Survey Best Practice

This guide is to assist you with the survey creation process. It is not designed to make you an expert in survey content but to provide tips and suggestions for best practices in data collection. Designing an effective survey involves several key practices to ensure the collection of reliable, valid, and actionable data. 

Here are some best practices for survey design:

Define the purpose of the survey. Before crafting your survey questions, clearly define their purpose. Know exactly what you want to achieve. For example, you may want to assess patrons’ satisfaction with the Call Center. A sample goal could be, “Demonstrate comprehensive high-quality customer service through accurate, friendly service and timely responses.” Therefore, you might ask questions like:

  • What is going well?
  • What areas need improvement?
  • What suggestions or ideas do you have for improving services?
  • Can you provide feedback on areas or individuals that deserve accolades?

Ensure each question adds value. Each question should directly relate to your research goals. Avoid irrelevant questions.

  • For instance, omit demographic information if it doesn’t contribute to the survey’s purpose.

Keep the survey brief. Make sure questions are clear and concise. Identify the data you need first, then write your questions.

  • Ensure the survey follows a logical order and is quick to complete. Aim for 10-15 minutes.

Ask one question at a time. Avoid combining multiple inquiries into one question, as this can cause confusion and inaccurate responses.

  • For example, avoid questions like, “Is your resident assistant available and visible regularly?” if these are separate issues.

Avoid leading and biased questions. Provide only the necessary wording for respondents to give informed answers. Focus on the respondent’s opinions, not your own. 

  • For example, avoid phrasing that assumes a certain view, such as, “Weather Now News is considered the most accurate. How often do you find their forecasts accurate?” 

Use simple language. Ensure questions are easy to read and understand. Use simple sentences and straightforward answer choices. Simplicity is key.

Ease respondents into the survey. Start with easy questions to engage respondents, then move to more complex or thought-provoking ones.

  • For example, avoid placing sensitive questions at the beginning, as they may deter participation.

Consider incentives to boost responses.

  • For example, offer incentives like gift cards or merchandise to encourage participation. Ensure your incentives are valuable to your target population.

Pre-test your survey no matter its length or simplicity, before full distribution.

  • For example, share it with at least two people to identify and fix any issues.

By following these best practices, you can create effective surveys that yield reliable and actionable data.

Suggested Survey Question Types

We strongly encourage the use of the following question types:

Single select- close ended; all respondents would have only one answer to choose from

  • Example: Are you currently enrolled at Delaware State University?  Yes / No

Drop down - extensive list, i.e. state of residency

  • Example: What is your state of residency?
    • Alabama,
    • Alaska,
    • Arizona, etc.

Likert Scale - responses to these questions range from one end of the spectrum to the other, gauge satisfaction; when you want to capture statistics that will provide rich data

  • Example: (Scale: 5 = strongly agree, 4 = agree, 3 = neutral, 2 = disagree, 1 = strongly disagree) 
    • The instructor spoke clearly.   5 4 3 2 1 
    • I turned in homework on time.  5 4 3 2 1  

Multi-select – these questions consist of a question followed by a list of answers to choose from; more than one response could apply to respondents. Consider always including “other” as an option since there isn’t a universally applicable response.

  • Example: What was your primary reason for choosing to live on campus? (check all that apply)
    •  Convenient location 
    •  Lower cost  
    •  Wanted to meet people 
    •  Safety and security 
    •  My parents wanted me to
    •  Other, please specify:

Text box – open ended; respondent feedback. The amount of space allotted for a response should provide an indication of the expected response length.  

  • Example: What did you learn from this activity?

Suggestions for Improving Survey Accessibility:

Page Titles:

  • Ensure each survey page has a clear title. If subsequent pages have similar content, consider adding “continued” to the previous page’s title to signify progression.

Image Descriptions:

  • Provide brief, descriptive image explanations when adding images to questions or answers. Note that for Baseline surveys, images must have public links.

Hyperlinked Text:

  • Include descriptive text when hyperlinking. Specify that the link will open in a new window for clarity and set the target to open in a new window.

Question Types:

  • Matrix and dual matrix question types may pose challenges for accessibility software. Prefer single-select questions whenever possible to ensure compatibility.

Question Numbers:

  • Keep question numbers visible to provide respondents with context about their progress in the survey.

Progress Bar:

  • Maintain the visibility of the progress bar to give respondents a sense of their advancement through the survey.