Wording Surveys Well Makes Them More Effective (Part 3)

Wording Surveys Well Makes Them More Effective (Part 3)

by Arden Harper    May 13, 2020 3:10 pm  
Written by Arden Harper – 4 minute read

This is part 3 of this series. Start on part 1 here

Part 2 discussed the importance of making questions objective instead of subjective. Part 3 talks about choosing an appropriate answer type. Answer types will depend on what type of survey is being conducted. An opinion survey, asking for feedback, will likely have a different type of answer. 

There are a lot of ways that survey wording can be improved, so we’ll break it down into different categories: asking direct questions, making questions objective, providing an appropriate type of response option, asking for one piece of information per question, and length. 

Choosing an Appropriate Answer Type

As if there weren’t enough options for questions themselves, there’s also the issue of types of answers the user can choose from. There are several types of answer formats, all with different merits. 

It’s important to understand that there are two major categories of survey, in my mind*: opinion-seeking & fact-seeking. An opinion-seeking survey is likely to be used after a product or service is used to gather feedback. An example would be a survey sent by a company after buying their product or service, asking for a review or how to improve. Fact-seeking surveys are looking for data, often for statistics. An example of this is the US Census. 

*please note that these are not official terms, but ones that I felt were descriptive and specific enough

The difference between these two questions isn’t the question itself, but is in the answers. N/A (or non-applicable) can show up in a variety of ways, especially as more and more people work freelance. We’ve had several clients bring us survey results where we were not part of the question-making process. Often they will express concern that their “N/A” responses are much higher than they anticipated, and they don’t know how those respondents should be classified. Carefully planning all possible answers ahead of time can affect the data is important in choosing the content within the possible answers. 

A blend of fact-seeking and opinion-seeking can be used, but the data type will differ from using all of one type or all of the other. For example, a survey will almost always look for a fact (usually a demographic such as name, age, address, and phone number), but will rarely, if ever, use purely opinion-seeking questions. Attributes like names and age are not subject to opinion. 

An example of a fact-based answer that is often not inclusive enough is questions of gender. Forcing users to choose between a binary of male or female doesn’t encapsulate all possible responses. 

Overall, objectivity and subjectivity are a precarious balance to achieve, but the takeaway is that the less emotionally-loaded language is included, the better. Keeping language objective makes for better data collection and a lowered likelihood of abandonment of the survey itself.

Boxplot has written effective surveys for many clients in the past, and provided in-depth analysis of the results. 

Email us today: info@boxplotanalytics.com


Need help applying these concepts to your data?

Chat with us about options.

Schedule a Meeting   


Continue to make data-driven decisions.

Sign up for our email guides that contains relevant tips, software tricks, and news from the data world.

*We never spam you or sell your information.

* indicates required

"Wording Surveys Well Makes Them More Effective (Part 2)"