Conducting a Survey
Conducting a successful survey requires knowledge in several different areas. Researchers need to know how ensure security and confidentiality, craft email invitations, and achieve high response rates.
Researchers have a professional and legal obligation to inform respondents about the security of their personal information and survey responses. Researchers must describe the ways in which the respondent’s identity and responses will be protected, be they anonymous or confidential. Below are the differences between anonymity and confidentiality and an example statement, demonstrating how you can assure your respondents their answers are secure.
In the case of anonymous surveys, no one, including the researcher, is able to connect a respondent to his/her answers.
Example Anonymity Statement: We will not be tracking any identifying information. Your responses are completely anonymous.
When conducting confidential surveys, the researcher may have the capability to match up responses. This can be done at the University of Minnesota using an Internet ID. However, answers will be decoupled and any information that could identify a respondent will be removed by the researcher or an independent third party.
Example Confidentiality Statement: You may be assured of complete confidentiality. Your email address will be stored only to track survey completion. The data will be reported only in the aggregate and no individual will be identified.
As the first point of contact, the invitation plays an important role in determining whether someone will respond to a survey. When inviting individuals to complete a survey, there are some things to include that may yield higher response rates.
All survey invitations should include:
- General Information: Let the individual know who is conducting the survey and what topics will be covered in the survey questions.
- Salience of the Survey Topic: Tell the respondent why the survey is important.
- Estimated Time of Completion: Studies have shown that participants who were told a survey would take a short time to complete (8 – 10 minutes) had a higher response rate than those who were told it would take longer (20 minutes). Include the estimated completion time if the time is short, but if the completion time is long, it may be better to exclude it from the invitation.
- Security Measures: To be in accordance with the law, researchers must note the measures, be they anonymous or confidential, taken to protect the respondent’s identity and survey responses.
Optional information to include in an invitation:
- Statement of Scarcity. Tell the individual that he or she was selected as a part of a small group to participate in the survey.
- Incentives. Promising an incentive of any kind generates higher response rates than offering none.
- Personalization. There is both positive and negative research in the ways of personalizing survey invitations. On one hand, recipients feel more valued and respected, which increases the probability that they will comply with the researcher’s request. On the other hand, the perceived level of anonymity and privacy decreases with personalization.
After your survey has been written, you will need to draft respondents. When you are asking individuals for their participation, there are a few considerations that will increase your response rates.
Here is a list of those considerations:
- Consider the survey length. The shorter the survey, the more likely a respondent is to answer your survey request.
- Consider timing. The time a survey is sent out can greatly affect whether or not a participant will respond. Identify when respondents are most likely to respond to a survey request, and send out the survey at that time.
- Be credible. Send out surveys from addresses that participants will recognize and trust.
- Include invitations. Well-written invitations should be able to persuade respondents to complete a survey..
- Keep it relevant. Solicit responses from specific populations who have a reason to be interested and explain the purpose of the survey thoroughly in the invitation.
- Use incentives. Promising an incentive of any kind generates higher response rates than offering none.
- Contact respondents more than once. Studies have suggested that sending out a pre-notification message, the survey, and a reminder message at incremental times helps achieve higher response rates.