In statistics, a manageably small subset of a population from which information can be collected and conclusions about the population can be inferred. Samples are used to make educated guesses and develop statistics about a certain population, idea, area etc. Researchers often use the information and expand upon it at which point they can develop experiments based on the seemingly common situations.
It would be nearly impossible to poll every person in the world, but if they polled some people from each continent from different financial statuses/situations with different values, then they could presumably deduce that commonalities exist within the whole population…this process is known as sampling.
Example: Based on interview results from 200 undergraduates, study results indicate that 80% of college students are not awake before 9 AM.