We say that a sample is unbiased when each part of a population is represented proportionally to the number of members of that part. A biased sample, on the other hand, is a sample in which some parts of population are overrepresented or underrepresented. Let's try to visualize the difference between unbiased and biased samples.
Here, the people are chosen randomly and the possibility that our sample is biased is reduced. We can imagine that we are choosing all U.S. citizens born in August. Next let's imagine that we are choosing only people from our family or our closest friends.
Now we can see that we have a misrepresentation of members of the population marked as blue. When using a biased sample, our conclusions and statistics may not describe reality accurately. That is why choosing an unbiased sample is very important if we want to conduct a reliable study and draw appropriate conclusions.
