28,000
Chat with our AI personalities
This is a non-random sampling technique where the initial respondents to a survey recruit others for a survey. Linear snowball sampling is when the first responent recruits one more person for the survey, who in turn recruits one more person for the survey. Number of people surveyed will increase in a linear manner. The related links provide good discussion of this form of surveying. It is a non-representative sample, but it is an effective means of surveying people with common traits who might otherwise be difficult to find. For example, I want a statistical profile of people that are "hackers" so every hacker I find, I would ask if I could interview his "fellow hackers." See related links.
It is more accurate, unbiased and includes every item in the population, whereas sampling may be biased, and sampling is not totally representative.
Self-selected sampling is a technique in data gathering which participants are taking initiative in the test or survey conducted. This brings results that are often biased and inconclusive.
Blocking is more for experimental design while strata is for survey sampling.
ome suggested ways: Larger samples, Better sample design, Better measurement, Better data validation, Better survey/questionnaire design.