Secrets and Features of Writing Quality Survey Questions
Learn to design surveys so that you get reliable results.
The process of writing survey questions is extremely demanding and, perhaps an even more important stage than direct questioning. The point is that it is precisely the qualitatively posed questions that determine the receipt of answers because correctly selected formulations allow you to get the most complete answer required in the situation.
However, when composing questions, you need to adhere to a certain algorithm of actions. First, you need to determine the type of questions you will be using. It all depends on the final goals of your survey. We can advise open-ended questions, which imply that the respondent will further expand the answer with his own point of view on a particular topic. However, closed-ended questions can also be highly effective, where a person is presented with a limited set of answers that they can choose from. It should be understood that a closed-ended question does not always imply only a “Yes / No” answer, here you can pre-select such categories that will allow you to differentiate the answers in the best way and cover as many of the options as possible. It can also use multiple-choice options, Likert rating scales, and more.
Even defining the list of questions and their type is not the end of the preparatory stage. It is extremely important to determine both the intonation and the ways to ask these questions. As part of this material, we will look at a list of approaches to create a first-class questionnaire. We will pay special attention to writing quality questions with neutral answer options, and develop a balanced set of answer options. Moreover, we will look at how to avoid asking about two things at the same time, teach you how to create good closed-ended questions for a questionnaire, and also talk about the principles of creating a combined questionnaire, where various question options are used.
7 rules for writing a quality survey:
1. Concentrate on creating closed-ended questions.
Open-ended questions imply that a person is able to present their own point of view. These detailed responses can be extremely useful, but in some situations, they can only harm the survey. Firstly, their compilation takes more time, which is not always acceptable in the context of the survey. Secondly, they allow too streamlined answer options, when the interlocutor misses the very essence of the question, leaving it undisclosed.
In this regard, it is extremely important to think over the survey in such a way that there are a minimum number of open questions in it. Often, survey creators adhere to the principle that there can be no more than two open-ended questions in one survey. At the same time, they try to place them at the very end of the survey, after the person answers the entire list of closed questions. This trick allows you to avoid a situation where the interlocutor decides to avoid answering an open-ended question since you can perceive his point of view from closed question
2. Maintain a neutral tone in your questions.
One of the most important conditions for a correctly asked question is its neutrality. Questions to which you attach a certain emotional colouring can act as a prompting question, nudging the person towards one or other answer. This can significantly distort the real answer that the interlocutor wants to give.
For example, you could ask the question:
“Our service is really great. How cool do you think it is? “
This format of the question implies that the respondent must agree with the presented opinion, and as a result, the question completely loses its meaning. A really useful question from a service organization perspective would be, “How useful or unhelpful is the service for you?” Formatted this way the question will allow you to find out the opinion of the interlocutor and understand the aspects of the work that suit him or otherwise.
3. The answers provided must be balanced.
This point is closely related to the previous point about the neutrality of questions. It should be understood that each respondent must have the opportunity to give an honest and thoughtful answer, otherwise their position, as well as trust in it, may be threatened. An incorrectly compiled list of answers can become another factor that will reflect the bias of one of the parties. Let’s look at a specific example. Let’s go back to the question: “How useful or unhelpful is the service for you?” Imagine that you have given a respondent the following answers as options:
- Incredibly useful;
- Very useful.
Obviously, such answer options reflect only the positive side of the question and cannot be credible. Any answer, like a question, must imply an objective tone, and therefore a more balanced set of answers is required, for example:
- Very useful
- Haven’t been influenced either way
- Really useless.
4. Don’t ask about two things at the same time.
The surveyor should understand that confusing respondents is as bad as influencing their answers. After all, both situations lead to the fact that they choose the wrong answer option as opposed to how they really feel. that they would like in reality.
A fairly common problem is the so-called “double question”, where the respondent is asked to simultaneously assess two different aspects at once. For instance:
“How would you rate the quality of our customer service and the quality of our product?”
This question asks the interlocutor to rate the customer service and the quality of the product, which involves considering two different topics. As a result, the respondent will be forced to consider only one of the questions. Because of this, you can miss an extremely important opinion on one topic or another. Moreover, “double questions” allow the respondent to avoid uncomfortable topics.
However, there is a fairly simple solution – you need to split these questions into two and ask them in turn.
5. Don’t ask similar questions.
Asking people about the same thing will generally not go down well. It is also true with a survey. Hardly anyone will appreciate if during the dialogue they constantly return to the same questions, topics, and or answer options. This leads to a negative reaction from the respondent, who will either start to ignore such questions or give dry direct answers, and this will result in the point being missed.
To solve this problem, it is necessary to analyze the existing list of questions in order to exclude those that are the same or overly similar to each other. Create and follow a survey structure to avoid repeating the same topics.
6. Make most of the questions optional.
It is extremely important to take into account that a person cannot have all the information, or that he does not have the desire or ability to share it. In this regard, the questions should leave the possibility that some questions will remain unanswered. The priority of mandatory and optional questions must be determined prior to the survey. However, remember not to overdo it here, as being forced to answer can cause an overwhelmingly negative reaction, due to which the respondent completely refuses to share information.
7. Check your questionnaire.
There is nothing worse than correcting errors in a questionnaire during a survey or after it has been submitted. In order to avoid such a problem, carefully read the list of questions, send them to relatives or other people so that they can also check it. A fresh pair of eyes will help you find potential mistakes quickly and easily.
Remember, the better your questions are, the better the answers you will get. Put these tips into practice to create a quality profile today!