What is a Likert scale?

By Dr. Rafiq Muhammad, MD, PhD · Updated June 2026

A Likert scale measures an attitude by asking how much people agree with a set of statements — the “strongly disagree…strongly agree” format you’ve answered a hundred times. It’s the workhorse of survey research. The detail that trips students up: a single statement is a Likert item; the Likert scale is the combined score of several items measuring the same thing — and that difference changes how you analyse it.

Item vs scale — and why it matters

A single Likert item produces ordinal data: the gaps between “agree” and “strongly agree” aren’t guaranteed equal. A Likert scale — the sum or mean of several items measuring one construct — is commonly treated as approximately interval, which is what lets you report means and use parametric tests. So: analyse a lone item as ordinal (median, non-parametric tests); analyse a multi-item scale as roughly interval (mean, parametric tests). Conflating the two is the most common Likert mistake.

How many points?

Five and seven points are the standard choices. Fewer loses sensitivity; many more than seven adds little and tires respondents. The other decision is the neutral midpoint:

• Odd number (with midpoint) — lets genuinely neutral people answer honestly, but can become a lazy default.
• Even number (forced choice) — pushes a leaning, useful when one exists, but risks manufacturing opinions.

There’s no universally right answer — choose deliberately and justify it in your methods.

Reliability of the scale

Because a Likert scale combines items, you should report whether those items actually hang together — that’s internal-consistency reliability (Cronbach’s alpha). A scale with low alpha isn’t measuring one coherent construct, and summing its items is then meaningless.