How to plan it?
While planning the survey research, it is important to:
- Identify the survey’s objectives in association with the general research question(s). Defining the objectives will help determine what kinds of information should be obtained (i.e., specific topics to cover) and from whom this information should be obtained (i.e., population of interest). One possible way of identifying important objectives (e.g., topics and hypotheses) is by conducting first qualitative research (e.g., interviews, focus groups).
- Conduct a desk research / literature review.This will help accomplish three important goals: 1) justify theory underlying the questions to be answered; 2) define variables and terms; 3) identify existing instruments that could be used or adapted.
- Choose a survey design. The survey’s objectives, the population of interest and the resources available (e.g., money, human resources, time) should be considered in selecting the appropriate design.
- Decide on sampling and determine the sample size. This decision should depend on survey’s objectives, the nature of the population being studied, the complexity of the survey design and the resources available. If the main goal is to generalize the findings of the research to the population of interest, one should use a probabilistic sampling method (e.g., simple random, stratified random, cluster) in order to ensure the representativeness of the target population. However, if generalization is not a major goal, non-probabilistic sampling methods (such as quota sampling or convenience sampling) may be more useful given that they are easier to do.
- Prepare the survey instrument. First, an identification of existing and appropriate instruments that could be used to cover the survey objectives should be performed. Questions from existing instruments should be critically revised and adapted if necessary. If questions from existing instruments do not cover some of the survey objectives, new items should be created.
- Re-examine items and prepare the instrument layout. The survey must begin with an introduction telling respondents what the questionnaire is about, what they are asked to do and why. Ensure confidentiality of answers. For self-administered surveys (e.g., mail or internet) it is also important to instruct respondents about what to do with the completed questionnaire. Next, the items should be presented with clear identification of transitions (e.g., change of topic) and with detailed instructions about how respondents can answer. Items should be organized from more general to more specific. Overly personal or sensitive questions should not appear in the beginning. All items should be evaluated. The questionnaire should end with a note of thanks and instructions regarding possible follow-ups.
- Pilot-test the instrument. Before conducting the full-survey, testing should be done to find out if respondents can understand the items, if they can perform the tasks that it requires, the time needed for administration and if the interviewers can and will read it as worded.
- Revise the instrument in accord with pilot-test results and final editing. If new items are added, it is advised to pilot-test them. Ensure that the final version of the questionnaire includes clear instructions to the person distributing it and gives an estimate of the time required by the average respondent to complete the questionnaire.