How will you sample?
- Population: The population is the set of people or entities to which findings will be generalized. The population must be defined explicitly before a sample is taken and findings must not be generalized beyond the population.
- Sample: The units that will be included in the study, for example people, organizations, neighbourhoods.
- The sampling frame: A list of everyone in the population, from which the sample is drawn. It is often difficult to generate complete lists, for example telephone directories are often used as sampling frames but tend to under-represent the poor (who have fewer or no phones) and the wealthy (who have unlisted numbers). Think about the ways that you can address any bias in your sampling frame.
- Sample size: The number of cases needed in the sample. This can be calculated based on the of the population and the level of confidence that you would like in your findings (for social science studies the confidence level is generally 5% and for medical studies the confidence level is generally 1%). (for small populations, under 2,000, the size of the population also affects the sample size needed).
- Significance: is the percent chance of concluding there is a relationship in the data when there is none. Using the general social scientists cutoff of .05: if there is 5% or less chance that a relationship is just due to chance, we conclude the relationship is real.
For an online site that will help you calculate the sample size needed see http://www.surveysystem.com/sscalc.htm