What is a Likert scale?

If you’ve ever answered a question that asked how much you agree or disagree with something, then you’ve answered a Likert scale question.

Likert scales are widely used to measure attitudes and opinions with a greater degree of nuance than a simple “yes/no” question.

Let’s explore this in more detail: what is a Likert scale, when should you use one and how can you create an effective Likert scale questionnaire? You’ll also see some examples of Likert scale questions used well.

The Likert rating scale is a type of survey scale. It’s a question with a series of answers to choose from, ranging from one extreme attitude to another, normally with a moderate or neutral option. It’s sometimes referred to as a satisfaction scale as it is ideal for measuring satisfaction with a product or service.

A good example would be, “How likely are you to recommend our veg box delivery service?”, with pre-defined answer options, ranging from “very likely” through to “highly unlikely”. A five-point Likert scale, where you have five different options to choose from, is the most common, although a seven-point scale is also used.

Named after their creator, American social scientist Rensis Likert, Likert scales are popular because they’re considered one of the most reliable ways to measure opinions, perceptions and behaviours.

Compared to binary questions, where you only have two answers to choose from, Likert-type questions give you more granular feedback. For instance, discover whether your product was just “good enough” or (hopefully) “excellent”. Likert questions can help you decide whether a recent company training session left staff feeling “very satisfied”, “somewhat dissatisfied” or maybe just neutral.

Using the Likert scale in your survey will help you uncover degrees of opinion that can give you a deeper understanding of the feedback you’re receiving. It can also help identify the areas in which you could improve your service or product.

These questions are used in many different types of surveys. Maybe you want to find out what your customers think about your latest product or online shopping experience, or find out how happy your staff are at work.

Likert scale questions are used in many different types of surveys, whether you are trying to find out how your employees feel about their work or what your customers think about your latest product.

To help you use Likert questions effectively, here are some examples of how they can look:

There are so many kinds of survey questions that it can be hard to know when to use which one.

Likert scale questions are great for looking at a specific topic in greater detail. They help you find out people’s thoughts about specific aspects of a wider topic. Think of Likert scale surveys whenever you need to find out more about…

… or any other questions where you need to measure sentiment about something specific, but in more detail.

In the methodology world, this greater level of detail is known as variance. The more variance you have, the better you know the nuances of someone’s thinking.

Another advantage of the Likert scale is that it can help you avoid some of the common pitfalls of survey design, like creating overly broad questions that can be tricky to answer.

It’s an easy mistake to make. If you’re designing your survey in a bit of a hurry, it can be tempting to try and take a shortcut by reaching for the broader types of questions, like “yes/no”, “select all”, open-ended, ranking or matrix questions. But these sorts of questions can make people frustrated and mean they start answering too quickly, which spoils the quality of your data.

Don’t worry, most of the time you can avoid this problem by using the trusty Likert scale. The five-point Likert scale in particular, with its simple language and list of clear, well-defined options, will keep those completing your questionnaire focused and happy.

Within your survey, it’s helpful to group together questions related to the same topic. Not only will this give it a more logical order, but it will also help when it comes to analysing your findings. It will allow you to group together the results and look at findings for a topic as a whole. Let’s put this into context with an example.

You could start with this initial question:

How satisfied or dissatisfied are you with the quality of the dinner you were served tonight?

And then follow up with:

How satisfied or dissatisfied are you with the quality of your starters tonight?

How satisfied or dissatisfied are you with the quality of the main course tonight?

How satisfied or dissatisfied are you with the quality of dessert tonight?

But here’s one question you should leave for another section of the survey:

How satisfied or dissatisfied are you with service from the waiting staff tonight?

You could then add up the responses related to one topic to get an overall score. In this case, a “Quality of food” score. This can help give you a more holistic measurement of the attitudes about the particular product, service or event you’re researching.

Be accurate. Likert scale survey questions must be written clearly in order to avoid confusion and increase their effectiveness. If you ask about your customers’ satisfaction with their online shopping experience, do you mean the ease of use of the website or the range of products? Do you want to know whether your customers were happy with the quality of the products, the speed of delivery or the quality of your customer service? In short—the more specific you can be, the more valuable the data.

Be careful with adjectives. Your response options need to include descriptive words that are easily understood to ensure people will know exactly what you mean. It needs to be clear which option or grade is higher or bigger than the next. For instance, is “pretty much” more than “quite a bit”? We recommend starting from the extremes (“extremely” and “not at all”) and then setting the midpoint of your scale to represent moderation (“moderately”) or neutrality (“neither agree nor disagree”) followed by clear, unambiguous terms like “very” and “slightly” for the remaining options.

Bipolar or unipolar scales? A bipolar scale is where your survey question asks participants to choose on a scale between two polar opposites. Meanwhile, a unipolar scale focuses on one specific characteristic and the extent to which it is present, or true. A good example of a unipolar scale is one that ranges from “extremely brave” to “not brave at all”. Meanwhile, a bipolar scale would range from “extremely brave” to “extremely shy”. In most cases, a unipolar scale is preferable because people find it easier to focus on one quality at a time. What’s more, because each end of the unipolar scale is the exact opposite of the other, it’s more methodologically sound.

Better to ask. It’s generally preferable to ask a question, rather than presenting a statement and asking the extent to which people agree with it. Why? Because most people will tend to agree rather than disagree with the statement since humans tend to prefer to be polite and respectful. This phenomenon is known as the acquiescence response bias.

You’re probably familiar with Likert scale—or satisfaction scale—questions. In fact, you’ve probably answered a fair few in your time, even though you may not have known their name. And now, hopefully you’ll understand how you can use them effectively to gather nuanced answers to your questionnaires.