Best Practices

Security

When conducting a survey or administering forms, it is important that potential respondents are informed about the security measures in place to protect their personal information and survey responses. Accordingly, the ETSU IRB requires researchers to explain these measures during the submision process. When talking about data security, the two most common ways are anonymous or confidential data collection. Read below to see the differences between anonymous and confidential data collection or click here.

Anonymous Data

When conducting an anonymous survey or administrating an anonymous form, NO ONE is able to connect a respondent to his/her answers, IP address, or other metadata that can be collected about a respondent. This includes the researcher, survey/form programmer, IRB, and/or Qualtrics. In Qualtrics, you can do this by publishing an anonymous web link or by selectng anonymize responses in survey settings for an email distribution. If you do want to identify the respondent (confidential data, see below), but you do not want to collect IP addresses, you can enable the anonymize responses feature and use embedded data to pipe in a respondents contact information from your contact list.

Example Anonymity Statement: Your responses to this survey are voluntary and COMPLETELY ANONYMOUS. All identifying information is automatically deleted by Qualtrics and is not accessible to anyone at ETSU at any time. 

Confidential Data


When conducting a confidential survey or administrating a confidential form, only the person collecting the data (and your Qualtrics Brand Administrator, if needed) will have the capability to connect a respondent to his/her answers. This is typically done when answers to a survey or form need to be matched to something else (pre/post tests, evaluations, etc.), when a survey follow-up is needed, when a researcher is required to keep respondent information, or when survey/form completion needs to be tracked. In many cases, identifying information is also stripped after confidential data collection and is missing from the final dataset and/or analysis.

Example Confidentiality Statement: Your responses to this survey are voluntary and COMPLETELY CONFIDENTIAL.The data will only be reported in group form and never for an individual. 

 

Anonymous

Confidential

Advantages

  • You can send out the same anonymous link broadly without having to individualize the link to each respondent. This provides the opportunity to more easily share on social media platforms. Just make sure you engage bot detection via Qualtrics if you are sharing on social media



  • Respondents are more likely to be honest in their responses

  • It is likely to reduce the burden of answering questions on sensitive topics

  • Possible higher response rates

  • Lower risk of social desirability bias 

  • You can pre-load demographic data (if available) and reduce the number of questions in your survey

  • You will be able to breakout your data by key demographics

  • You will be able to follow up with respndenst to address feedback and issues if necessary

  • Survey questions and blocks can be targeted or displayed to specific demographic groups

  • You can track who has responded

  • You can troubleshoot if a respondent has an issue with their survey

Disadvantages

  • It is harder to troubleshoot if a respondent has an issue with their survey

  • When using an anonymous link, you cannot track who has taken the survey or how many times they have taken it

  • It is harder to identify bots and spam responses

  • Possible lower response rates

  • Respondents are likely to be less honest for fear of retribution

  • Higher risk of social desirability bias

Email Invitations

When distributing your survey/form via email, the language you use is very important. After all, the email will be the basis of your first impression to the respondent. Also, in most cases as a researcher, the ETSU IRB will require you to include an Informed Consent Document (ICD) in the email or at the beginning of the survey. Often times, the language in your invitation and ICD document will overlap. Click here to download ETSU's ICD Template.

All survey invitations should include:

  • Project Overview: Tell the respondent who you are and what topics will be covered in the survey questions.

  • Survey Importance: Tell the respondent why the survey is important.

  • Estimated Completion Time: You will need to test your survey and see how long it takes to complete it. Example: This survey should take 5-10 minutes, depending on your answers. It is important to say depending on your answers because of skip patterns, etc. that will make the time different for different respondents

  • Security Statement: IRB requires researchers to disclose if the survey is anonymous or confidential, and the steps taken to ensure data security.

For further reading on survey best practices, see Internet, Phone, Mail, and Mixed-Mode Surveys: The Tailored Design Method 4th Edition

Share