Introduction to research methods in psychology/IRMP102/Survey Research
A survey is a way of gathering information by the use of old-fashioned questionnaires, verbally, or the Internet. Surveys consist of administering a series of predetermined questions to a group of individuals. The individuals respond to the survey either in an oral or written form, and the researcher then summarizes the responses by averaging across the individual responses. Compared to a study conducted in a psychology laboratory, surveys can reach a much larger number of participants at a much lower cost.
When an attempt is made to include every member of a specified population in the study, the survey is called a census. More often, however, data are collected from a subset of the population. The subset is called a sample, and the investigator conducting a survey usually chooses people for the sample by a method intended to ensure that the distribution of important characteristics is the same in the sample as it is in the population (you will learn more about these methods in the section "Conducting Surveys" in Chapter 9 of your textbook). To the extent that this goal is achieved, the investigator will be able to generalize the survey results from the sample that was studied to the wider population.
Advantages of survey research
- Surveys can be done relatively quickly.
- When respondents’ anonymity is ensured, surveys are effective in studying embarrassing, highly personal topics.
- Large sample sizes can be obtained relatively inexpensively.
- Surveys that are administered online can be configured to counterbalance the order of questions, or even to tailor subsequent questions based on the respondent's earlier responses.
- Survey results describe relationships between variables that may help us confirm or reject theories and suggest new experiments.
Limitations of survey research
- A respondent may not understand the questions asked. For example, an illiterate person may be mailed a written survey, or a person may not understand the vocabulary of an interviewer but be too embarrassed to ask for clarification.
- The order of the questions may affect respondents’ answers. With mail surveys, there is no way to ensure that the respondent completes the survey in the presented order.
- Results are only as good as the sample that completes the survey. Some people have no chance of being included in a survey. For example, a telephone survey would not include homeless people without phones. Those people who do not complete the survey may differ substantially from those who do. Therefore, the sample may well be biased.
- People’s responses are not based only on their opinions. People’s responses are influenced by other factors, including their willingness to respond “yes” in order to make themselves look good, as well as who it is that is asking them the survey questions. These extraneous influences are called "Demand Characteristics" and they refer to the possibility that participants who form an interpretation of the researcher's purpose may, consciously or non-consciously, change their behaviour to fit that interpretation.
- The format of the question may also influence responding. For example, whether the questions require selecting from listed options or require that the respondent develop his or her own options.
