Formatting the Survey
Keep it as short as possible.
Writing a survey requires finding the ideal balance between being concise and comprehensive. Each survey must be long enough to capture all of the content to be measured, but not so long that respondents are intimidated or frustrated by the length and complete it with reduced accuracy or not at all. Short surveys are more likely to be returned as they place less of a demand on the respondents. Consider the value of the survey information against the time the respondent must spend on the survey. Ideally, aim for no more than two sides of one sheet of paper when printed (using a readable font).
Give the survey a title which concisely reflects the content of the survey instrument and who is conducting it.
Write an introduction to the survey.
The introduction should include the following information:
• the purpose of the survey • whether or not participation is voluntary • the confidentiality and/or anonymity of survey responses (see below) • how survey findings will be used (including with whom they will be shared and when) • the approximate time it will take to complete the survey • the name and phone number or e-mail address of a contact person for those who have questions or concerns regarding the survey
Explain the anonymity and/or confidentiality of survey responses in the survey's introduction.
Anonymous data do not include names, addresses, or any personal information that would make it possible to associate a response with any given individual. Data that are confidential contain information that may identify an individual respondent, however only a limited number of people (i.e., the researchers) have access to individual responses. Results of confidential surveys are reported in aggregate in order to respect the privacy of the respondents. Files containing individual identifiers must be stored with considerable attention to data security and access. There are significant advantages to collecting identifiers, including the ability to do "pre and post" studies through linked data files and to follow-up on non-responders to increase response rate.
Provide directions that describe exactly what respondents are to do.
Organize questions in the survey according to the topic area.
When possible, use headings to label each different category of questions. However, with shorter surveys, use of headings should be limited so as not to clutter the page and detract from the survey content.
Group items with the same response scale together.
Introduce the scale once for a group of items (immediately before the items). In addition, for electronic surveys, ensure that the scale can always be seen by the respondent.
Do not break questions between pages.
Position questions for a high probability of completion.
Start with an easy-to-answer question that is clearly related to the subject matter in order to increase comfort and confidence with the survey. Put sensitive or challenging questions at the end of the survey. Respondents are more likely to answer more personal and/or difficult questions once they have already made an investment in filling out the survey. In general, "behavior" questions are more sensitive and personal than "knowledge" questions. Similarly, demographic items should be placed at the end of the survey, especially when asking for potentially sensitive information.
Provide an open-ended question at the very end of the survey so respondents can comment on anything you did or did not ask them.
This space allows respondents an opportunity to share their comments freely and without restrictions. While many respondents will not take the time to write additional comments, those who do usually have strong opinions that they feel compelled to share.
Thank respondents for their participation at the end of the survey.
Proofread for spelling and grammar.