Rating scales in questionnaires: how to know which one to choose?

Rating scales are a key part of a market survey questionnaire. Choosing the most appropriate rating scale is easy if the right criteria are followed. But be careful, choosing the wrong scale can make the results of that survey question unusable. Therefore, today we explain everything you need to know about rating scales and give you some tips on how to use them more effectively in your questionnaires.

What is a rating scale?

A rating scale is a way of offering response options to a survey question. Rating scales capture different degrees of agreement, satisfaction, frequency or other perceptions and attitudes of respondents to a question. Rating scales are therefore much more nuanced than dichotomous questions that simply give two options, such as ‘yes’ or ‘no’.

Types of Rating Scales

Rating scales are used in all types of market surveys and are one of the most common questions to be used in any form:

• Likert scale: This is one of the most commonly used and measures the degree of agreement or disagreement with a statement, such as: ‘Strongly disagree’, ‘Disagree’, ‘Neither agree nor disagree’, ‘Agree’ and ‘Strongly agree’. The Likert scale provides information on the intensity of opinions or perceptions.
• Frequency Scale: Measures the frequency of a behaviour or action, with options such as ‘Never’, ‘Rarely’, ‘Sometimes’, ‘Frequently’ and ‘Always’. It is useful for understanding patterns of behaviour.
• Satisfaction Scale: Assesses the level of satisfaction with a product or service, using options such as ‘Very dissatisfied’, ‘Dissatisfied’, ‘Neither satisfied nor dissatisfied’, ‘Satisfied’ and ‘Very satisfied’.
• Numerical Scale: Consists of a series of numbers, e.g. from 1 to 5 or 1 to 10, where the respondent selects the numerical value that best represents his or her opinion. It can range from ‘Not at all satisfied’ (1) to ‘Very satisfied’ (5 or 10), depending on how wide the scale is.

How to design appropriate rating scales?

Designing rating scales in questionnaires requires attention. A good scale can improve the quality and accuracy of responses, facilitating the collection of useful and reliable data. A mistake in designing the rating scale can ruin the question. That is why at We are testers we always recommend using the following tips in your survey:

• Make sure that the ends of the scale are balanced. The ends of the scale should represent opposite versions of an idea (e.g. ‘not at all satisfied’ and ‘completely satisfied’ where ‘not at all’ and ‘completely’ are opposite terms). Avoid any imbalance and you will achieve a symmetrical and unbiased scale.
• Decide on the appropriate number of points on the scale. Odd scales usually have 5 to 7 points, which is the range that generally gives good accuracy without overloading the participants. A 5-point scale (strongly disagree to strongly agree) is easy to understand and gives sufficient range for differentiated opinions. Seven-point scales offer a little more detail, useful for issues that require more sensitivity in responses. By default, use five-point scales and only use seven-point scales when you have good arguments for doing so.
• Include neutral options only if necessary. If it is relevant to know whether participants do not have an opinion or feel neutral, include an option such as ‘Neither agree nor disagree’. But if you want to force an opinion, you can opt for even scales (such as 4 or 6-point scales), which have no middle option. Odd scales are very common, and many surveys use them by default, so they are absolutely fine. But keep in mind that while a high degree of ambivalence is in itself a result, it will make it difficult for you to make decisions based on the data. And it can often be very comfortable for respondents to opt for the neutral options that require less thought and do not engage them. Therefore, forcing them to choose one side of the scale may help you more in many cases. We recommend that whenever possible, use the even scale.
• Order the scale logically. Align the scale in a natural way, such as from minor to major or from bad to good. If labels do not follow a clear progression, they can confuse respondents. For example, if you use numbers, make sure they are in ascending order for positive responses.
• Use descriptive labels instead of numbers. Putting labels on all items on the scale (e.g. ‘very dissatisfied, dissatisfied, neutral, satisfied, very satisfied’) helps to reduce ambiguity, as each option has a clear meaning. If you only use numbers, such as 1-5, participants may interpret them differently, affecting the consistency of the results.
• Avoid bias in labels. Make sure labels are neutral and do not suggest an answer. For example, avoid overly loaded terms such as ‘excellent’, ‘perfect’, as they are not really the opposite of ‘very bad’ or ‘poor’ and will bias your scale. Use neutral words, avoid exaggerations and choose options that all respondents can interpret similarly and objectively.
• Do not introduce too many scales in a single questionnaire. Repeated use of rating scales can lead to fatigue and less reflective responses, especially in long questionnaires. If you need to use several scales, interspersing them with other types of questions (such as multiple choice or open-ended questions) can make the response experience more dynamic and reduce fatigue.
• Maintain consistency in the format of the scales throughout the questionnaire. Although you should not have too many rating scales or too close together, you can use them several times in your questionnaire. If you do, keep them consistent. This helps to avoid confusion and makes the response process smoother. For example, if you are using a five-point scale, try to keep that range throughout all questions that use rating. Do not use different ranges (such as five points on one scale and seven on another).
• Consider visual scales in some cases. A visual scale such as a slider bar or stars (if it’s satisfaction) can be more appealing to respondents, especially in online market surveys. This can help maintain interest during the interview and improve questionnaire completion rates. However, make sure that the design does not sacrifice accuracy in data collection.

As you can see, every element of the scale, from the range of items to the labels to the order of presentation, affects how respondents interpret and respond. A well-designed scale not only facilitates response but also contributes to the accuracy and validity of the data collected. And as we always recommend above all else, test your scale and the rest of the questionnaire with several people before launching the survey. Even the most experienced researcher can overlook details. Ask your colleagues to alert you to any ambiguities or improvements that can be made to any item in the scale and the rest of the questionnaire. Testing the questionnaire with several people will help you identify problems before you release the questionnaire to the sample.

Surveys with We are testers

Remember that in We are testers you have everything you need to create your surveys. You can do it yourself with an easy and intuitive research platform, and with a lot of versatility to cover all kinds of needs. If you prefer, we can create the surveys for you, so you don’t have to worry about the details of your questionnaire. Leave the design of the rating scales and the rest of the questionnaire in the hands of our research experts. And remember that we use a consumer panel of 130,000 people to find the perfect sample for your survey. Contact our experts now to find out all the possibilities.