What you're describing is known as cluster sampling. In this sampling method, the population is divided into distinct groups, often called "clusters," which can represent the population as a whole. A random sample of these clusters is then selected, and either all members of the chosen clusters are surveyed, or a random sample is taken from each selected cluster. This method is often used when it is logistically easier or more cost-effective to reach groups rather than individuals scattered across a large area.
If you have further questions about sampling methods or need more detailed information, feel free to ask!