Recall that our population refers to the group of things that we want information about. And a sample refers to part of the population that we take out to examine and draw conclusions from. In this video, we will be looking at the different methods of obtaining a sample. But first, let's look at the types of biased samples. Biased samples occur when one or more parts of the population are favored over others. The two types of biased samples include the convenience sample and the voluntary response sample. A convenience sample only includes people who are easy to reach. If this is our population and a researcher comes along to interview people, then he would only talk to the people that are closer to him to be part of the sample.
This is a biased sampling method, because not everyone in the population has an equal chance of being part of the sample. Only people that are of convenience to the researcher will be interviewed. Now a voluntary response sample consists of people that have chosen to include themselves in the sample. So the researcher lets people come to him. This is a biased sampling method because people with a strong interest for the survey topic are the ones who are most likely to respond. Whereas the people who don't feel as strongly about the topic may not even care to respond. Remember that a good sample is one that is representative of the entire population, and it gives each thing an equal chance of being chosen.
When you have these conditions, you have what is known as an unbiased sample. We will be looking at three different types: stratified random sampling, multistage sampling, and simple random sampling. The most basic type of sampling is the simple random sample, also known as an SRS. Since an SRS is unbiased, each individual has an equal chance of being chosen to be surveyed—in other words, to be part of the sample. You can think of an SRS as putting names into a hat and selecting N of them. So if I wanted a sample size of six, I would select six papers and come up to the randomly chosen people to interview them.
For a stratified random sample, we take the population and we divide it into something called a strata. Strata refers to the groups of similar people. Within each stratum, we take an SRS and combine the SRSs to get the full sample. For example, we could take an SRS of two people from each group so that we get the total of six people. A stratified random sample is good for making sure that whoever's administering the sample gets in contact with each kind of group.
The last type of sampling is called multistage sampling. For multistage sampling, we use a combination of two or more simple random samples. As the name suggests, multistage sampling means you have to go through different stages to find where your sample comes from. For example, if we have three groups, stage one could be selecting which group will be picked using an SRS. Let's say that I picked out group one. Then that means I would only look at group one. Then for stage two, I would do another SRS to get the six random people. We go through different stages of simple random samples to get the actual sample. And this is why this is called multistage sampling.
I'd like to point out that instead of putting names in a hat, there's another way to pick things randomly. We can use something called the random digits table. The random digits table consists of a long string of random numbers, and it can help us do an SRS. To use it, I would first have to label each member of the population with a number. We have 30 people in this population, so I will label each person from one to 30. Notice how I have written zero one, instead of just one. Doing this helps us use the random digits table. Since each label has two digits, we will read the string of numbers, two digits at a time.
So let's say I want a sample size of four. We will use the random digits table to randomly select four people. The first number on the table is 19. So person 19 will be part of the sample. The second number is 22. So person 22 will be part of the sample. The third number is 39, but our sample size doesn't go up to 39, so we will ignore it. We will also ignore 50 and 34, but we will keep number five. We will also ignore 75, 62, and 87, but we will keep number 13. As a result, these are the people we would survey.