Survey Research

The term survey research refers to a family of non-experimental research methods using the collection of self-report data to assist the researcher in more fully understanding attitudes of participants regarding a topic of interest. Unlike experimental studies, survey studies attempt to avoid interventions with participants so as to capture reality as it exists at a given point in time. Survey research is important in the field of curriculum studies and is adaptable across a wide range of research problems. For example, survey data can be useful in understanding attitudes of educators, students, and others regarding new curricula, teaching practices, and curricular reforms.

Surveys may include collection of quantitative, qualitative, or mixed mode data, and data may be collected directly (e.g., via interview) or indirectly (e.g., via mail) from either individuals or groups. Highly sensitive topics (e.g., participant involvement in illegal activities) are often best addressed using indirect anonymous surveying methods. Survey data collection typically requires three elements: a survey tool (items), a sample of respondents (informants), and an interviewer. If surveys are administered indirectly (e.g., via mail, e-mail, or a Web site), no interviewer is needed. Careful development of survey tools is essential. Survey items may be open- or closed-ended depending on the type of responses sought. Items should be written to evoke the type of responses desired without being too leading. Moreover, care should be taken to sequence the questions so that more positive, higher-interest items come before negative or more tedious questions. It is also important to keep a written questionnaire or interview to a reasonable length to avoid participants' loss of interest or failure to complete the survey.

Surveys may be administered in a variety of ways. Direct methods, such as personally administered pencil and paper surveys, face-to-face interviews, and telephone polling, provide the researcher with opportunities for personal interaction with the participants as well as the ability to ask clarifying follow-up questions and to identify problems with administration of the survey. Focus groups allow the researcher to assess the views of a number of participants simultaneously, and the social interaction provided in this setting may increase the quality of the data gathered. Advances in technology during the last two decades have yielded new methods (e.g., personal response system “clickers”) for gathering numeric data from a live group of participants. Indirect methods (e.g., mail, [Page 832]e-mail, Internet surveys) allow surveys to be distributed to large samples at a relatively low cost; however, response rates tend to be low (often around 20%), some respondents fail to answer all of the questions, and it is difficult to determine who is actually responding to the surveys and whether there are problems with survey administration. When response rates are low, follow-up reminders can sometimes prompt nonrespondents to complete mail or Internet surveys.

Surveys may be used to provide descriptive data relative to a phenomenon of interest or to investigate relationships between variables. Survey data lend themselves to a variety of research designs and data analytic procedures depending of the type of questions asked. Continuous response formats (e.g., Likert scales, semantic differentials) generate numeric data that can be compiled and summarized in a variety of ways using descriptive statistics. Cross-tabulations may be used to compare survey responses across demographic subsets and allow the researcher, if desired, to test causal comparative or correlational inferences. Interviews and free-response written questions yield verbal data that can be subjected to content analysis or coded for use in ethnographic or grounded theory studies.

...