What are dual-meaning questions

It's no secret that customer feedback is important. In order to be able to evaluate the satisfaction and loyalty of your customers, you must of course first find out how they really think about your products or services think.

But how do you get this information? Very simple: with a survey.

You may already have a process for creating surveys, but now you might be wondering what kinds of questions to ask? To do this, you first have to ask yourself a question: What information do we want to find out?

Once you know that, you can set up your survey. In this article, we'll show you different types of questions and what information they can provide. We'll also give you a few examples and best practices to illustrate your point.

What types of questions are good for an effective survey?

  1. Multiple choice
  2. rating scale
  3. Likert scale
  4. Order of precedence
  5. Polarity profile
  6. Alternative questions
  7. Closed questions
  8. Open questions

1. Multiple choice

As the name suggests, respondents have multiple choices to choose from when answering a question. Most of the time there is also an option like “Other” with a free field in which respondents can enter their answer if none of the given answers are suitable for them.

Multiple choice questions are very popular in surveys because they are easy to fill in and provide clear data that is easy to analyze. A multiple selection is suitable for collecting demographic data, information on the use of the product or service and the priorities of the customers.

Single choice

In the case of multiple choice questions, you can ask your participants to make only a single choice from a list of options. Round buttons (so-called radio buttons) are often used online.

Multiple choice

In the case of multiple-choice questions, you can also ask your survey participants to mark all of the answer options that apply in a list. Checkboxes are often used for this, which participants can click on.

2. Evaluation scale

With rating scale questions (also called ranking or ordinal scale), respondents should rate a situation using a numerical scale, the end values ​​of which are assigned a certain mood. With this question you can e.g. B. measure customer satisfaction on a scale from 1–10. You can choose which of the numbers are positive and which are negative.

Rating scale questions help measure progress over time. If you send the same rating scale several times over a period of time, you can measure whether the mood of your survey participants is developing positively or negatively.

The Net Promoter Score® (NPS) is a typical use case for rating scale questions.

3. Likert scale

Likert scale questions indicate whether a respondent agrees or disagrees with a question. The scale usually has five or seven points and then ranges, for example, from “very unlikely” to “very likely” or “totally disagree” to “completely agree”.

A Likert scale is suitable, for example, to evaluate customer satisfaction.

4. Ranking

For these questions, respondents are asked to rate various answers according to personal priority or importance. These questions provide qualitative feedback about the respondents, but with no explanation for the answers.

A ranking should be used to determine the needs and behaviors of your customers and to analyze how they use your product or service and what they are still missing.

5. Polarity profile

For polarity profile questions, participants are also asked to rate something on a scale. However, the end of the scale is a different, contradicting statement. So instead of answering the question, “Do you agree or disagree with X?”, The participants have to give answers that correspond to their perception or their emotional reaction.

For example, a polarity profile question might be, “On a scale of 1 to 5, how did you find the service?” Where 1 is terrible and 5 is excellent. These questions are about evaluating the respondents' intuitive response. However, they may be more difficult to evaluate than predefined answers, e.g. B. Express approval or disapproval.

You should use polarity profile questions if you want to collect clear, qualitative feedback from your customers.

Likert scale compared to the polarity profile

With both a Likert scale and a polarity profile, you ask participants to provide an answer using a scale. The difference between the two systems is how the questions are asked. In Likert scale questions, participants receive a statement that they either agree or disagree with. With the polarity profile, participants are asked to complete a statement in a question, for which they can choose from different, opposite words and phrases.

6. Alternative questions

Alternate questions provide only two answers that respondents must choose from. These questions are easy to answer for the participants and easy for you to evaluate. However, there is not much room for interpretation.

Alternative questions provide you with clear data that you can analyze easily and quickly.

7. Closed questions

Closed questions have a set number of possible answers that respondents have to choose from. All question types described so far are closed questions. When creating your survey, you choose whether there are several answer options or just one selection.

8. Open questions

For the question types mentioned so far, the answer is possible in the form of a selection from several specified options. In the case of open questions, on the other hand, there is usually an empty text field in which participants can enter an individual answer to the question.

This qualitative feedback can be very helpful if you want to understand and interpret the mood and problems of your customers. However, the results are not easy to analyze and trends and changes of opinion are difficult to deduce. To interpret this qualitative feedback and analyze it for mood and tone of voice, you need human support. Often there are also spelling errors that cause problems in a machine evaluation.

We recommend that you ask open questions together with at least one closed question. In this way you can collect clear data that you can analyze and forecast over a period of time and also receive direct feedback from your customers.

Sample questions:

1. Multiple choice

Here is an example of a multiple choice question using Single choice:

Here is an example of a multiple choice question using Multiple choice:

2. Evaluation scale

Here is an example of a rating scale question in an NPS survey.

3. Likert scale

Here is an example of a Likert scale question using five points:

4. Ranking

Here is an example of a ranking question:

5. Polarity profile

These are examples of questions with a polarity profile:

6. Alternative questions

Here is an example of an alternative question:

We'd love to hear what you think of our customer service. Please take a moment to complete our survey by clicking on the appropriate link:

How would you rate our customer service?

Well i'm satisfied.

Bad, I am not satisfied.

7. Closed questions

Here is an example of a closed multiple choice question:

8. Open questions

You could include this example of an open-ended question in a survey:

"Why did you decide to register with us?"

How to formulate questions for surveys

  1. Avoid leading questions
  2. Avoid technical jargon
  3. Avoid unnecessary or double negations
  4. Avoid double questions
  5. Leave no room for vague answers

1. Avoid leading questions

This refers to questions that suggest the participants which answer to choose. Here's an example:

"What is your favorite HubSpot tool?"

The question is suggestive because the participants may not even like to use HubSpot. So a list of different tools may not lead to an honest answer here. The question can be improved by adding the “I don't have a favorite tool” option.

Questions that induce your respondents to answer a certain way because the question is phrased accordingly produce unreliable results. To avoid this, you should make sure to always formulate your questions clearly, precisely, impartially and neutrally. As little as possible should be included in your question that might induce respondents to give your preferred answer. Maybe even have a third person with no subject matter look over your survey for feedback.

Leading questions usually contain emotionally charged assumptions that might lead participants to respond in a certain way. One way to avoid such emotional components in your questions is to avoid adjectives.

 

Pay attention to neutrality with your questions and include various answer options in order to give your survey participants a suitable answer.

2. Avoid technical jargon

Technical jargon can make participants feel like they don't know what is going on. Use clear, clear language so that your respondents don't have to go through a dictionary. Here is an example of a technical jargon question:

"How is your relationship between CAC and CLV?"

With such a question, you are assuming that your participants know both acronyms and their relationship to one another as a measure. That may not be the case. If acronyms, abbreviations, or other technical terms appear in your survey, simplify and / or explain them to improve understanding. The survey should explain the terms and provide a response option for those respondents who do not have this information on hand.

3. Avoid unnecessary or double negations

Unnecessary or double-negative phrases are confusing and could even cause participants to become so irritated that they do not want to complete your survey. Here is an example of a question that contains an unnecessary negative because two feelings are in conflict:

"Don't you like to use Google?"

Instead, you should phrase the question like this: “Do you enjoy using Google?” This makes the meaning very clear.

On top of that, such formulations unnecessarily complicate the analysis of the answers. Think of it this way: If the question is already formulated in a confusing way, how are you going to determine which statement the respondent agrees with?

4. Avoid double questions

Double questions ask two questions at once. If you use such questions, your respondents will not know which to answer. As a result, the results may be misleading. Here is an example of a double question:

"Are you satisfied or dissatisfied with your salary and career with your current employer?"

If respondents are happy with their salary but not with their career advancement, they won't know whether to answer “happy” or “dissatisfied”.

You should also include such different aspects in two different questions. This way you don't confuse your respondents and get clearer results.

5. Leave no room for vague answers

Since you are conducting a survey to ask for opinions, you should not allow your survey participants to provide answers such as “no comment” or “not relevant”. Instead, you should offer a variety of answer options that allow the most informative information possible. You should also enable answers like “I'm not sure” to learn more about how your respondents' skills or knowledge are distributed.

Net Promoter, Net Promoter System, Net Promoter Score, NPS and the NPS-related emoticons are registered trademarks of Bain & Company, Inc., Fred Reichheld and Satmetrix Systems, Inc.

Originally published June 7, 2019, updated January 08, 2021

Recommend this article: