and then decide from these categories where to collect data next. Your rolling analysis and emergent categories define the next step.
Recruit based off theoretical purpose and relevance
Keep your recruiting criteria flexible, and know that your purpose is to generate new theory, and not to just verify what you already know. After doing analysis from earlier rounds you should have an idea of categories that were derived from your data. Look at gaps in those categories, and take not of new questions that emerge. Use these gaps and new questions to guide your criteria for recruitment in the next round of data collection.
Select comparison groups based on theoretical purpose and relevance
Many research studies, especially those meant for verification or description, may pre-prescribe which sets of groups should be compared in analysis. The issue with this approach is that the data you collect based off this pre-set criteria may be irrelevant to your emergent theories. In theoretical sampling, you do not decide ahead of time what groups to compare, and instead use your data to determine what groups are most relevant for comparison.
What is theoretical saturation in qualitative research?
When doing theoretical sampling, you need some way to determine whether you’ve collected enough data. You don’t determine the amount of data to collect ahead of time. Instead, you collect data until you reach theoretical saturation.
Theoretical saturation occurs when adding additional data doesn’t contribute any more properties to your categories. As you collect data in grounded theory, you’re looking to extend and elaborate on your categories, and push for diversity in your recruiting to ensure that you’ve covered everything you can learn related to each category.
Once you have a reasonable amount of confidence that each category you derived from your data is saturated, you have reached theoretical saturation.
Data collection, coding, and analysis happen at the same time
There are many research methods out there where data collection, coding, and analysis are distinct steps in the process, with one phase leading into the other. When you’re doing theoretical sampling and practicing grounded theory, however, these 3 steps happen simultaneously. As you collect data, you code and analyze it, and use what you learn to determine what data to collect next. It’s a continuous process of looking for emergent categories, reformulating them, pruning your list, and then continuing to build upon your theory.
Watch Webinar
See how theoretical sampling works in grounded theory.
In conclusion
Overall, theoretical sampling is a great way to recruit based on what you learn along the way, to ensure that the data you collect is relevant to the theory you’re developing. To learn more about the coding process in grounded theory, check out our article How To Do Open, Axial and Selective Coding in Grounded Theory.
If you’re interested in learning about a different method of analysis, where you do determine the recruiting criteria and research timeline in advance, check out How to do thematic analysis.
Qualitative analysis doesn't have to be overwhelming
Take Delve's free online course to learn how to find themes and patterns in your qualitative data. Get started here.
Try Delve, Grounded Theory Software
Online software such as Delve can help streamline how you’re coding your qualitative coding. Try a free trial or request a demo of the Delve.
References
Glaser, B. G., & Strauss, A. L. (1967). The discovery of grounded theory: Strategies for qualitative research.
Cite this blog post:
Delve, Ho, L., & Limpaecher, A. (2020a September 11). When to Use Theoretical Sampling https://delvetool.com/blog/theoretical-sampling
