My Skill Lab
Data ScienceIntermediate

Statistics for Data Science

From distributions to A/B tests you can trust.

Taught by Niraj Thapa · Analytics Lead

What you'll learn

  • Summarize any dataset with the right measures of center and spread, and spot when the mean lies to you
  • Recognize common distributions and explain why the normal curve appears so often in practice
  • Apply the central limit theorem to reason about sample means and standard error
  • Build and correctly interpret confidence intervals instead of fixating on a single point estimate
  • Run a hypothesis test end to end and explain a p-value to a stakeholder without overclaiming
  • Design an A/B test with adequate power and avoid peeking, multiple-testing, and other classic traps

About this course

Statistics is the grammar of data science, and this course teaches it the way working analysts actually use it: intuition first, formulas second. You will move from describing a single column of numbers to designing and reading A/B tests whose results you can defend, learning exactly what a p-value does and does not promise along the way. By the end you will reason fluently about uncertainty, sampling, and statistical significance without drowning in proofs.

Course content

8 lessons
  1. 1Describing Data: Center, Spread, and ShapeTurn a raw column of numbers into an honest one-sentence summary using mean, median, spread, and shape.Preview
  2. 2Distributions: The Shapes Data TakesMeet the handful of distributions that describe most real data and learn what each one's parameters mean.Preview
  3. 3Sampling: Learning About a Whole From a PartUnderstand why we sample, how bias creeps in, and the difference between a population and a sample.
  4. 4The Central Limit Theorem and Standard ErrorThe one result that makes inference possible: why sample means are normal even when the data isn't.
  5. 5Confidence Intervals: Honest Error BarsReplace a single point estimate with a range, and learn what '95% confident' really means.
  6. 6Hypothesis Testing: Asking Data a Yes/No QuestionThe structured way to decide whether an observed effect is real or just noise.
  7. 7P-values: What They Mean and What They Don'tDemystify the most misunderstood number in statistics and learn to read it without overclaiming.
  8. 8A/B Testing: Designing Experiments You Can TrustPut it all together to design, power, and read an A/B test while dodging the classic pitfalls.