Likert scale: meaning and example

Where the term comes from

The format goes back to psychologist Rensis Likert, who introduced a technique for measuring attitudes in 1932. The basic idea is that instead of answering only yes or no, a person can choose from several ordered levels to express the degree of agreement more precisely. These levels are converted into numbers so that responses to several statements can be combined into an overall score.

In everyday language, any multi-level response scale is often called a "Likert scale." In methodology, a more precise distinction is useful. A single graded item is a Likert item. In the narrower sense, a Likert scale is created only when several such items are combined into a common score.

Item or scale, a simple example

The statement "I quickly feel exhausted in groups" with five levels from "strongly disagree" to "strongly agree" is a single Likert item. If you use several statements on the same topic, assign a value to each level (for example, 0 to 4), and add them together, you get a Likert scale. A purely mathematical example: With three items scored 4, 3, and 4, the total is 11 out of a possible 12 points. These numbers are chosen only to show the principle. They do not say anything about a test result.

Not every multi-level response is the same

A common misunderstanding is that every question with several response options is automatically a Likert scale. That is not the case. The Masking Test with the CAT-Q, for example, uses seven levels of agreement, which is a typical Likert response format. The Depression Test with the PHQ-9, by contrast, asks on how many days a symptom occurred, so it measures frequency. It also has several response levels, but methodologically it is not the same as an agreement scale because it asks about a different type of information. The Self-Esteem Test based on the Rosenberg scale, in turn, classically uses four levels of agreement. These differences in response format are not a minor detail. They affect how a score is calculated and what that score means.

Common confusion

Along with confusing an item with a scale, people often assume that more response levels are always better. That cannot be said in general either. Whether four, five, or seven levels make sense depends on the purpose, the target group, and whether a neutral middle option is intended. What matters is that respondents can clearly distinguish between the levels. Very fine distinctions that people cannot reliably tell apart do not improve accuracy.

What this means for self-tests on medtests.net

Many psychological self-tests on medtests.net use Likert response formats. The Big Five Test and the Masking Test let you rate statements using graded response options, and several items are combined into a score for each scale. This format explains why such tests use degrees of agreement rather than right or wrong answers. They measure degrees of a characteristic, not facts. What an individual total score means depends on the specific instrument and is explained on the corresponding test page. Such a score is a guide, not a diagnosis.

What the term does not mean

A Likert scale is a measurement and response format, not a quality seal. The fact that a test uses a Likert scale does not by itself tell you whether it measures reliably or validly. That is what reliability und validity address. A high total score is also not proof of a condition. It is an indication that a professional can evaluate further.