Survey

What is a Survey?

“The customer’s perception is your reality.” It’s important to listen to what the customer has to say, and a survey is a great way to do so. This method allows you to collect quantitative and/or qualitative data on attitudes, beliefs, opinions, preferences, motivations, and self-reported behavior through a series of questions.

Why Conduct a Survey?

Surveying is a versatile technique that allows you to reach large groups of people, and can be easily combined with other methods, like observation, focus groups, and usability tests. With a variety of question types and formats to choose from, you can create your survey so that it will yield the type of data that you want. Additionally, there is flexibility in the vehicle with which you choose to deliver the survey. You may choose to use phone, web, email, or paper.

How to Conduct a Survey?

Define Objectives: Identify the population, the specific information you want to gather, how you will gather it, and the type of analysis you plan to do.
Determine if You Will Need PRA Approval: Begin the PRA process if necessary.
Determine Survey Format: Surveys can include both closed-ended and open-ended questions. You can include one or both, but make sure the types of questions you use will provide you with the data you need (nominal, ordinal, interval, or ratio).
- Closed-ended questions have a predefined list of answer options and include, but are not limited to multiple choice, yes/no, numeric scaled, and image choice questions.
- Open-ended questions allow the respondent to provide an answer in their own words and help uncover motivations.
Write the Survey: This process takes several revisions, reviews, and pretests to get it just right.
Conduct the Survey: Conduct the survey with your selected survey method and sample. The survey process often takes longer than expected and you may have to send several reminders to receive the response rate that you desire.
Analyze the Data: Your synthesis method will vary based on the type of data you collected: qualitative or quantitative. You may want to use an Affinity Diagram to analyze qualitative data. You could also code data or make text answers countable to pull out bigger trends from non-numeric responses. Report quantitative data in the way that makes the most sense (graphs, charts, key takeaways, etc.)