Hypothesis Testing in Python
| Course Details | ||||||||
|---|---|---|---|---|---|---|---|---|
|
||||||||
| Languages | ||||||||
|
||||||||
| Outcomes | ||||||||
|
||||||||
| Price & Terms | ||||||||
|
||||||||
| Technical | ||||||||
|
||||||||
| Specification | ||||||||
|
||||||||
Learn to choose, run and interpret t-tests, ANOVA, proportion and chi-square tests in Python, and when to use non-parametric tests instead.
Hypothesis testing gives analysts a disciplined way to judge whether a pattern in their data reflects a real effect or simple chance. The lessons work through that judgment in Python: framing null and alternative hypotheses, setting a significance level, reading p-values, and choosing among t-tests, ANOVA, proportion tests and chi-square tests. Real-world data on Stack Overflow user feedback and medical supply shipments anchors the exercises, and non-parametric alternatives show what to do when a test’s assumptions fail.
What You Will Achieve
- Set up null and alternative hypotheses, a significance level and the correct tail direction for an analytical question
- Pick a suitable test, such as a t-test, ANOVA, proportion test or chi-square test, for a given question and data type
- Interpret p-values, confidence intervals and test statistics from Python libraries to decide whether to reject the null hypothesis
- Check sample size and other assumptions, then switch to a non-parametric test when they are not met
Program Overview
Hypothesis Testing Fundamentals
A one-sample proportion test serves as the worked example for the full testing workflow and the errors a conclusion can carry.
- z-scores and p-values
- Left-tailed, right-tailed and two-tailed tests
- Statistical significance and confidence intervals
- Type I and Type II errors
Parametric Tests for Means and Proportions
Means and proportions are compared across two or more groups using the t-distribution and the chi-square distribution.
- Two-sample and paired t-tests
- ANOVA and pairwise t-tests
- One-sample and two-sample proportion tests
- Chi-square tests of independence and goodness of fit
Test Assumptions and Non-Parametric Tests
The assumptions behind parametric tests are checked, with alternatives for samples that break them.
- Common assumptions and sample size checks
- Wilcoxon signed-rank test
- Wilcoxon-Mann-Whitney and Kruskal-Wallis tests
Certificate Awarded (Paid)
Job Outlook
Hypothesis testing in Python supports data analysis and data science work, such as comparing how groups perform, validating A/B test results and checking survey or supply-chain data before drawing conclusions.
Good for: Analysts working in Python who need to decide which significance test fits a question and read its output with confidence.
Keep in mind: Most of the course centers on classical parametric tests; non-parametric options get one closing chapter on Wilcoxon and Kruskal-Wallis tests.
Bottom line: A focused Python course that pairs each common test with the assumptions behind it and closes with what to use when those assumptions break.
Frequently Asked Questions
Which datasets do the exercises use?
The exercises draw on real-world data, including user feedback collected from Stack Overflow and supply-chain records for shipments of medical supplies, so each test is applied to a practical question.
Which Python functions appear in the lessons?
Lessons use functions such as ttest() for comparing means and proportions_ztest() for comparing two proportions, after first calculating z-scores, test statistics and p-values step by step to show how each result is built.
How are Type I and Type II errors introduced?
The opening chapter compares a hypothesis test to a criminal trial, then explains false positive and false negative errors and what each one means for the conclusion an analyst draws.
Are transcripts or a glossary available?
Yes. The videos have live transcripts that can be shown with the Show transcript option below each video, and a course glossary sits in the resources section beside the lessons.
User Reviews
Write your review
You must be logged in to post a review.
Reviews can only be posted by registered learners. Log in or create a free account, then you will return here to post.
There are no reviews yet.