Learn why hypothesis testing is useful, and step through the workflow for a one sample proportion test. In doing so, you'll encounter important concepts like z-scores, p-p-values, and false negative and false positive errors. The Stack Overflow survey and late medical shipments datasets are introduced.

Hypothesis tests and z-scores

Uses of A/B testing

Calculating the sample mean

Calculating a z-score

p-values

Criminal trials and hypothesis tests

Left tail, right tail, two tails

Calculating p-values

Statistical significance

Decisions from p-values

Calculating confidence intervals

Type I and type II errors

Introduction to Hypothesis Testing

Learn how to test for differences in means between two groups using t-tests, and how to extend this to more than two groups using ANOVA and pairwise t-tests.

Performing t-tests

Hypothesis testing workflow

Two sample mean test statistic

Calculating p-values from t-statistics

Why is t needed?

The t-distribution

From t to p

Paired t-tests

Is pairing needed?

Visualizing the difference

Using t.test()

ANOVA tests

Visualizing many categories

Conducting an ANOVA test

Pairwise t-tests

Two-Sample and ANOVA Tests

Learn how to test for differences in proportions between two groups using proportion tests, extended it to more than two groups with chi-square independence tests, and return to the one sample case with chi-square goodness of fit tests.

One-sample proportion tests

t for proportions?

Test for single proportions

Two-sample proportion tests

Test of two proportions

prop_test() for two samples

Chi-square test of independence

The chi-square distribution

How many tails for chi-square tests?

Performing a chi-square test

Chi-square goodness of fit tests

Visualizing goodness of fit

Performing a goodness of fit test

Proportion Tests

Learn about the assumptions made by parametric hypothesis tests and see how simulation-based and rank-based non-parametric tests can be used when those assumptions aren't met.

Aannames bij hypothesetoetsen

Veelvoorkomende aannames bij hypothesetoetsen

Steekproefgrootte testen

Het raamwerk "Er is maar één test"

There is only one test

Specificeren en hypothesen opstellen

De infer-pijplijn voortzetten

Genereren en berekenen

Geobserveerde statistiek en p-waarde

Niet-parametrische ANOVA en ongepaarde t-toetsen

Simulatie-gebaseerde t-toets

Rangsomme-toetsen

Gefeliciteerd!

Non-Parametric Tests

Late Shipments

Late shipments Bootstrap Distribution

Democratic Presidential Candidates by County

StackOverflow Survey

Late Shipments w/ Freight Cost Group

Met hypothesetoetsen kun je vragen stellen over je gegevenssets en die op een statistisch verantwoorde manier beantwoorden. In deze cursus leer je hoe en wanneer je veelgebruikte toetsen inzet, zoals t-toetsen, proportietoetsen en chi-kwadraattoetsen. Je krijgt grondig inzicht in hoe ze werken en welke aannames eraan ten grondslag liggen. Je leert ook hoe verschillende hypothesetoetsen samenhangen via het "There is only one test"-framework, en je gebruikt niet-parametrische toetsen waarmee je de vereisten van traditionele hypothesetoetsen kunt omzeilen. Door de hele cursus heen verken je een Stack Overflow-gebruikersenquête en een gegevensset met vertraagde leveringen van medische hulpmiddelen.

Sampling in R

Leer over hypothesetests in R, inclusief het gebruik van t-tests, proportietests en chi-kwadraattests.

Veelvoorkomende aannames bij hypothesetoetsen

Hypothesis Testing in R

Praktische interactieve oefening