When you're doing research, you're not always able to ask everyone you'd like about your topic. For example, if you're doing a survey, you can't ask everyone in the world to answer your questions. However, with a basic understanding of samples and populations, you can get good results that are useful and fair. This knowledge is valuable for research in business, social science, biology and many other fields.

## Creating and Analyzing Your Sample

### Step 1

Define the characteristics of the large group you wish to research. For example, you might be interested in the laundry detergent preferences of Hispanic women who live in urban areas. This group of people is the population whose preferences you will study. (A population does not have to be people; it could be inanimate objects, animals or anything else you can count and analyze.)

### Step 2

Select a sample. To find out about these people's laundry detergent preferences, create a sample of this population -- a smaller group of people who meet your desired description. You may do this by assigning members of the population numbers and then selecting numbers randomly; you might choose people walking through a mall at random to ask them to take your survey. There are many methods of choosing random samples, which are generally considered to be the most scientifically sound. If you choose people who are convenient for you or because they're your friends, they might not actually represent the people you're trying to study.

### Step 3

Remember the importance of generalizing. You must be able to show that your sample is truly representative of the entire population. For example, say you choose to interview Hispanic women only in Los Angeles and Houston about their laundry detergent preferences. Is it fair to say that they represent Hispanic women in all urban areas? You are asking whether it is possible to generalize the information from these women in these cities to women in other urban areas. If you think this is a fair generalization, proceed with your analysis; if not, you'll have to broaden your sample to include women in other cities, or change your expectations for the research.

### Step 4

Do your research. For this example, you would likely do a survey of the women in the sample to see which detergent they prefer.

### Step 5

Analyze your results and draw conclusions. If you are confident that your sample is truly representative of the population you're interested in, you can draw generalizable conclusions from your research.