question writing dos and don'tsDos and Don’ts

Rethink how to address your target participants

When you have decided what to ask, it’s time to determine how to ask it. Your survey participants might belong to different demographic groups. Your target group might be specified in regards to age, level of education, profession, etc… Therefore you have to use language that is clear and understandable for your audience.

In other words, you should speak their language. You should consider what they already know, and use terminology that is understandable to them. Always try to write your questions in a way that is coherent and easy to grasp.

Be clear and concise

In a way, it doesn’t really matter who is going to participate in your survey. You have to be clear with what you ask and avoid long and complicated questions anyway. Use words with clear meanings. If you have to use any less familiar expressions, jargon, terms or acronyms, make sure you provide descriptions for them.

If your participant has to read a question two times to understand it, this means you have already lost them. A good exercise is to practice writing questions that you could see yourself asking friends or colleagues.


What was the state of your room when you arrived at our hotel in terms of cleanliness and tidiness?
How clean was your room when you arrived?
Do you use any medicines several times per week or per month?
Do you regularly use medicines?

Be humble

Sometimes people are frightened to respond to surveys. If respondents feel they are not qualified or good enough to answer a question, or if they feel sharing their true opinions might sound stupid to the creator of the survey, they just might leave your survey.

If you really want participants to share their true opinions with you without feeling intimidated, due to their level of knowledge, beliefs, or profession, then you have to make them feel comfortable by using the right tone in your survey questions.

So, when you ask questions, use phrases such as “do you think …”, “do you feel …”, “according to you …” or “in your opinion”. In this way, respondents to your survey will not think they are participating in a quiz, and they will not fear their answers will be seen as right or wrong.


Which types of services should our company offer, to be considered a good company?
In your opinion, which types of services should our company offer, to be considered a good company?
Is there any problem with the choice of colors in our product?
Do you feel there is any problem with the choice of colors in our product?
Some questions might be sensitive, and therefore participants might become defensive and hide their real opinions. To avoid such situations, you can make them feel that they are not alone, or they are not the only people who may have such opinions.


Do you think that the administration of your department needs to be changed?
Talking about reforms in the company, we have found out some of your colleagues think that the administration of your department needs a change. What about you? Do you share this opinion?

Provide a timeframe

In many cases, the answer you are looking for is meaningful when put into a timeframe.

For example, you want to see whether or not your advertising campaigns have successfully reached local people in the past month. So, you make a survey and ask your participants: “Do you follow the local media?”

Well, this might be really hard for them to answer, or the answers might be very misleading for you. Maybe a respondent does usually follow the media, but for the past month they have been on a trip or for some reason could not follow the media, therefore missed your advertising campaign. So, if they answer “yes”, you might be upset with your campaign because you may think it has not reached them or has not been interesting enough.

So as you see, asking the right questions will be something like this:

“Have you followed the local media in the past month?”

More examples:

Do you visit any health care professionals?
In the past year, have you visited any health care professionals?
How much time do you spend watching TV?
In a typical day, about how much time do you spend watching TV?

In you do not have a clear or specific time period in mind, it can be a good idea to begin your questions with phrases or words such as “In general”, “Overall”, “Typically”, “Usually”. This make the participants answer based on their average behaviour or usual habits.

However, such questions are always harder to answer and it is more difficult for participants to evaluate something in vague timeframes.

If possible, you can place the timeframe in the beginning of your survey questions, to put more emphasis on it.

Be unbiased, ask unbiased questions

Well, the whole point with doing a survey is getting the real opinions of your target group, not getting a result that you “like” or “prefer” to get. It’s all about being real and dealing with reality.

Survey experts count five common mistakes that might cause poor survey results.

1. Using leading questions

If you create biased questions in your survey, this can cause your audience to provide you with answers that might not be real. You should avoid leading them to an answer that you may prefer to hear.


Has our amazing new product made your life easier?
How do you feel about our new product?

2. Asking questions with loaded language

You usually have some personal opinions or preconceptions about the subject for which you have created the survey. But beware of allowing your own ideas to leak into your survey questions. Such mistakes can influence the participants and their responses.

When you ask questions with loaded language, you are putting emotions or connotations into  the questions, that may effect participants as well.


In the past week, how much time did you waste checking your Facebook?
In the past week, how much time did you spend checking your Facebook?

3. Asking double-barreled questions

Ask about one thing at a time. Try to make a distinct question for each subject that you need information about. Avoid asking questions that contains more than one subject. For example:

“Do you play computer games and watch TV a lot in your free time.”

How would the participant answer if they only watch TV andDon’t play any games, or vice versa? What is the answer to this question really? Yes? No? Kind of?


Do you exercise a lot and eat healthy food?
– In the past two months, how much time was spent on exercise?
– In the past two months, about how many days did you eat healthy food?

4. Asking unbalanced questions

Give readers a possibility to express their opinions more freely, by putting opposing answers into the questions. In this way, the questions will have more balance.


Do you like mathematics?
– Do you like mathematics or not?
– Do you like mathematics, or do you not like mathematics?

An unbalanced question is one that has a question stem that does not provide the reader with all reasonably plausible sides of an issue.

The problem of balance in a survey question also can apply to the response alternatives that are presented to respondents.

Unbalanced questions generally are closed-ended questions, but it is possible to use open-ended questions in which the question stem is unbalanced.

However, an unbalanced question may not always lead to biased data, but that is the concern in most instances.

5. Asking overly broad questions

Avoid asking questions that expect respondents to write an article as an answer. For instance, if you ask a broad and very general question such as: “How do you feel about our product?” and then give them a comment box to fill in, what do you expect as an answer?

A product can be criticized or praised from many different angles. You should be much more specific and narrow down the topic; then write one question per topic.


What do you think about our product?
– Do you like the tactile quality of our product or not
– Do you like the color that we use in our product or not?
– How easy is it to handle our product?
– Etc…

Put your questions in the right order

When ordering your survey questions, you should try to put them in a logical order, and group questions for similar topics together. A good idea is to use section breaks, or page breaks (both available in SurveyLegend fields) between groups of questions, based on your surveying strategy. Also, it’s better to put a brief overview of your survey structure in the beginning of your survey.

Put easy questions first

If possible, easier questions should come earlier in the survey. This will create a pleasant start for your respondents, and it will be more likely that they continue answering your whole survey. It’s always good to give this impression that the survey they are going to participate in is not that hard. In oral surveys, this also helps the interviewer build a good relationship and trust with the respondent.

Put difficult or sensitive questions last

Conversely, put the more difficult questions near the end of the survey. If your respondents face a difficult question right at the beginning, they might assume that all the questions are going to be that difficult. Therefore, taking the survey starts to look like a real hassle, right from the beginning.

But if they see the hard or sensitive questions at the end, they may put in the effort to finish the survey, because they have already done most of the work, and know that there are not many questions left. Note that SurveyLegend shows a nice progress bar in each survey, which lets the participants know how much more of the survey is left.

Plus, even if participants leave the survey or skip the hard questions, at least you will have most or at least half of the survey to analyze, instead of none.

Become a Legend Now ;)

Make your 1st free online survey today!

starMobile-ready surveys
starUnlimited responses
starLots of amazing features!