Height is neither nominal nor ordinal. It is a ratio variable, the highest level of measurement in statistics. Height has equal intervals between units, a true zero point, and can be meaningfully added, subtracted, multiplied, and divided. If you’re working through a stats assignment and trying to classify height, ratio is the correct answer.
That said, height can be deliberately converted into ordinal or even nominal data in certain situations. Understanding why height is ratio by nature, and when researchers choose to downgrade it, will help you classify any variable you encounter.
The Four Scales of Measurement
Statistics organizes data into four levels, each building on the one below it. Knowing what operations each level permits is the key to classifying any variable correctly.
- Nominal: Categories with no inherent order. You can only check whether two values are equal or not. Examples: blood type, eye color, country of birth.
- Ordinal: Categories that have a meaningful order, but the gaps between them aren’t necessarily equal. You can rank values as greater or lesser, but arithmetic doesn’t work. Examples: pain rated 1 to 10, education level, race finishing positions.
- Interval: Numeric values with equal spacing between units, but no true zero. You can add and subtract meaningfully. The classic example is temperature in Celsius: the difference between 20°C and 30°C is the same as between 30°C and 40°C, but 0°C doesn’t mean “no temperature.”
- Ratio: Everything interval has, plus a genuine zero point that means “none of this thing exists.” You can multiply and divide. Examples: height, weight, distance, income.
Why Height Is a Ratio Variable
Height meets every requirement of the ratio scale. First, the intervals are equal: the difference between 150 cm and 160 cm is the same 10 cm as between 180 cm and 190 cm. Second, zero means something real. A height of 0 cm means no height at all. This isn’t an arbitrary starting point like 0°F; it’s a true absence of the thing being measured.
That true zero is what separates ratio from interval. Because height has one, you can make meaningful ratio statements: a person who is 180 cm tall is exactly twice as tall as a child who is 90 cm. You can’t make the same kind of statement with interval data. Saying 40°C is “twice as hot” as 20°C doesn’t hold up, because the zero point on the Celsius scale is arbitrary.
Height is also continuous, meaning it can take any value within a range rather than being restricted to whole numbers. Clinical protocols measure adult height to the nearest millimeter using standardized equipment, and researchers routinely average three separate readings to get a precise result. This level of precision reflects the continuous, ratio nature of the variable.
Why Height Isn’t Nominal or Ordinal
Nominal data is just labels. If you tried to treat height as nominal, you’d be saying that 170 cm and 171 cm are simply “different” with no way to determine which is greater. That obviously doesn’t match reality.
Ordinal data lets you rank values but strips away the meaning of the gaps between them. T-shirt sizes (S, M, L, XL) are ordinal: you know XL is bigger than M, but you can’t say the jump from M to L equals the jump from L to XL. Height doesn’t have this limitation. The difference between any two heights is precisely quantifiable in centimeters or inches, and those differences are consistent across the entire scale.
When Height Gets Converted to Ordinal Data
Researchers sometimes deliberately convert height from a continuous ratio variable into ordered categories. The BMJ describes a common approach: splitting height into groups like “short,” “average,” and “tall” using cutoff points. A pediatrician might classify a child’s height as “below the 5th percentile” or “above the 95th percentile” rather than tracking the exact measurement.
Once you do this, height becomes ordinal. You know that “tall” is greater than “average,” but the categories no longer tell you by how much. This conversion always loses information. You sacrifice precision in exchange for simpler groupings that can be easier to work with in certain analyses or clinical decisions.
You could even push it further and turn height into nominal data. Imagine a database with a field labeled “meets height requirement: yes/no” for an amusement park ride. At that point, height is reduced to unordered categories, and even the ranking information is gone.
Why the Distinction Matters
The scale of measurement determines which statistical tools you can use. With ratio data like raw height measurements, everything is available to you: means, standard deviations, correlations, multiplication, division. You can say the average height in a sample is 168.3 cm and that number is genuinely informative.
If you convert height into ordinal categories, you lose access to most of those tools. You can’t calculate a meaningful average of “short,” “average,” and “tall.” You’re limited to rank-based statistics like medians and percentiles. And if you somehow reduced height to nominal categories, you’d be restricted to counting how many people fall into each group.
So when a stats question asks whether height is nominal or ordinal, the answer is neither. Height in its natural, measured form is ratio. It only becomes ordinal or nominal if someone deliberately bins it into categories, and at that point, you’re no longer measuring height itself. You’re measuring a simplified version of it.