Gain an understanding of A/B tests and design. Learn about the aspects to be aware of to ensure appropriate handling of the data and analyses.

A/B testing and design

A/B subject design

Format and histograms

Considerations in A/B testing

Type I error in A/B tests

Sampling

Family-wise error rate

Power and sample size

Power impacts

Determine the ideal sample size

Regression to the mean

Introduction to A/B Tests

Learn common analyses to compare A/B groups. Understand the appropriate approach to each test given their assumptions and limitations.

Independent-sample t-test

Independent t-test sample size

Independent t-test analysis

Mann-Whitney U test

Mann-Whitney U analysis

Mann-Whitney U power assessment

t-test vs. Mann-Whitney U test

Chi-Square test of independence

Chi-Square test of independence sample size

Chi-Square test of independence analysis

Fisher's Exact test

Fisher's Exact test analysis

Choosing tests

Comparing Groups

Learn to analyze the trend and relationship of variables in A/B groups. Understand how to assess and present the results to any audience.

Introduction to correlations

Options

True

False

Correlation notes

Correlation coefficient

Pearson correlation

Pearson sample size

Assessing Pearson assumptions

Pearson ignoring groups

Pearson within groups

Spearman-rank correlation

Spearman correlation analysis

Pearson Correlation

Spearman Correlation

Correlation options

Reporting A/B test results

Choosing a data visualization

Tests

Bar or Box Plot

Scatter Plot

Plots for analyses

Creating bar plots for A/B design

Correlation and plotting

Associations of Variables

Understand the basis of regression and regression lines. Learn to run regressions, predict data based on the regression model, and visually present the results.

Introduction to linear regression

Scatter and points

Regression facts

Linear regression

Linear regression model

Linear predictions

Multiple linear regression

Logistic regression

Logistic regression model

Logistic probabilities

Plotting regression

Logistic plot

Linearity

Wrap-up

Regression and Prediction

absent

covid

webdata

gamedata

SiteSales

A/B testing is a common experimental design for human behavior research in industry and academia. In this course, you’ll learn what questions the A/B tests can address, the important considerations to be aware of in A/B tests, how to answer the questions at hand, and how to visualize the data. You’ll also learn how to determine the sample size needed in an experiment, conduct analyses appropriate for the data and hypothesis at hand, determine if the results can be regarded with confidence, and present the results to an audience regardless of statistical background.

A/B testing is a common experimental design for human behavior research in industry and academia. A/B tests compare two variants to determine if the measurement shows different performance and if measurements vary in a meaningful way. By learning about A/B testing and presenting the results, you can make data-driven decisions and predictions.
<br><br>
<h2>Build an Understanding of A/B Design</h2>
<br><br>
In this course, you’ll learn what questions the A/B tests can address, the important considerations to be aware of in A/B tests, how to answer the questions at hand, and how to visualize the data. You’ll also learn how to determine the sample size needed in an experiment, conduct analyses appropriate for the data and hypothesis at hand, determine if the results can be regarded with confidence, and present the results to an audience regardless of statistical background.
<br><br>
<h2>Learn How to Analyze A/B Test Data</h2>
<br><br>
This course covers parametric and non-parametric A/B tests, such as t-tests, Mann-Whitney U test, Chi-Square test of independence, Fisher’s exact test, and Pearson and Spearman correlations. Additionally, you’ll explore a power analysis for each test.
<br><br>
<h2>Predict Outcomes Based on Data</h2>
<br><br>
As you progress, you’ll also learn to run linear and logistic regressions to predict outcomes based on data and previous findings.
<br><br>
<h2>Present Results to Any Audience with Visualizations</h2>
<br><br>
By the time you complete this course, you’ll have a thorough understanding of A/B tests, the analyses you can perform with them, and how to relay the results with data visualizations.

Hypothesis Testing in R

Learn the basics of A/B testing in R, including how to design experiments, analyze data, predict outcomes, and present results through visualizations.

A/B Testing in R | Analyze, Predict & Visualize Results

A/B Testing in R

Stronger correlation coefficients allow more accurate predictions to be made

Correlations are particularly susceptible to outliers

A/B design correlations can ignore groups or assess within groups

Correlation assesses the strength and direction of two variables

Correlation notes

A/B Testing in R

exercicio interativo prático