Skip to content
All apps

Levels of measurement

About this app

Drag variables into the five levels of measurement (nominal → ordinal → interval → ratio → absolute). Each step adds a property; the operations panel shows which statistic is allowed from which level — mean from interval, ratio from ratio level.

What levels of measurement are

The level of measurement of a variable determines which calculations make sense with its values. The levels form a staircase, each step adding a property. Nominal: values can only be told apart. Ordinal: they also have an order. Interval: they also have equal distances but no true zero. Ratio: they also have an absolute zero. Absolute: they also have a natural counting unit. Hence a mean only makes sense from interval level up, and a ratio such as “twice as much” only from ratio level.

What the app shows

The five levels stand side by side as a staircase. Variables such as country, postcode, school grade, a Likert item, temperature in °C, revenue or number of children wait as chips. You drag each chip onto its step and immediately get a reason, also when the answer is wrong. Tap a step and the operations panel shows which statistics are allowed there: mode and frequencies, median, differences, arithmetic mean, ratios.

What to watch for

Look out for traps: postcodes are numbers but only labels. School grades have an order, yet the gap from 1 to 2 is not defined as equal to the gap from 5 to 6. At 0 °C nothing is “empty”, so “twice as warm” is meaningless. Allowed operations are cumulative: each step inherits everything below it. “Show solution” lights up the correct steps, and “New round” reshuffles.

Frequently asked questions

What are the four levels of measurement?
The classic four are nominal, ordinal, interval and ratio. German textbooks often add an absolute scale for pure counts. Interval and ratio scales together are also called metric or quantitative scales, and each higher level allows more operations.
Can you calculate a mean of Likert scale data?
A single Likert item is ordinal, because equal distances between the answer options are not guaranteed. Calculating a mean assumes equal distances, so it is an assumption that needs justification. Median and frequencies are safe at ordinal level.
Is temperature interval or ratio data?
Temperature in °C or °F is interval data, because the zero point is set arbitrarily. Differences are meaningful, but ratios such as “twice as warm” are not. Temperature in kelvin has an absolute zero and is ratio data.
How does the exercise in the app work?
You drag variable chips onto the matching step and get a short explanation for each placement. A counter shows how many are correct. Tapping a step lists the statistical operations allowed at that level.

Subject: Data & decisions

More from Data & decisions

A VisuApp by heyprof: interactive, free in your browser, no sign-up. What is a VisuApp? · All apps