Statistics for Data Work

About this course

Numbers move every week. Statistics tells you which changes are real and which are noise, how sure you can be, and how big a test needs to be before anyone looks at the result. This course teaches the part of statistics that data work uses, with business questions instead of formulas for their own sake: describing skewed data honestly, putting an interval around every estimate, reading a p-value without fooling yourself, designing and reading an A/B test, fitting a regression and knowing what it can and cannot tell you.

You work with Tomás, the data analyst at Mangrove Market, a fictional grocery chain with 320 stores and a delivery app. Did the new checkout really lift conversion? How sure can a stock count in 40 stores be about all 320? How many minutes does rain add to a delivery? Why did the ten worst stores improve after a visit that changed nothing? Each idea starts with a picture and a small simulation before any formula, and every method is shown the spreadsheet way and the Python way.

School algebra is enough. Auto-checked exercises in your browser compute medians, intervals, test statistics, sample sizes and regression lines, and the capstone has you analyse a real-looking checkout experiment and write the recommendation a leadership team can act on.

What you will be able to do

• Describe data with the right summary, spot skew, outliers and mixed groups, and always report the count
• Explain sampling, standard errors and confidence intervals in plain words, and compute an interval for a mean or a proportion
• Run and read a hypothesis test, and separate statistical significance from a difference that matters to the business
• Design an A/B test with a unit, a primary metric, guardrails and a sample size, and read the result honestly without peeking
• Fit and interpret a regression, read its output table and explain why correlation is not cause
• Recognise Simpson's paradox, survivorship bias, regression to the mean and multiple comparisons before they reach a decision

Syllabus

6 stops, about 3.5 h in total. Each stop fills in when you complete the lesson.

1. 1

   Describing data honestly

   Before any test or model, someone asks "what does a typical order look like?", and the first number you give shapes every decision after it. This lesson teaches you to describe a column of numbers so that the summary matches the data: the right centre, the right spread, the outliers, the shape, and how many values it rests on.

   35 minIntermediateEpisode2 exercises

   Descriptive statisticsStatistical trapsPython

2. 2

   Samples, uncertainty and confidence intervals

   You rarely measure everything. You count stock in 40 stores, audit 400 orders or survey 1,000 customers, and then someone asks what that says about all of Mangrove Market. This lesson shows you how a fair sample speaks for the whole, how much it can wobble, and how to report a number with an honest range around it.

   35 minIntermediateEpisode2 exercises

   Confidence intervalsProbabilityDescriptive statisticsPython

3. 3

   Hypothesis tests without the fog

   A hypothesis test answers one narrow question: if nothing had changed, how surprising would data like ours be? This lesson builds that question from a simulation, runs the two tests analysts use most (the two-proportion z-test and Welch's t-test) in a spreadsheet and in Python, and shows how to report an effect with its interval so nobody mistakes "significant" for "important".

   35 minIntermediateEpisode1 exercise

   Hypothesis testingA/B testingConfidence intervalsPython

4. 4

   Designing and reading an A/B test

   Most A/B tests go wrong before the first customer sees the variant, or in the week someone looks at the result too early. This lesson is the statistics under a trustworthy test: the unit you randomise, where the sample size formula comes from, the checks you run before reading, why peeking breaks the 5% promise, and how to read the result against a rule you wrote first.

   40 minIntermediateEpisode2 exercises

   A/B testingHypothesis testingConfidence intervalsPython

5. 5

   Regression and correlation

   Riders say rainy evenings are slower, and everyone knows far orders take longer. This lesson turns those hunches into numbers you can defend: how tightly two things move together, how many minutes each extra kilometre and each rain shower add, how sure you can be of those minutes, and when a fitted line says nothing about cause.

   40 minIntermediateEpisode1 exercise

   RegressionDescriptive statisticsStatistical trapsPython

6. 6

   Traps that fool smart people

   Every number in a report can be correct and the story it tells can still be false. This last lesson of the course shows six traps that catch experienced analysts (Simpson's paradox, regression to the mean, many comparisons, survivorship, selection and ignored base rates) and the check that catches each one before it reaches a decision.

   35 minIntermediateEpisode2 exercises

   Statistical trapsHypothesis testingDescriptive statisticsPython

7. Capstone: Did the new checkout really work?

   A realistic project with a rubric and a model answer.

8. Final exam and certificate

   10 questions drawn at random, 70% to pass.

Capstone project

Check what you know

Questions