For research studies seeking data for the general population, it is a good idea to use quotas to ensure that the distribution of respondents accurately reflects the population of the country. This is known as a Census Representative Sample, as the data used is taken from the most recent government Census. For example, a Census Representative Sample based on the the most recent Census conducted in 2010 would aim to have 51% of the respondent pool be female, 37% of the respondent pool be between the ages of 18–44, and so forth.
Census representative samples are not always the best option. Some research will only be targeted at certain groups, and therefore will be better off restricting the study to only the relevant population.
A study seeking to measure nationwide use of social media sites would benefit greatly from a Census Representative Sample.
When performing market research to determine interest in a potential new product targeted exclusively at women, a Census Representative Sample would not be necessary.