Instant Census Blog

Back to all blog posts

Survey Design: The First Steps

By Chris McCarthy on August 6, 2015

Designing a survey is no small task, but there’s no need to be intimidated by the challenge.

In this week’s post, we’ll cover the beginning steps to designing a survey. This will include determining a survey topic, creating a panel based on your specific demographics, and compiling survey questions. Next week, we’ll move further into the process, covering topics such as pretesting questionnaires, interview and data collection methods, and how to analyze your data.

To begin the process of survey design, it’s important to ask yourself the following questions:


What are you looking to learn?

If you’re interested in conducting a survey, you most likely already have an area of study in mind. If not, many studies are conducted on the following topics:

  • Consumer insights / Customer feedback

  • Event feedback

  • Education

  • Human Resources

  • Politics/Political Campaigns

  • Health/Medical related research (i.e. Flu Studies)

  • Potential markets for a new product

This topic should guide all the other decisions you make about your survey. Your demographics, questions, and methods should all be designed to elicit precise information about the area of study, and nothing else.

Who do you want to interview (i.e. panel demographics)?

Your panel’s demographics will depend on your survey’s topic and target population. If your survey targets a specific audience, it’s important to enlist a panel that meets your demographic needs to ensure your data’s accuracy. Target audiences are based on a variety of traits, often including age, location, education, ethnicity, religious affiliation, etc.


Keep in mind, your panel should be informed enough and have the expertise to answer your questions truthfully and accurately. It is also important to gauge the willingness of your demographic to answer. You can only ask so many questions before people stop responding, and it is important to remember what that limit is for your particular demographic when making other design decisions about your survey.

What questions need to be asked?

There are two styles of questions that can be asked: open-ended or closed-ended questions. Open-ended questions prompt respondents to formulate their own, custom answers, while closed-ended questions require respondents to choose answers from a list of options, or answer “yes” or “no.”


Determining what style of question to ask depends on your survey topic and style. Open-ended questions are best for longer surveys as they are more exploratory. They also allow researchers to gain audience insights and opinions they may not have considered when developing the survey. Be sure not to overuse them, however. Open-ended questions can often be more burdensome for the respondent. They require more proactive thinking on the part of the respondent, which can create higher response fatigue. Furthermore, Open-ended questions can often be too open-ended. If a question requires multiple sentences to answer correctly, instead of a single word or phrase, this will even further increase the response fatigue induced in the respondent.

Closed-ended questions do not allow for respondents to provide unique or unexpected answers. As they are bound to specific answers, such as “Yes,” “No”, or multiple choice selections, researchers expect to receive conclusive, quantifiable results. It is important when writing such questions that you manage the careful balance in the number of answers. Too many, and respondents can lose focus or become unable to properly respond; too few, and you might not be able to adequately reflect some responses. You must also think about presentation of options, and how that might unduly influence a respondent.

At Instant Census, we’re here to assist with your survey design process. Our team of survey design experts are on hand to transform your survey into an adaptive study customized to fit your needs.

Our advanced text message survey software and capable of running complex custom question logic and answer parsing; sending one-time surveys, surveys on particular events, and scheduled recurring surveys; and corresponding survey conversations with thousands of respondents simultaneously. This means our users have the freedom to ask open and/or closed ended questions and deploy survey questions on a schedule that best fits their needs.

Be on the lookout for next week’s post as we’ll delve further into the survey design process by discussing interview and data collection methods, the importance of pretesting questionnaires, and how to analyze your data.

If you’re interested in learning more about Instant Census, our text message surveys and rapid response rates, get in touch today!

Want to learn more about our capabilities and what we do? Get in touch!

Whatever your target audience, discover how easy it is to start connecting and get results.