Difficulty: Beginner | Prerequisites: Section 1.1 (Overview of Statistics)
Once you know what data is (Section 1.1), the next step is learning how to classify it. This matters because the type of data you have determines which statistical methods you are allowed to use. You would not calculate an average of zip codes, for example, even though zip codes are numbers. This section teaches you two classification systems: qualitative vs. quantitative, and the four levels of measurement (nominal, ordinal, interval, ratio). Both will come up repeatedly in every later chapter when you need to choose the right graph, the right summary measure, or the right test.
Data is either qualitative (categories, labels) or quantitative (numbers you can measure or count). Within those, data falls into one of four measurement levels: nominal, ordinal, interval, or ratio. The level determines what mathematical operations make sense and which statistical tools apply.
Qualitative data (categorical data)
Data consisting of attributes, labels, or non-numerical entries that can only be classified into categories. Think of it as data you describe with words, not numbers. Examples: eye colour, college major, place of birth.
Quantitative data (numerical data)
Data consisting of numbers that are measurements or counts, measured naturally with a numerical scale. In simple terms, this is data you can do arithmetic on meaningfully. Examples: age, temperature, weight.
Nominal level of measurement
Qualitative data only. Data is categorised using names, labels, or qualities. No mathematical computation can be performed. Think of it as pure labels with no ranking or order. Example: movie genres (action, comedy, thriller).
Ordinal level of measurement
Data that can be arranged in order or ranked, but the differences between values are not meaningful. In simple terms, you know which is "more" or "less", but you cannot say by how much. Example: a pain scale of 1 to 10 (the gap between 1 and 2 may not equal the gap between 9 and 10).
Interval level of measurement
Quantitative data that can be ordered, and the differences between values are meaningful. Zero is a position on the scale, not the absence of the quantity. Think of it as: you can subtract meaningfully, but you cannot form meaningful ratios because zero does not mean "none". Example: temperature in Fahrenheit (0°F does not mean no temperature).
Ratio level of measurement
Similar to interval, but zero is an inherent zero that implies "none". Ratios of two data values are meaningful, and one value can be expressed as a multiple of another. In simple terms, this is the most informative level. Example: distance (0 feet means no distance; 6 feet is twice 3 feet).
Ask yourself: can I meaningfully add, subtract, or average these values?
Yes: quantitative (age, wait time in minutes, weight).
No: qualitative (hair colour, opinion poll responses, zip codes).
Numbers that act as labels are qualitative, not quantitative. Zip codes and jersey numbers look like numbers but represent categories. You would never average two zip codes.
Hair colour of classmates: qualitative (colour is not a number).
Wait time in minutes at a checkout: quantitative (time is a measurement).
Responses on an opinion poll ("Is global warming real?"): qualitative (these are opinions, not measurements).
Zip code: qualitative, because the numbers are not measurements or counts; they represent locations.
Types of series on Netflix: qualitative, because the type of series is a category, not a number.
The levels build on each other. Each higher level has everything the one below it has, plus more.
Nominal – Labels only, no order.
Example: movie genres (action, comedy, drama). You cannot rank them in a universally meaningful way.
Example: jersey numbers of baseball players. Player 2 is not "lower" than player 12 in any statistical sense.
Ordinal – Labels with a meaningful order, but differences between ranks are not consistent.
Example: top 5 US occupations by job growth (1st, 2nd, 3rd...). You know the ranking, but the gap between 1st and 2nd is not necessarily the same as between 4th and 5th.
Example: pain level 1 to 10. The difference between 1 and 2 may not be the same as the difference between 8 and 9.
Interval – Ordered, differences are meaningful, but there is no true zero.
Example: temperature in Fahrenheit. The difference between 30°F and 35°F is the same 5 degrees as between 35°F and 40°F, but 0°F does not mean "no temperature".
Example: years of Red Sox World Series wins (1903, 1912, ..., 2018). The differences between years are meaningful, but year 0 does not mean "no time".
Ratio – Ordered, differences are meaningful, and zero means the absence of the quantity. Ratios are meaningful.
Example: number of people at Disney World. Zero means nobody; 200 is twice 100.
Example: heart rate of a student during an exam. Zero heart rate means no heartbeat.
Example: distance between two books (3 ft, 9 ft, 18 ft). Zero means no distance; 18 ft is twice 9 ft.
Can the data be ordered? No → Nominal.
Are differences between values meaningful? No → Ordinal.
Does zero mean "none" (an inherent zero)? No → Interval. Yes → Ratio.
Knowing your data's level of measurement prevents you from running the wrong analysis. In a business context, you might survey customer satisfaction on a 1–5 scale (ordinal) and be tempted to average the scores. Statisticians debate whether this is valid because ordinal gaps are not guaranteed to be equal. In healthcare, body temperature in Fahrenheit (interval) can be subtracted meaningfully, but saying "100°F is twice as hot as 50°F" is not correct because the Fahrenheit zero is arbitrary.
Students often assume that if something is a number, it must be quantitative. Zip codes, jersey numbers, and phone numbers are all numerical but qualitative, because arithmetic on them is meaningless.
Students frequently mix up interval and ratio. The key question is whether zero means "nothing". Temperature in Fahrenheit: 0°F is not the absence of heat, so it is interval. Distance: 0 feet is the absence of distance, so it is ratio.
Ordinal data tempts students into calculating averages. Be cautious: the difference between rank 1 and rank 2 may not equal the difference between rank 4 and rank 5.
Students sometimes think nominal data cannot involve numbers at all. It can (jersey numbers, zip codes), but the numbers serve as labels, not measurements.
⚠️ Expect questions that give you a data example and ask you to classify it as qualitative or quantitative, and then identify its level of measurement (nominal, ordinal, interval, ratio).
⚠️ The zip code question is a perennial exam favourite. Zip codes are qualitative and nominal, even though they look like numbers.
⚠️ Be able to distinguish interval from ratio by asking: "Does zero mean none?" Temperature in Fahrenheit = interval. Temperature in Kelvin = ratio (0 K is absolute zero, true absence of thermal energy).
⚠️ Jersey numbers and similar label-numbers are tested frequently. They are nominal.
True or False: Zip codes are quantitative data because they are made up of numbers.
Fill in the blank: Data that can be ranked but where differences between values are not meaningful is at the ______ level.
True or False: Ratio-level data has an inherent zero.
Fill in the blank: Eye colour is an example of ______ data.
True or False: 0°F means there is no temperature, making Fahrenheit a ratio scale.
Answers: 1. False (qualitative/nominal). 2. Ordinal. 3. True. 4. Qualitative (nominal). 5. False (0°F is a position on the scale, not the absence of temperature; Fahrenheit is interval).
Q: A UCF instructor asks her students if they have scholarships. Is the response quantitative or qualitative?
A: Qualitative. The response is "Yes" or "No", which are categories, not numerical measurements.
Q: Classify the following data by level of measurement: number of people at Disney World.
A: Ratio. The data can be ordered, differences are meaningful, and zero is an inherent zero (zero people means nobody is there). You can also form meaningful ratios (200 people is twice 100 people).
Q: Classify the following: top 5 US occupations with the most job growth.
A: Ordinal. The data has a meaningful ranking, but the difference in job growth between rank 1 and rank 2 is not necessarily the same as between rank 4 and rank 5.
Q: Classify the following: body temperature of a patient in degrees Fahrenheit.
A: Interval. Temperature can be ordered and differences are meaningful, but 0°F does not mean the absence of temperature. You cannot say 100°F is "twice as hot" as 50°F.
Q: Why are jersey numbers of baseball players classified as nominal rather than ordinal?
A: Because jersey numbers are labels with no inherent ranking. Player 2 is not statistically "lower" or "less" than player 12. The numbers identify players, nothing more.
Data classification feeds directly into choosing the right descriptive tools (Chapter 2): bar charts and pie charts suit qualitative data, while histograms and stem-and-leaf plots suit quantitative data. It also matters for inferential statistics later on: different hypothesis tests apply to different data types (chi-square for categorical data, t-tests for numerical data).
qualitative data, quantitative data, categorical data, numerical data, nominal ordinal interval ratio, levels of measurement, data classification statistics, types of data statistics, zip code qualitative or quantitative, interval vs ratio, measurement scales statistics, principles of statistics chapter 1 section 1.2