Supported by:
Brown School of Public Health   The Rockefeller Foundation

launching a
public health survey

When getting started on an online survey, the most important steps are to identify your survey objectives and choose a target audience. Are you interested in learning more about a specific audience’s barriers to COVID-19 testing? Is your focus on the entire community or specific populations? Clearly defining what you want to learn, and from whom you want to learn, sets the stage for how you develop and administer a successful survey.

Next to the questions you ask, how you structure your survey is also important. Consider elements such as format, layout, length, and organization. For example, surveys that are long and poorly organized may discourage the audience and result in fewer people completing them.

The vast majority of work on a survey happens before it goes live. Once it is out in the world and results come in, you don’t want to make changes that would impact the validity of your results.

Here is our step-by-step guide to creating and implementing an online survey:

online surveys:
a step-by-step guide


Define your Goal
Define the purpose of your survey in clear, unambiguous terms. It sets the direction for everything you do. Key questions to consider:

  • What question(s) am I trying to answer with this survey?

  • What will survey results help me decide or accomplish?

  • What gaps in my understanding of a given situation or experience will this survey fill?


Identify Your Audience
To collect data that meets your goal and answers your questions, you’ll need to identify the target audiences from whom you’d like to learn. Are you targeting all adults within a specific zip code? Or only people who are at higher risk from COVID-19? Do some basic research on your audience’s composition and any key demographic characteristics so that as you begin collecting responses, you’ll know if your preliminary data is representative of your target audience.

In this step, you should also confirm that an online survey is the best method to gather insights from specific individuals or communities. While most online survey programs offer paid services to reliably recruit people from specific locations, ages and genders, it’s harder to control for race, ethnicity, socio-economic status, and COVID-19 risk factors. If relevant, these questions can be asked in the form of screening questions (see below).

Key questions to consider:

  • Who are the people I need to reach and gather information from?

  • How many people do I need to hear from to feel confident in the findings?

  • What are the demographic characteristics of those people? What’s their gender, race or ethnicity, age range, employment status, and income level? Do they live in a particular location or zip code?

  • Do they have access to the internet, have an email address, and will they be easy to reach with an online survey?

  • Screening questions can be used to ensure you’re surveying people who have something in common besides demographics. Example screening questions include, “Have you ever been tested for COVID-19?” or “Have you ever been diagnosed with COVID-19?

  • Where should I promote the survey in order to reach these individuals? Are there partners who can help me share it?


Organize the Survey
Dividing your survey into sections helps you and the survey taker. Always provide an introduction or welcome message and inform survey takers about estimated completion time, generally how the data will be used, and any privacy and reporting protocols. We suggest that Health Departments run anonymous and de-identified surveys, meaning that either identifying information is not collected or respondents’ answers are stored separately from any identifying information. In most cases, you will not need to ask for any contact information in your survey, though sometimes meta-data like their IP addresses are collected automatically. If collected, identifying information should not be shared publicly in any way.

When setting up the survey flow, group questions about a specific topic or theme together. Ask screening questions at the beginning of the survey or at the start of each section to make it easy to filter out people who don’t qualify. Group demographic questions (e.g. gender, age, race, etc.) together and include them at the end of the survey.


Select a Survey Platform
Select a survey tool that meets your budget (if you have one) and enables you to efficiently collect responses and analyze results.

Key questions to consider:

  • How many questions do I need to ask?

  • How many people do I want to take the survey?

  • What type of questions am I asking?

  • What is my budget?

  • What’s the fewest number of questions I can ask to meet my primary objective?

  • Do I want the survey tool to find people to take the survey or do I want to recruit on my own?


Define Your Questions
What questions you ask should tie back to your initial goals and the questions you want answered. It can be helpful to brainstorm all possible questions and list the insights you’ll glean from each, as well as any decisions you’ll be able to make based on the data. Address any question gaps and pare down the final question list ensure your survey isn’t too long or repetitive. Ideally, online surveys should take between 5 to 15 minutes to complete. The longer the survey, the more likely someone won’t start it or finish it.


Write the Questions
Survey questions should be written carefully to avoid confusing the survey taker and to ensure you get the most useful data. The good news is that most online survey platforms offer a library of question templates and examples, and provide some guidance on what questions to use when. Here’s a helpful resource from Humans of Data.


Test Your Survey
It may sound obvious but testing survey questions before they are released is crucial for many reasons, most importantly to avoid having to make changes after it has launched. When testing, make sure the questions follow a logical flow and their format makes sense. Define any technical terms and spell out acronyms. Time how long the survey takes to complete, and share this information with participants in the survey’s introduction and any recruitment materials. Most survey platforms allow you to share the draft survey electronically with others. Ask your colleagues and partners to test the survey for flow, format, readability, and length. Ask testers to include the word “TEST” in any of their responses so you can filter them out in analysis.


Launch the Survey
Launching (or fielding) the survey means different things depending on how you’re planning to share the survey, recruit participants, and collect responses. Once the survey is built, most online survey platforms provide a customizable link that you can include in a recruitment email, on social media, or a poster. If you’re using this option, it’s up to you to find and recruit survey respondents. When promoting the survey via email, social media, or postering, let people know there’s a set period of time you’ll be collecting responses to create a sense of urgency. If recruiting online, remind people to complete the survey as you get closer to the deadline.

If you’re paying a survey platform to recruit participants on your behalf, you will need to take an extra step and identify the key characteristics of the people you’d like to target—i.e., number of respondents, age, gender, location, etc.


Analyze the Survey Results
Most survey programs allow you to see completed responses and initial trends in real time, but it’s important to avoid drawing conclusions until you’ve reached your goal number of responses (also called meeting your quota). Survey programs also let you view the data in a variety of ways, either as a list of individual responses or aggregated, in the form of a chart or graph. When the survey is complete, you can also export the raw data as an Excel file and conduct your own analyses. This is especially useful when doing more advanced analysis and cross tabulations.

To Top

      ©2021 | Brown University School of Public Health