Levels of measurement describe how data values are organized and what their numbers or labels actually mean. This matters because the type of measurement determines which comparisons, calculations, graphs, and statistical tests are appropriate. The four main levels are nominal, ordinal, interval, and ratio, often arranged like a ladder from least to most mathematical information.
Choosing the correct level helps prevent misleading averages, invalid comparisons, and incorrect conclusions.
Key Facts
- Nominal data use categories with no natural order, such as blood type, major, or eye color.
- Ordinal data have a meaningful order, but differences between ranks are not guaranteed to be equal.
- Interval data have equal spacing between values, but no true zero, so differences are meaningful but ratios are not.
- Ratio data have equal spacing and a true zero, so differences and ratios are meaningful.
- Allowed operations increase by level: nominal = classify, ordinal = rank, interval = add and subtract, ratio = multiply and divide.
- Common statistics: nominal uses mode and proportions, ordinal uses median and percentiles, interval and ratio often use mean, standard deviation, correlation, and regression.
Vocabulary
- Nominal level
- A level of measurement where values are names or categories with no natural order.
- Ordinal level
- A level of measurement where values can be ranked, but the size of the gaps between ranks is not necessarily equal.
- Interval level
- A level of measurement with ordered values and equal intervals, but without a true zero point.
- Ratio level
- A level of measurement with ordered values, equal intervals, and a true zero that means none of the quantity is present.
- True zero
- A zero value that represents the complete absence of the measured quantity.
Common Mistakes to Avoid
- Treating nominal labels as numbers, such as averaging jersey numbers or ZIP codes. These numbers identify categories and do not represent amounts.
- Assuming ordinal ranks have equal spacing, such as saying the gap between 1st and 2nd place equals the gap between 2nd and 3rd. Ranks show order but not exact distance.
- Using ratios with interval data, such as saying 40 degrees Celsius is twice as hot as 20 degrees Celsius. Celsius has no true zero, so ratios are not physically meaningful.
- Choosing a statistical test before identifying the measurement level. The level of measurement controls whether statistics like mean, standard deviation, and correlation are appropriate.
Practice Questions
- 1 A survey records students' favorite school subject: math, biology, history, art, or music. Identify the level of measurement and name one appropriate summary statistic.
- 2 A race records finishing places as 1st, 2nd, 3rd, 4th, and 5th. Identify the level of measurement and explain whether calculating the average finishing place is always meaningful.
- 3 A data set includes temperature in Celsius, height in centimeters, satisfaction rating from 1 to 5, and car color. Classify each variable as nominal, ordinal, interval, or ratio, and explain which one allows the statement twice as much.