Often, we are faced with a problem that requires conducting a study.

For instance, we are interested in finding out how to get more people to be physically active in our local community.

The first step would be to assess current levels of physical activity.

In addition, we would like to know their awareness and perceptions regarding physical activity.

Finally, we would like to know what forms of physical activity are popular in the community (and why), and request suggestions to increase physical activity.

Now, that is a lot of work. If the local community is very large, (say, 10,000 people), then we would take a long time to get any answers to our original questions.

The obvious question at this point is, “Do I need to interview every single one of the 10,000 individuals to get the answers I seek?”

Fortunately, the answer is “No”.

It is possible to interview some members of the local community and still get what we want.

[How is that?

Assume you are asked to assess how a cake has come out.

You will not need to eat the whole cake to describe it: a (small) piece would suffice. You would still be able to accurately describe the cake, its flavors, etc.]

If we interviewed all 10,000 members of the local community, a statistician may say, “..the   entire population was interviewed..”

If we interviewed only 1,000 members of the local community, the same statistician would say, “..a sample of 1,000 people was interviewed…”

Summary:

Population: All the individuals who are eligible to be included in a study, for instance.

Sample: A selection of individuals from the Population who are included in the study, for instance.