Surveys are a great way to gather feedback from your customers, employees, or other stakeholders. But success depends on the ability to ask a good survey question.
A well-crafted survey question can help you get the information you need to make informed decisions and improve customer satisfaction.
However, poorly written questions can lead to inaccurate results that don’t tell you anything useful. So how do you write effective survey questions? Let’s dive in.
The Types of Survey Questions
- Closed-ended questions
- Open-ended questions
- Multiple choice questions
- True or false questions
- Rating or Scale questions
- Ranking questions
Closed-ended Questions
Closed-ended questions require respondents to select from a list of predetermined answers. They are useful for collecting quantitative data and quickly gathering information from a large group.
These questions should focus on specific topics, be clearly worded, and have options that cover all possible responses.
Open-ended Questions
Open-ended questions are useful for gathering qualitative data and getting more detailed feedback. These questions allow respondents to provide their own responses without being limited to predetermined answers.
Open-ended survey questions should be framed in a way that encourages participants to share as much information as possible, so they should include clear instructions and avoid leading or biased language.
Multiple Choice Questions
Multiple-choice questions provide respondents with a list of options to choose from.
They are useful for gathering quantitative data, but they should be carefully crafted to avoid bias and the temptation for participants to select the most appealing option without considering all the alternatives.
Multiple-choice questions should include clear instructions and cover all possible answers.
True or false Questions
True or false questions are useful for testing knowledge and quickly gathering data. However, they should be used sparingly as they can limit respondents’ ability to provide more detailed information.
When using true or false questions, make sure the answer options are clear and unambiguous.
Rating & Scale Questions
Rating and scale questions allow respondents to rate their answers on a scale.
This type of question is useful for collecting more detailed quantitative data, but it should be used in conjunction with other survey methods to get an accurate picture of participants’ opinions.
When using rating or scale questions, make sure the options are clearly worded and cover the full spectrum of possible answers.
Ranking Questions
Ranking questions allow respondents to rank a list of items or ideas in order of preference. This type of question is useful for exploring preferences, but it should be used in conjunction with other survey methods to get a complete picture of the data.
Ranking questions should include clear instructions and give participants enough options to accurately express their preferences.
Tips for Creating Good Survey Questions
No matter what type of survey question you’re asking, there are a few key tips to keep in mind for crafting effective survey questions:
1. A Mix of Open and Closed-ended Questions
It’s important to use a mix of open and closed-ended questions in order to get both quantitative and qualitative data.
Closed-ended questions are useful for quickly gathering data, while open-ended questions allow you to get more detailed feedback.
A mix of these questions or a survey with entirely closed-ended questions and a subsequent answer box for anyone who wants to elaborate will help to keep the survey short and sweet but impactful.
The last thing you want to do it create an hour-long survey that forces participants to write a minimum of 50 words for every question, particularly when the survey is voluntary.
Be sure to switch between these two main types but also include a different question asking a variety like ranking or true and false, to avoid monotony.
2. Keep Questions Clear and Concise
When writing survey questions, make sure they are clear and concise. If a question is too long or wordy, respondents may skip it or provide an inaccurate answer.
Not only will they skip it if they overtly don’t understand it, but they may also mistakingly believe they understand the question and then supply an inaccurate answer.
It’s best to avoid any fancy jargon or flowery language that can impact participants of all backgrounds in their participation.
For example, if you’re a grocery store sending out a general survey and your customer base varies in demographic across languages spoken, sayings only native English speakers understand may be lost in translation.
3. Avoid Leading Questions
Leading questions can influence how participants respond, which can skew the results of your survey. Make sure your questions are neutral and don’t suggest a particular outcome.
Leading answers – as in there is a clear desired answer the survey giver is seeking – can also affect the honesty of survey responses. The point of a survey is to garner raw, inclusive data to help dictate the direction of your business.
Swaying the results will only hurt your business in the long run and negate the purpose of the survey. Pointed language such as which of our products is the best skips over the fact that the participant may not like or have tried any of the products in question.
Another example of a leading question is a simple yes or no sent out by a store following a visit with the allure of a prize for the survey response, “did you enjoy your visit? yes or no.” the obvious desired answer here is yes.
4. Include a Demographic Question
Demographic questions can provide valuable insight into who is responding to your survey and help you better understand their responses. It’s important to ask demographic questions in a respectful way so as not to offend participants.
Some useful demographic questions include age, gender, location, education level, occupation, industry, etc. This information will help you determine if certain demographic groups are responding more or less favorably to certain questions, which can inform the decisions you make.
Depending on who you ask, some recommend those demographic questions come at the beginning of the survey to ensure they’re answered, while others think they’re best reserved for an easy end-of-the-survey sendoff.
Regardless, you’ll want to ensure that answers in the demographic section offer exclusive choices just like the rest of the survey.
5. Avoid Non-mutually Exclusive Questions
Non-mutually exclusive questions are those that offer options that overlap. For example, if you have a question like “What is your favorite color?” and include the options of red, blue, green, and yellow; this wouldn’t be mutually exclusive because participants could select multiple colors.
This means you won’t be able to accurately analyze the data later on. To avoid this, you can try something like “What is your favorite color: red, blue, green, or yellow?” and make sure they only select one option.
Similarly, if when you ask a demographic question like age, you say choose one: 25-30 years old or 30-35 years old, the 30 years old who takes the survey will not know which answer is the right one.
6. Carefully Consider Question Order
The order in which you present your survey questions can also have an effect on the answers you receive. This is called the order effect.
If a survey starts off with two or three negative questions and then transitions to positive ones, respondents may provide less favorable answers to the initial questions because of the negativity.
Similarly, one question might influence how a respondent perceives and answers later ones. In other words, the order you present your questions matters.
To avoid this, it’s best to randomly order your questions and not include any priming questions or transitions between sections.
7. Avoid Asking Two Questions in One
When creating survey questions, make sure that each one is asking only one thing. Otherwise, you may end up with inaccurate data as respondents are more likely to interpret the question differently than intended.
Additionally, loading a single question with multiple components makes it harder for participants to understand and answer accurately.
For example, if a computer company wanted to scope out the market and they posed the question, “Are you planning to purchase a PC in the next six months, if so, which one?” and the answers were exclusively different computer types, this excludes the answer of whether the participant is planning to pick out a computer in the near future in the first place and assumes they would pick one of the types listed.
If the computer company was planning to run a marketing campaign in the near future based on the three-month purchasing trajectory in this survey question, the results would not be conclusive.
8. Provide Context as Needed
When it comes to collecting information, especially demographic data for surveys, it’s important to provide context as to why this information is being asked.
In general, demographic questions are used to understand the characteristics of survey respondents, which can help inform decisions made from the survey results. This includes things like age, gender, location, education level, occupation, and industry.
Providing context can help increase the accuracy of survey answers and promote the importance of as many answers as possible.
A simple sentence at the beginning of the survey or above sensitive sections (like demographics) stating a piece of information collected is used to better categorize answers and inform changes.
9. Don’t Force Surveyors to Answer Questions
If you want truthful and accurate answers to your survey, it’s important not to force the respondent to answer a question. This includes things like making questions mandatory or rejecting incomplete surveys.
If someone feels coerced into providing an answer, they may provide inaccurate information or worse yet, abandon the survey completely. Additionally, if you don’t give participants the option to leave a question blank, you risk skewing your data.
In the case of getting the most out of a survey, any answers are likely better than no answers at all. Casual survey takers do not want to feel like they’re doing work.
Without any real value for them and the forced answers, you run the risk of having little to no responses, ultimately affecting statistical significance.
10. Avoid Bias in Your Survey Questions
The goal of any survey is to gain unbiased, truthful results. However, this can be difficult when crafting questions that don’t lead a respondent to answer in a particular way.
Common sources of bias include some of the themes mentioned above, including leading questions (those that suggest a certain response), loaded language (where words sway respondents towards one opinion over another) and double-barreled questions (those that ask two distinct questions at once).
Bias can also come in the form of biased question order. As previously mentioned, if one question directly follows another, the respondent may answer based on their perception of how they should respond to the first question.
To create unbiased surveys, it’s important to phrase questions carefully and make sure the order of the questions does not influence answers.
Additional Reading: Survey Bias: What It Is and How to Prevent It
11. Avoid Lengthy Surveys with Too Many Questions
Lengthy surveys with too many questions can create survey fatigue, which is when respondents become bored and fatigued by answering an overwhelming amount of questions. As a result, respondents may answer quickly or abandon the survey altogether.
To avoid this, it’s important to limit surveys to only necessary questions. Additionally, if there’s any chance of adding long and complex questions, break them up into shorter, easier-to-answer questions.
Keeping surveys short and sweet is the best way to get honest responses, paired with the ability to skip any questions that don’t appeal to the participant. A general rule of thumb is that most closed-ended questions take 30 seconds to answer, and a five to ten-minute survey is most ideal for participation.
Within this framework, that’s 10-20 brief questions. Hoping for some longer responses? You’ll need to greatly cut down the number of questions to ensure more answers.
12. Test Survey Before Distribution
Testing your survey is important and will help you spot any potential problems with the questions, structure, or design. Start by testing it on a few colleagues, employees, friends, or family members who fit the target audience profile, depending on the nature of the survey.
Of course, large-scale surveys may require A/B testing amongst a smaller target group to narrow down any pain points before being distributed on a wide scale.
Ask them to complete the survey as if they were responding for real, but also ask them to look out for typos and awkward phrasing.
While you can’t eliminate all bias, testing the survey is a good way to spot any unintentional bias in the questions. Once you’ve spotted any changes that need to be made, go ahead and make them before sending it out to your target audience. This can help to determine pain points and help your survey run efficiently.
Examples of Good Survey Questions
Finally, here are some examples of good survey questions that follow the principles outlined above. Keep in mind that answers supplied in the case of close-ended would include “never,” “not applicable,” or “don’t know” where applicable to ensure participants are not forced to answer positively or inaccurately.
• What age range best describes you?
• What is your highest level of education?
• Are you currently employed?
• How often do you use our product/service?
• On a scale from one to five, how satisfied are you with our product/service?
• What is the primary reason for visiting our website today?
• How likely are you to recommend our product/service to a friend or colleague?
• Is there anything else we can do to improve your experience?
Each of these questions aims to provide necessary context, doesn’t force respondents to answer, and are written without bias. With this framework in mind, you can create survey questions that provide accurate, useful feedback.
Final Thoughts
Surveys are an incredibly useful tool for gathering feedback and gaining insights into customer behavior. However, creating unbiased surveys can be tricky, especially if you’re not familiar with the process.
With this guide in hand, you’ll now have a better understanding of how to create surveys that don’t lead respondents in any particular direction.
Just remember to keep surveys short, use open-ended questions whenever possible, and keep any close-ended questions as impartial as possible.
Proper testing, context, and a thank you will go a long way in helping to bolster the responses to your survey and ensure response rates and survey experiences are positive.