Although most of us know what a survey is, many do not know all the tools necessary to create a survey that will yield quality data. In this lesson, you will learn about such tools.
Survey Study Defined
The survey is a method for collecting information or data as reported by individuals. Surveys are questionnaires (or a series of questions) that are administered to research participants who answer the questions themselves. Since the participants are providing the information, it is referred to as self-report data. Surveys are used to get an idea of how a group or population feels about a number of things, such as political debates, new businesses, classes, and religious views. Additionally, surveys can be a way for people to measure how often or how little people engage in different behaviors, such as smoking or drinking alcohol.
Surveys can be administered in various forms, including in-person interviews, through telephone interviews, through the Internet, or a paper questionnaire that requires participants to write their answers. The ways in which surveys are administered, however, are not the most important aspect in getting valid survey results.
When conducting a survey, it is of the utmost importance that the participants answering the questions are a representative sample of your target population. To be clearer, imagine you were trying to get an idea of the political views of the average American college student. It would be impossible to survey every single college student in the U.S., so you would pick a smaller sample that represents the overall views of the country’s college students.
If, for instance, you only surveyed students at Ivy League schools, you would only be getting an idea of how students in those schools feel and would largely be ignoring students in small community colleges. To make your sample of participants representative, you would have to administer the survey to each type of college.
Another important part of designing a valid survey is to word questions carefully. The way people respond to questions on surveys heavily depends on how a question is worded, which is referred to as wording effects. For instance, people feel much differently about hearing the phrase ‘revenue enhancers’ than ‘taxes.
‘ Revenue enhancers sounds relatively innocent, while nearly everyone groans upon the mention of taxes. Therefore, when you ask people if they are in favor of more revenue enhancers, they will most likely say yes, whereas if you ask people if they are in favor of higher taxes, they most surely will say no. Researchers are usually aware of the power of wording effects, and organizations with an agenda will word questions in such a manner that people will give the answers that the organization wants.
Yet another potential problem to watch for in survey data is the courtesy bias.
The courtesy bias refers to when a person gives answers to survey questions not based on their true opinions but based on what they think would be socially desirable or acceptable. Of course, when people aren’t providing truthful data, then that affects the validity of the survey data. For example, sexual behavior/thoughts are common survey subjects because that is often the only way to measure such behavior. If a survey contained a question asking people if they have ever had an extramarital affair, it may be likely that they will say no even though they have because extramarital affairs are typically associated with a great deal of social stigma. This has found to be the case even when participants are assured that their answers will be anonymous.
A survey is a method of collecting information or data in which people self-report their own opinions/behaviors in response to questions. They can be administered via pencil and paper, telephone interview, in-person interview, or online. Although in many cases surveys are the only way to get the information that researchers are seeking (such as with sexual behavior), they must be carefully examined and looked at with an eye of skepticism due to well-documented problems that can affect the validity of the data collected.
One such problem involves representative sampling, in which some researchers do not make sure that they collect a sample of participants that truly represents the population they are targeting. Another is wording effects, in which organizations with an agenda word survey questions in such a way that people will provide the answers that the organization wants. Lastly, surveys can be subject to the courtesy bias, in which people do not provide truthful answers but rather answers that they feel will be seen as socially desirable or acceptable.