We have all been asked to participate in a survey at one time or another, whether it was when you went to a restaurant, visited a website, or bought a product. When asked to participate in a survey, initial reactions vary depending on a number of factors, we’re often always in a rush, someone had a rough day, or you’re just enjoying some free time with your friends and don’t want to be disturbed. But when asked properly and weigh the balances of time and feedback, successfully completing a survey gives useful insight into the mind of the customer. White it usually only takes a couple of minutes to fill out a survey, hours of thought have gone into perfecting the questions included in it.
Most information collected through a survey is extremely important to a business as surveys are used to collect data on customer wants and needs.
Surveys help answer the questions businesses have by encouraging discussion and feedback with customers. This in turn allows businesses stay competitive by adapting to the ever changing customer tastes and lifestyles. (SnapSurvey)
Surveys are usually not composed of just one type of question. Usually, a survey will have multiple types of questions and answers. Whether you’re planning your own survey or just want to understand the thinking behind most survey questions, the 4 types of questions that are most effective and commonly used in surveys are explained below.
1. Open-Ended Questions
If you have ever filled out a survey after dining at a restaurant, you are probably familiar with questions like "What did you enjoy the most about our restaurant?". You have likely come across questions such as “What products would you like to see on our website?” when you are shopping online. Then there are the general question we ask each other like “Why did you decide to buy X?” that would lead to a more in-depth conversation that offers to pull more information from the respondent.
These types of questions are open-ended questions and allow the participant to elaborate on their answer. They are typically used when a fixed set of answers can not commonly include all the acceptable alternatives. Open ended questions are best when used at the beginning, as surveys are most effective starting from general questions that would lead up to more specific ones.
Asking open-ended questions will increase the time it takes to complete the survey, so keep in mind that while you may be able to capture more detail, you are creating a more taxing experience for the respondent and might reduce the number of surveys completed. Likewise, it can be extremely difficult to analyze the data collected from open-ended questions since the answers will vary immensely.
2. Multiple-Choice Questions
A multiple-choice question has fixed-alternatives for the participant to choose from. Some examples of these types of questions are:
How did you hear about us?
A) Television
B) Public transit Ad
C) Friend
D) Newspaper
E) Internet
Other:______ (please specify)
Will you visit our business again?
Definitely Probably
Probably not
Definitely not
No opinion
Here’s what to keep in mind when incorporating multiple-choice questions in a survey:
Having an "other" option can be positive or negative depending on the survey and the results you are aiming to achieve. “Other” is positive when there isn’t an option that the participant finds suitable. Having stated that, if most participants select this option, the purpose of the question will be useless since you might not know what “other” means in each participant’s mind unless you probe further. Therefore, it is important to have a list of alternatives that is comprehensive for the participant to choose from.
Having roughly 4-5 options to choose from will help get more specific answers from respondents. On the other hand, if the objective of the question is to get a simple idea of how the participant feels, having 2 to 3 options will suffice.
The biggest limitation to a multiple-choice question is the possibility that respondents will have a much more complex answer and will face some difficulty attempting to select an answer from the list of choices given to them.
Depending on the question you might want to have the option to either “check only one” or “check all that applies”. For example, given a “will you visit our business again?” question, a “check only one” makes the most sense, but for a “how did you hear about us?” question, you can definitely ask the participant to "check all that apply".
Finally, one of the most important details to remember when making a survey is to have mutually exclusive answers, as this will both avoid participant confusion and produce more accurate data. Consider this as an example of a question that is not mutually exclusive: How much are you willing to pay for our marketing services?
$1000-$2000
$2000-$3000
Since the answers are not mutually exclusive, a participant who is willing to pay exactly $2000 would not know which alternative to select.
3. Rating Scales & Matrix Questions
There are different types of rating scales. The two most commonly used rating scales are Likert type scales and semantic differential scales in which fixed alternatives act as scales for the participant to choose from.
Likert type scales typically have 5 or 7 intervals inquiring the participant to either agree or disagree with the statements presented. There is a score associated with each answer which can be used to analyze the data at the end of data collection. An example of a Likert type scale question:
How did you hear about us?
A) Television
B) Public transit Ad
C) Friend
D) Newspaper
E) Internet
Other:______ (please specify)
Will you visit our business again?
Definitely Probably
Probably not
Definitely not
No opinion
Here’s what to keep in mind when incorporating multiple-choice questions in a survey:
Having an "other" option can be positive or negative depending on the survey and the results you are aiming to achieve. “Other” is positive when there isn’t an option that the participant finds suitable. Having stated that, if most participants select this option, the purpose of the question will be useless since you might not know what “other” means in each participant’s mind unless you probe further. Therefore, it is important to have a list of alternatives that is comprehensive for the participant to choose from.
Having roughly 4-5 options to choose from will help get more specific answers from respondents. On the other hand, if the objective of the question is to get a simple idea of how the participant feels, having 2 to 3 options will suffice.
The biggest limitation to a multiple-choice question is the possibility that respondents will have a much more complex answer and will face some difficulty attempting to select an answer from the list of choices given to them.
Depending on the question you might want to have the option to either “check only one” or “check all that applies”. For example, given a “will you visit our business again?” question, a “check only one” makes the most sense, but for a “how did you hear about us?” question, you can definitely ask the participant to "check all that apply".
Finally, one of the most important details to remember when making a survey is to have mutually exclusive answers, as this will both avoid participant confusion and produce more accurate data. Consider this as an example of a question that is not mutually exclusive: How much are you willing to pay for our marketing services?
$1000-$2000
$2000-$3000
Since the answers are not mutually exclusive, a participant who is willing to pay exactly $2000 would not know which alternative to select.
3. Rating Scales & Matrix Questions
There are different types of rating scales. The two most commonly used rating scales are Likert type scales and semantic differential scales in which fixed alternatives act as scales for the participant to choose from.
Likert type scales typically have 5 or 7 intervals inquiring the participant to either agree or disagree with the statements presented. There is a score associated with each answer which can be used to analyze the data at the end of data collection. An example of a Likert type scale question:
Semantic differential scales, on the other hand, have opposing statements on the opposite ends of the scale. For example:
On a scale of 1-5, where 1 is sad and 5 is happy, describe how you feel when visiting our business.
Rating scale questions are best used when attempting to measure unobservable characteristics such as emotions, feelings, or opinions of an individual. Advantages of using rating scales in your survey is that they make the data collection process easier and faster. These scales can give more precise data than “yes/no” or “true/false” questions. However, just like multiple-choice questions, scales do not ask the question “why” and the answers given by participants can be biased since different participants may have different definitions of what certain words mean. For example, in the semantic differential scale example, the words “sad” and “happy” could have different meaning for different participants.
4. Rank order
Rank order questions can require a respondent to choose between two alternatives or to rank the alternatives in order of importance.
Examples: When you are deciding what kind of restaurant to dine in, which is more important: price or the atmosphere? Look at the list presented and rank them from 1-10, 1 being the most important and 10 being the least important.
Rank order questions are are known to avoid the order bias that multiple-choice questions typically hold. To avoid the order bias on multiple-choice questions, you should create multiple versions of a survey with alternatives for questions ordered in different orders. Since a rank order question is specifically asking the participant to rank the alternatives in order of importance, the participant will have to read all the alternatives and think through the answers. However, if the participant is in a rush when given multiple-choice questions, they might select the first answer that applies to them without reading the rest of the answers, which might contain an answer that applies better to the participant.
To sum up, when designing surveys, it is important to make sure the wording is simple for the participant to understand, avoid ambiguous questions and answers, implicit assumptions, and generalizations. Keeping this in mind will help you understand when to use which type of questions in surveys and help you make a great survey with actionable results!
4. Rank order
Rank order questions can require a respondent to choose between two alternatives or to rank the alternatives in order of importance.
Examples: When you are deciding what kind of restaurant to dine in, which is more important: price or the atmosphere? Look at the list presented and rank them from 1-10, 1 being the most important and 10 being the least important.
Rank order questions are are known to avoid the order bias that multiple-choice questions typically hold. To avoid the order bias on multiple-choice questions, you should create multiple versions of a survey with alternatives for questions ordered in different orders. Since a rank order question is specifically asking the participant to rank the alternatives in order of importance, the participant will have to read all the alternatives and think through the answers. However, if the participant is in a rush when given multiple-choice questions, they might select the first answer that applies to them without reading the rest of the answers, which might contain an answer that applies better to the participant.
To sum up, when designing surveys, it is important to make sure the wording is simple for the participant to understand, avoid ambiguous questions and answers, implicit assumptions, and generalizations. Keeping this in mind will help you understand when to use which type of questions in surveys and help you make a great survey with actionable results!