My target population is the Facebook followers of three fashion-wear retailers in SL. In this kind of situation, what will be the most appropriate sampling method?
having a random sample is impractical for most of the online studies, specially for social media studies. Hence, convenient sampling is suitable to adopt as along as you do the power analysis to satisfy sampling adequacy for given statistical technique.
You have very nicely explained and nicely elaborated sampling methods for online environment: Sampling Methods for Web and E-mail Surveys - > http://faculty.nps.edu/rdfricke/docs/5123-Fielding-Ch11.pdf
Sampling of online social networking site data is particularly tricky because of the elusiveness of a reliable sampling frame. Since we can't know that all social media units of analysis are equally eligible for selection, the probabilistic nature of this sampling type is compromised. For this reason, simple random sampling is likely not possible. A good alternative is systematic sampling where we start with a randomly selected member of the population, then select every nth case at equal intervals after that. An important note is that the population being sampled cannot already be ordered, as in sorted by popularity, length of video, etc.
- Badges
- Science topic
- Similar topics
- Statistics
- Survey