Clearstistics

Statistics · epidemiology · the maths underneath

Statistics, explained simply.

Clearstistics explains statistics, epidemiology and the maths behind them — one concept at a time, in plain English, with visuals that do the explaining. No formulas thrown at you, no prior knowledge assumed.

Watch in order ↗ Take the quiz Subscribe

Start here · new to the channel? This is the front door — then watch the series in order ↗.

How sharp is your statistics sense? · 6 questions

No sign-up, and nothing leaves your browser. You get a score — and if you stumble, we point you at the exact video to start with.

Your progress

Tick off videos as you watch them, take the quiz under each one, and your level score grows. Saved only in this browser — nothing is sent anywhere. Reset progress.

Videos watched 0 / 14

Quizzes taken 0 / 15

Level score 0 / 50

No quizzes taken yet

The videos · 14

Newest first. Every video stands on its own, but they also build on each other — new here? Watch the series in order ↗. Subtitles are on the YouTube player (the CC button), and under every video sits a three-question quiz. Each one also has its own page with the answer, the chapters and the full transcript.

Video 14 · 5:12 · Oct 2026

Types of Data: Categorical vs Continuous

One school fun run, eight runners, six columns. A spreadsheet will happily average the bib numbers (148.5) and the finishing times (27:30), but only one of those averages tells you anything about the runners. Swap two runners' bib numbers and nothing about the race changes; swap their times and a different runner wins. That is the difference between a category and an amount. This video gives you the average test for any column, and shows categorical data (names for groups), ordinal data (an order without equal steps), dichotomous data (two options), and amounts that are continuous (measured) or discrete (counted). The habit to take away: before you average a column, ask what each number stands for.

What is the difference between categorical and continuous data? → · Open on YouTube ↗

Quiz — 3 questions

Video 13 · 5:26 · Oct 2026

Misleading Graphs: Why the Axis Matters

One town, one month of buses. On one chart, bus 4 looks twice as late as bus 7; on another chart of the very same two numbers there is hardly any difference. No number changed: the axis did. This video shows why a bar's length is its number, so a bar chart starts at zero; why a line chart may zoom in, and why your eye still makes the change look bigger; and how two scales on one chart can make almost any two lines look like partners. The one habit to take away: read the axis, where it starts and where it ends, before you read the shape.

How can a graph mislead you without changing the numbers? → · Open on YouTube ↗

Quiz — 3 questions

Video 12 · 7:02 · Sep 2026

Correlation vs Causation: Do Big Feet Read Better?

At one primary school, children with bigger feet read better: the correlation is plus 0.75. Everyone knows the slogan "correlation does not imply causation", and yet in one study two in three psychology students who had already passed statistics still found a correlation-only study convincing. This video gives you one habit that works instead: every link comes with three stories. A causes B, B causes A, or a third factor pushes on both. Colour the same sixty dots by the children's age and the link disappears inside every age group. Then three rules: to show a cause, hand it out by lottery; read the verb in the headline; and a link is still a clue.

Why does correlation not imply causation? → · Open on YouTube ↗

Quiz — 3 questions

Video 11 · 6:46 · Sep 2026

Correlation: What r Really Tells You

Correlation is one of the most quoted numbers in statistics, and one of the most misread. In everyday talk it means any link at all, so a correlation of zero sounds like no link whatsoever. A ball thrown straight up follows its path perfectly, and its correlation is exactly zero. This video builds r from scratch with twenty-five athletes, their sprint times and their long jumps: every athlete pulls r up or down, and r is roughly the average pull. Then three rules for reading it: r only sees straight lines, your eye cannot measure it, and one far dot can move it a long way.

What does the correlation coefficient r actually measure? → · Open on YouTube ↗

Quiz — 3 questions

Video 10 · 6:58 · Sep 2026

Scatterplots: Two Variables at Once

Twenty-four students, two numbers each: hours on the phone and hours asleep. Two lists tell you almost nothing; one picture keeps every student in the room. This video builds the scatterplot from scratch and then springs the trap: two classes that share every single number — the same average, the same spread, even the same histograms — and look nothing alike. The difference is the pairing, and a summary of one column at a time cannot see it. Then the four steps for reading any cloud of dots, including the one almost everyone gets wrong: strength is how tightly the dots keep to the band, not how steeply it falls.

How do you read a scatterplot? → · Open on YouTube ↗

Quiz — 3 questions

Video 9 · 5:43 · Sep 2026

Variance: Why We Divide by n − 1

Two players, eight turns each, and exactly the same average reaction time: 280 milliseconds. On paper they are identical; watch them play and you would never swap them. This is the ruler that tells them apart. Why the obvious idea always gives zero, why squaring fixes it and punishes a big miss four times as hard, and why the square root brings you back to milliseconds. Then the part almost nobody shows: slide the point you measure from, watch the total area bottom out exactly at your own average, and see why a sample always understates its own spread. That is what n − 1 repairs.

What is the difference between variance and standard deviation? → · Open on YouTube ↗

Quiz — 3 questions

Video 8 · 6:21 · Sep 2026

Histograms: Bin Width Changes the Story

Twenty-eight students, one class, how many minutes each needs to get to school — and the list tells you almost nothing until you cut it into bins and stack what lands in each. That stack is a histogram, and its height means one thing: the count. Then the part nobody mentions: the bin width is a choice. Five minutes shows two hills, twenty shows one slope, one shows a comb — same numbers, three stories. Plus the rule of thumb, the habit that beats it, and the mean that lands where nobody is.

What is a histogram, and how do you read one? → · Open on YouTube ↗

Quiz — 3 questions

Video 7 · 6:51 · Sep 2026

The Log Transformation: One Axis Trick

Three hundred incomes, one billionaire, and a page that gives up: everyone collapses into a single smear against the wall. Then one thing changes — not the data, the ruler — and the smear opens into a bell. What a logarithmic scale is, why equal steps mean equal ratios, which average the picture balances on, where you meet this ruler in the wild, and the three limits you have to respect.

What is a log scale, and when do you use one? → · Open on YouTube ↗

Quiz — 3 questions

Video 5 · 5:20 · Aug 2026

Mean vs Median: Which One to Use When?

The mean is the balance point of your data, the median is the middle person. Twenty incomes on a seesaw, then a billionaire walks in: the mean rockets from 41,000 to almost two million while the median slides exactly one seat. Both are correct — they answer different questions. By the end you know which one to reach for.

Mean or median — which one should you use? → · Open on YouTube ↗

Quiz — 3 questions

Coming: a full beginner course · with workbook

On top of these videos we're building a complete statistics course for beginners, with a workbook to practise on — online with instant marking, as a PDF, and in print. The quizzes and the progress bar above are its first pieces.

220 questions on five levels123 written questions with model answersevery number checked in code against the data in the videos14 videos, 15 chapters so far

Want to hear when it's ready — or shape what goes in? Say hello at info@clearstistics.com.

Coming: a teacher licence · follow your class

One key for the teacher, the whole class underneath it. You see who has watched which video, how far each student has come in the workbook, and what they scored on every quiz — per student and per chapter. Progress and score are two different things, so you get both.

Teacher key

Your class

28 students · 7 lessons

You invite your students; each one gets their own account, so their answers and progress stay theirs. You see the whole class in one view.

  • Amara92%9/10
  • Joris78%6/10
  • Lieke64%8/10
  • Sam45%5/10
  • Noor31%7/10

For a class

€15 about $17per student, per year · early bird · excl. VAT

An online workbook account for every student, the class view above, and the answer key and quiz variants for you. Ten students minimum. Invoice on request.

Workbook, on its own

€18 about $21one-off, PDF · early bird · incl. VAT

The book for a single reader, with model answers. Every number in it is checked in code against the data in the videos. In print, in colour, for €26.

Indicative prices for something still being built. Nothing is for sale yet and there is no checkout. If you teach statistics and want to try it with a class first, write to info@clearstistics.com — pilot classrooms shape what goes in. Early-bird prices; they rise by a third once all 15 chapters are finished. The full picture is on the course page.

What Clearstistics is

A YouTube channel about statistics and methodology — with an epidemiological slant — and the maths that sits underneath. Made for students, researchers brushing up, and anyone trying to read a scientific paper without getting lost.

What you'll find

The building blocks: the normal distribution, mean, median and mode, standard deviation and standard error, confidence intervals and p-values. And the methods behind medical research: relative risk, odds ratios, survival analysis and the models that sit underneath them.

How the videos are made

Every chart is drawn in code, straight from the numbers — so what you see on screen is what was actually computed, not an illustration of it. Greek letters always come with their name (μ is “mu”, σ is “sigma”), and every spoken number is checked against what's in the picture before a video goes out.

One colour language, every video

The same idea always wears the same colour, so once you've seen it once you'll recognise it everywhere:

mean μ (mu) spread σ (sigma) median & uncertainty mode highlight error

Questions

Got a question, or a concept that never quite clicked? Leave a comment under the video — they're all read, and they often become the next video. Email works too: info@clearstistics.com.