urveys are another way to learn about and learn from a group of people. According to the book, How to Ask Survey Questions (Fink, 1995), the definition of a survey is, “a system for collecting information to describe, compare, or explain knowledge, attitudes, and practices or behavior.” Doing a survey requires planning—planning what you want to ask; how to ask it; how manypeople to survey and how to reach them, either by mail, in person, or by telephone. Will you have team members asking the questions or will the respondents, those completing the survey, fill it out themselves? This guide addresses these issues step-by-step, so that you can make choices and initiate a survey as systematically as possible. Although surveys are a great way of learning about apopulation, remember that the information you learn may not always be representative of the views of the population that you’re trying to assess. If only a small group of people completes a survey, there is always the risk that their views differ in some way from the opinions of the rest of the larger group. This issue will be discussed in more detail in the following pages. This section of the manualwill first address the issue of developing a survey instrument, the list of questions that you’ll be asking. Included in this section will be what to keep in mind in developing your questions and how to use a qualitative process, like the one described in Part II, to develop a questionnaire. Also included is information on drawing questions from existing surveys. After that comes sampling, theprocess of establishing a “sample” of a larger group and the size of this sample. Along with this, we’ll illustrate the variety of sampling methods that you can choose from to find out more information about your population.
Developing a Survey
lthough we often think of surveys as pieces of paper or sets of questions, a survey is something bigger than that. It’s the entire process that we’llbe discussing in Part III of this guide, from deciding who will complete the questions and how to reach them to determining what questions to ask and what to do with the information once the survey questions have been completed. Usually, the first issue involves the survey design. What is a design? It’s the structure of the survey based on what you’re trying to accomplish. If you were trying todetermine the difference an intervention makes, such as the impact of a prevention strategy on condom use, you might have two groups. One would take advantage of the prevention strategy (known as the intervention group) and one would not (also known as the control group). Then you could use a questionnaire to see what changes have taken place in the use of condoms among the two groups. You canalso use a strategy to measure what changed in terms of the use of condoms by surveying a group of people as a pretest, delivering the prevention strategy, then surveying them again afterwards to see what changed. Measuring the difference a prevention strategy makes by using a survey is very complex and is best done with the help of a statistician, an epidemiologist or someone who is trained instatistics and survey design. For the purpose of this guide, we’re assuming that you would want to do a survey in order to be able to describe what’s going on in the community. This is called an observational design. Conducting a survey is just another form of assessment—the use of a set questions asked verbally or on paper to find out more. So that’s where we’ll start—developing the questionnaire.Where do we start?
There are several ways to approach developing a survey. If you followed the step-by-step approach presented in Part I, you already have a good start toward developing the survey. By this time, you will have established some goals and objectives and the “big questions” that you want to know about. You may have even followed the steps in Activity 1, the...