Skip to content
Statculator

Learn statistics online

You can learn statistics online for free with the guides below. Each one explains a single idea in plain language, gives the formula, works a full example by hand and links to a calculator so you can check every step with your own data.

If you are new to the subject, follow the five units in order: describing data first, then samples and inference, probability and the normal curve, hypothesis testing, and finally comparing means with t-tests and ANOVA. If you already know what you need, jump to a topic or a single guide.

  • 5 units, 14 guides
  • About 3 hours of reading
  • Introductory level
  • Free, no sign-up

By Hira Noor, BS Mathematics · Updated

Where to start

Start with Unit 1 and Measures of Central Tendency: Mean, Median and Mode. Read the guide, redo its worked example on paper, then enter the same numbers into the linked calculator and compare each step. Repeat that read, solve, check loop for every guide in the path.

A learning path in five units

The units follow the order most introductory statistics courses use, so each one builds on the last. Every unit lists the guides to read, the calculators to practise with and, where they apply, the statistical tables you will meet in exams.

  1. Unit 1: Describe a dataset

    Summarise a list of numbers by its center and spread, and know which formula your problem needs.

    1. Measures of Central Tendency: Mean, Median and Mode11 min read
    2. Sample vs Population Standard Deviation: Which Should You Use?8 min read
  2. Unit 2: Samples, populations and inference

    Tell the data you measured apart from the population you want to describe, and see why estimates carry uncertainty.

    1. Population vs Sample: Definitions, Differences and Examples10 min read
    2. Parameter vs Statistic: Difference, Symbols and Examples9 min read
    3. Descriptive vs Inferential Statistics: Differences and Examples9 min read
  3. Unit 3: Probability and the normal curve

    Find probabilities for events, binomial counts and normally distributed values, and convert a value to a z-score.

    There is no written guide for this unit yet. Each calculator page below has the formula, a step-by-step solution and a worked example, so use those as the lesson.

  4. Unit 4: Hypothesis testing

    Understand what a p-value measures, the two errors a test can make, and how to pick the right test for your data.

    1. What Is a P-Value? Meaning, Calculation and Common Mistakes11 min read
    2. Type I vs Type II Errors (Type 1 vs Type 2): Definitions and Examples14 min read
    3. How to Choose the Right Statistical Test13 min read
  5. Unit 5: Compare means with t-tests and ANOVA

    Run and interpret the one-sample, independent and paired t-tests, then extend the idea to three or more groups with ANOVA.

    1. One-Sample t-Test: Formula, Assumptions and Worked Example10 min read
    2. Independent Samples t-Test: Formula, Assumptions and Worked Example12 min read
    3. Paired t-Test: Formula, Assumptions and Worked Example11 min read
    4. One-Way ANOVA: Definition, Formulas and a Worked Example by Hand14 min read
    5. Two-Way ANOVA: Main Effects, Interaction and a Worked Example15 min read
    6. One-Way vs Two-Way ANOVA: Differences and When to Use Each11 min read

All statistics guides

How to figure out a statistics problem

Most statistics problems, from homework questions to a real dataset, can be worked through with the same five steps. The guides above go deeper on each one.

  1. Name the question. Are you describing data, estimating a population value, testing a claim or predicting one variable from another? The answer decides the family of methods.
  2. Identify the data. Note whether each variable is numerical or categorical, how many groups there are, and whether the observations are independent or paired. How to Choose the Right Statistical Test turns these answers into a test.
  3. Write down the formula and the hypotheses. For a test, state the null and alternative hypotheses and the significance level, usually α = 0.05, before you calculate anything.
  4. Calculate, then check. Work the numbers by hand once so you understand them, then repeat the calculation in a calculator to catch arithmetic slips.
  5. Interpret in context. Report the result in the language of the question, check the assumptions of the method, and say what the result does not show. A p-value alone is not a conclusion.

What math do you need to learn statistics?

An introductory statistics course needs arithmetic, order of operations, fractions and decimals, square roots and basic algebra, such as rearranging a formula to solve for one value. You will also meet summation notation (Σ), which simply means “add these values up”.

Calculus is not needed for an introductory course. It becomes useful later, in probability theory and mathematical statistics, where continuous distributions are defined with integrals.

How these guides are written

  • One idea per guide, with a short summary and key takeaways at the top so you can check whether it answers your question before reading on.
  • A worked example solved by hand, with every intermediate value shown, then the same numbers in a calculator so you can verify the result.
  • Notation stated, not assumed. Where textbooks differ, for example on dividing by n − 1 or n, the guide names the convention it uses.
  • Further reading at the end of each guide points to the textbooks and references behind it.

The guides are written and reviewed by the Statculator editorial team. The calculators they link to are checked against more than 200 reference values from SciPy and statsmodels; the Methodology page explains how.

Learning statistics: FAQs

Can I learn statistics by myself?

Yes. Introductory statistics is well suited to self-study because each idea can be practised with small datasets. Work through the five units above in order, redo every worked example on paper, and use the linked calculators to check your answers. If you are preparing for an exam, also practise reading the z table and t-table, because many exams do not allow software.

How long does it take to learn basic statistics?

It depends on your goal and how much time you put in. A university introductory statistics course usually runs for one semester and covers roughly the material in the five units on this page. Reading one guide and redoing its example is a sensible study session.

Do I need R or Python to learn statistics?

No. The concepts, formulas and interpretation can be learned without any programming. R and Python become worth learning once you analyse real datasets with many variables or need a reproducible analysis for a report or thesis. Until then, a calculator that shows its working is the faster way to learn the method. After the five units, correlation and regression are the usual next topic, and the Correlation Coefficient Calculator and Linear Regression Calculator are good places to start.

Is statistics hard to learn?

The arithmetic is usually the easy part. Most students find interpretation harder: what a p-value does and does not mean, which test fits a design, and when an assumption is broken. That is why the guides here cover interpretation and common pitfalls alongside the arithmetic.

Are these statistics guides free?

Yes. Every guide, calculator and statistical table on Statculator is free, with no account, sign-up or paywall.