Statistical Surveys: Choosing Sample Populations

Choosing a sample population is all about sampling the right people. For example, If you want to survey people from your school about if they go see movies or not, would it be smart to only ask people at the movie theater? No! Because that would produce a biased sample. Of course everyone at the movies from your school sees movies, they are already there! A lot of choosing sample populations is about common sense.

Choosing a sample population

In statistics we conduct surveys or studies on many different topics in order to learn something about the population we are studying. Often times the population we want to study is too large and it is impractical or not possible to survey or study the entire population. In situations like this we choose a sample population. Sampling allows us to learn about a population without having to study each individual member of that population.

There are several ways to choose a sample population. Some methods are listed here. In order to choose a good sample population it is important to avoid bias and to choose a sample population that is representative of the larger population we are attempting to learn about. For example, if we want to learn how many students in your high school like baseball, we shouldn't survey your classmates at a baseball game. Instead, it would be a better idea to survey random students at your school.

Sample Populations

The goal of statistics is to discover the underlying trends or correlations within an entire group. An issue that comes up is that most of the time is that the group, or population, you're trying to study is simply too large. If you're trying to learn about the favorite foods of every person in the country and compare it to where they live, there would be too many people to survey. It would be almost impossible not only to ask every person in the country what their favorite food was and where they live, but also to get every person you ask to respond. In these cases you would want to use a sample population.

A sample population is a smaller group within the entire population you are trying to study that can be surveyed to find similar results. There are more than $325$ million people living in the United States right now, making it impossible to ask everyone. However, you could ask $1000$ people from each of the $50$ states. This means you still have to survey $50000$ people, but this is much more manageable than $325$ million!

If your population is small enough, you may not have to use a sample population. For instance, you wouldn't need a sample population to ask everyone in your class what their favorite color is.

Let's look at a population and see if we can break it down.

You're trying to figure out what kind of vehicle is more popular to drive in your state: cars or trucks.

Here you would definitely need to use a sample population. Looking at Pennsylvania as an example, there would be over $8$ million cars to analyze, and that's only if everyone you ask responds to you. You would want to break this down into a smaller sample population.

(Video) Sampling Basics

by Wake Take Business Analytics

In this video, we learn about random sampling.

$\large{\text{Summary}}$

You can choose a probably sampling by randomly choosing from a population.

If you want to sample people who watch tv during a show, you select a sample $n$ viewers. A random sample is only random if there is an equal probability for each person to be selected from the population.

The video also discusses the difference between stratified and clustered random sampling.

Stratified: taking a random sample from each strata separately. Each strata is a different subgroup in a population.

Example: Age groups.

Clustered: Population is separated into clusters and then random sampling is taken from the clusters.

Example: Regions in a country.