Have you ever pressed one of those smiley face feedback buttons at the airport or a store?
Perhaps you pushed a button because it seemed fun. Maybe you ignored it because it seemed pointless. As it turns out, these simple machines collect incredibly useful data for improving customer experience.
These HappyOrNot terminals measure people’s moods and experience in real time, converting the smiley faces into quantitative data.
But long before smiley faces dotted airport terminals, Likert scale questionnaires have been doing the same thing: they quantify attitudes, feelings, opinions, and other subjective experiences.
Unlike binary yes or no questions, the Likert scale offers deeper insights into what people are thinking and feeling.
In this guide to Likert scale surveys and questionnaires, you’ll learn exactly what a Likert scale is and the best way to create and analyze your Likert scale surveys.
A lot of people define Likert scales as questionnaires where you choose among five- or seven-point scales, such as Strongly Agree—Strongly Disagree. But I strongly disagree—those are Likert-type responses.
A Likert scale is what emerges from the collective responses to a set of statements designed to measure a complex concept. That’s kind of confusing, so here’s an example:
We can all agree that there’s no single question that designates someone as politically liberal or conservative. Political affiliation is complicated—a person who is liberal on matters of health care could be conservative on questions of international aid.
So instead of asking one question, you measure how much people agree or disagree with various statements about political policy. Then when you combine or average a person’s responses, you get a more accurate measure of their liberal or conservative opinions.
In short, a Likert scale is the result of the survey. The individual statements and response-types are a methodology for scaling—or measuring—attitudes.
Okay, so how do you craft a Likert scale questionnaire that gets you results you can count on? That’s the hardest part. But not to fear, these 4 steps will help you get it right:
Once you’ve crafted the questionnaire for a Likert scale survey, most of the hard work is already done. If the statements in your questionnaire measure a particular concept when combined, that’s a Likert scale. Averages and standard deviations among your participants describe the data.
If you have used Likert-type items in a survey, then averages, medians, and frequencies are the tools you need for analysis. The tendencies in the data will give you answers to the questions that prompted this survey.
Although Likert scale surveys are relatively simple, there’s a big weakness to keep in mind when interpreting your results: respondent biases.
As with most surveys, social desirability bias can affect the reliability of your Likert scale survey. Even when you tell respondents their responses will remain anonymous, many people still try to give socially acceptable rather than being honest. This can be minimized with an even-number scale, which prevents people from sitting on the fence.
Likert scale surveys are particularly susceptible to central tendency bias: people avoid choosing the most extreme responses such as Very Helpful or Strongly Disagree. You can reduce the effect of this bias with some clear definitions, such as “In this survey, Very Helpful means you got everything you needed from our customer service agent.”
At the other end of the spectrum is extreme response bias. This happens when people only choose the extreme options. There are several causes of this, including cultural attitudes and IQ. But the one you have control over is the wording of the question. Use neutral language and don’t ask leading questions.
Leading questions ‘lead’ respondents to the survey creator’s preferred answer or desired outcome. This creates bias. For example, in 2014, the Scottish government wanted to ask, “Do you agree that Scotland should be an independent country?” The electoral commission deemed this to be a leading question and recommended, “Should Scotland be an independent country?”
People tend to agree with statements to please others. This is known as acquiescence response bias—otherwise known as the when-in-doubt-just-agree bias There are two ways to minimize its effect:
1) Phrase your statements as questions
2) Include a positive and negative statement and evaluate the pairs for consistency.
Likert scale surveys are overall, pretty simple to create, easy to complete, and provide highly reliable data. They allow you to capture the variation and complexity of people’s attitudes, giving you deeper insights into what people are thinking and feeling.
Getting your Likert scale survey right means loads of useful data to help you get to know people better. Ready to get started? We can help you create a survey today!