Data Saturation
The point at which newly collected qualitative data becomes completely redundant with already collected data.
What is Data Saturation?
Data saturation specifically refers to the informational redundancy in the data collection phase. It is achieved when the researcher stops hearing or seeing any new information, concepts, or variations in the responses of new participants, indicating that the sample is adequately diverse and comprehensive for the research question.
Why Data Saturation Matters
It provides a pragmatic and methodologically sound stopping criterion for field work. It prevents the unnecessary expenditure of time and resources on continuing to collect data that will not substantively contribute to the analysis.
Example
A researcher conducting focus groups on consumer preferences for eco-friendly packaging stops scheduling new groups after the sixth session, as participants are only echoing the exact same environmental concerns and price sensitivities raised in the first five.
Common Mistakes
- Failing to concurrently collect and analyze data; if analysis waits until all data is collected, it is impossible to accurately assess when data saturation occurred.
- Asserting data saturation in a study with a highly heterogeneous population without sampling across the different subgroups.