T1.1 Data, variables, and the map of this course Start ★ 0
T1.1Playlist 1, episode 1

Label or amount?

The first question in statistics. You ask it about every column of data, before any chart or formula. Make a guess first: you will see it again at the end.

Builds on: nothing, start hereLeads to: T1.2 and T5.1
1Lina’s notebook

One cup, one line

Each time Lina sells a cup, she writes one line. Every line answers the same five questions.

A dataset is a table: one row for each thing, one column for each question.
In French: un jeu de données
1Lina’s notebook

Four words, one table

Statisticians give each part of the table a name. Tap a word and watch the notebook.

In French: élément, observation, variable, valeur
2Labels and amounts

Same digits, two jobs

Numbers you meet every day. Some name a thing, some say how much. Tap Name or Amount on each card.

0 of 6
The same digits can name a thing or measure an amount. Ask what the number does, not what it looks like.
2Labels and amounts

Can you do math with it?

Meet Tag and Ruler. Pick a column, and the machine adds the answers from cup 1 and cup 2.

TagLabels. Answers “which one?” Categorical.
RulerAmounts. Answers “how much?” or “how many?” Quantitative.
If the math means something, it is an amount: quantitative. If it is nonsense, it is a label: categorical.
In French: variable qualitative, variable quantitative
?
makes sense?
2Labels and amounts

Count or measure?

Ruler’s numbers come in two kinds. Drag both sliders and watch how each one moves.

Counting jumps from value to value: discrete. Measuring glides anywhere: continuous.
In French: variable discrète, variable continue
Toppings are counted discrete
Wait time is measured continuous
3Four levels of measurement

Four levels, four powers

Climb the stairs. Each step unlocks one more thing you are allowed to say about two values.

Nominal and ordinal are Tag’s levels. Interval and ratio are Ruler’s.
In French: échelle nominale, ordinale, d’intervalle, de rapport
3Four levels of measurement

Is 20° twice as warm as 10°?

Today it is 20°, yesterday it was 10°. Switch the unit and watch “twice” fall apart.

No real zero, no “twice”. Degrees Celsius are an interval scale. Kelvin starts at “no heat at all”, so it is a ratio scale.
today ÷ yesterday2.00×

4Photo or film

Photo or film?

Five stands sell at the Saturday market, week after week. Look across all stands on one day, or follow one stand over time.

Photo: many things, one moment. That is cross-sectional data. Film: one thing, many moments. That is a time series.
In French: données transversales, série chronologique

5Population and sample

From my cups to everyone

Until now, Lina described her own six cups. Now she wants to know about people she has never met.

The population is everyone (or everything) your question is about.
In French: la population

5Population and sample

Who is the population?

Every question is about some group. Find the group for these four questions.

A population can be people, batteries, apples, even soup: all the elements your question is about.
5Population and sample

You don’t eat the whole pot

A cook checks the salt with one spoonful. Try the three ways to check this pot.

Checking everyone is a census. Checking a few is a sample. Stirring first means picking the sample at random, so the spoon looks like the pot.
Too many: millions of people Too slow and too costly Some tests destroy what they test
In French: recensement, échantillon
5Population and sample

Two numbers, two names

Each question has a number for the whole population and a number from the sample. They get different names.

Population Parameter
Sample Statistic
The parameter stays fixed, but nobody knows it. The statistic is known, but it changes with every new sample.
In French: paramètre, statistique
5Population and sample

Short names for the numbers

Statisticians write these numbers as symbols. Tap a row to see Lina’s version. No calculating today, just names.

Population: μ, σ, p. Sample: x̄, s, p̄. The bar on x̄ and p̄ means “from the sample”.
5Population and sample

One jar, many spoons

Back to Lina’s taro question. All Québec fans are a jar. Each time Lina asks 20 new fans, she takes a new spoon from the same jar.

The jar’s number (the parameter) never moves. The spoon’s number (the statistic) jumps from scoop to scoop.
Instrument: Jar and spoons. Back in T1.5, T3.3 and Themes 5 to 8.
In French: fluctuation d’échantillonnage
All fans: the population
20 asked: the sample
5Population and sample

See it, or guess it?

Describing tells what is in your hands. Inferring makes a careful guess about the whole jar, and says how sure it is.

Themes 1 to 4 teach you to see. Themes 5 to 8 teach you to guess well.
In French: statistique descriptive, statistique inférentielle

See it

Describe the data in your hands. “6 of my 20 fans chose taro.”

Guess it

Infer something about the whole jar. “About 30% of all fans, give or take.”

6The trap

Digits in disguise

Lina writes 1 for “drink here” and 2 for “to go”. Her spreadsheet happily averages them.

Before you average, ask: is this number an amount, or a name tag?

Statisticians have argued about this for a long time. A famous 1953 article by Frederic Lord, about the numbers on football jerseys, started a debate that is still going.

Average of the codes
1.33
Share of cups drunk here
67%

7The map

Your map of the whole course

Tag’s line: labels, counts, proportions p̄Ruler’s line: amounts, averages x̄, regressionTwo lines, every theme. Tap any stop.
8Practice

Practice arcade

Seven short levels, a boss where you build the answer, and a detective tool. Your stars stay in this browser.

9Remember

Six things to remember, and one trap

Sources and textbook links

The definitions on this page follow chapter 1 of the course textbook: elements, variables and observations; categorical and quantitative data; the four scales of measurement; cross-sectional and time series data; population, sample, descriptive and inferential statistics. The discrete/continuous distinction follows OpenIntro Statistics, chapter 1. The four levels come from Stevens (1946); Lord (1953) and Velleman and Wilkinson (1993) are the classic critiques. “In French” labels give common French statistical vocabulary, not quotations from the book. All data on this page are invented for teaching.

  1. Anderson, D. R., Sweeney, D. J., Williams, T. A., Camm, J. D., & Cochran, J. J. (2015). Statistiques pour l’économie et la gestion (5e éd., traduction de la 7e édition américaine d’Essentials of Statistics for Business and Economics par C. Borsenberger). De Boeck Supérieur. Chapitre 1.
  2. Stevens, S. S. (1946). On the theory of scales of measurement. Science, 103(2684), 677–680. https://doi.org/10.1126/science.103.2684.677
  3. Lord, F. M. (1953). On the statistical treatment of football numbers. American Psychologist, 8(12), 750–751.
  4. Velleman, P. F., & Wilkinson, L. (1993). Nominal, ordinal, interval, and ratio typologies are misleading. The American Statistician, 47(1), 65–72. https://doi.org/10.1080/00031305.1993.10475938
  5. Diez, D. M., Çetinkaya-Rundel, M., & Barr, C. D. (2019). OpenIntro Statistics (4th ed.). OpenIntro. Chapter 1. https://www.openintro.org/book/os/