It is seldom impractical to interview the entire population when conducting surveys. Therefore, sample design and selection procedures are critical components of every survey. Ideally, a sample should be random, representative of the population from which it is selected and sufficiently large to accommodate the level of dis-aggregation required during data analysis as well as facilitate the statistical procedures to be used when analyzing data.
Based on the nature of your study, your research questions, project budget and the availability of proper sample frames - among other considerations- we can develop an appropriate sampling strategy that will reinforce the external validity of your study allowing you to generalize your findings to the entire population.
A well-designed sample minimizes the possibility of biased results.