Experimental Design in R
In this course you'll learn about basic experimental design, a crucial part of any data analysis.
Start Course for Free4 Hours13 Videos53 Exercises
Create Your Free Account
or
By continuing, you accept our Terms of Use, our Privacy Policy and that your data is stored in the USA.Loved by learners at thousands of companies
Course Description
Experimental design is a crucial part of data analysis in any field, whether you work in business, health or tech. If you want to use data to answer a question, you need to design an experiment! In this course you will learn about basic experimental design, including block and factorial designs, and commonly used statistical tests, such as the t-tests and ANOVAs. You will use built-in R data and real world datasets including the CDC NHANES survey, SAT Scores from NY Public Schools, and Lending Club Loan Data. Following the course, you will be able to design and analyze your own experiments!
- 1
Introduction to Experimental Design
FreeAn introduction to key parts of experimental design plus some power and sample size calculations.
- 2
Basic Experiments
Explore the Lending Club dataset plus build and validate basic experiments, including an A/B test.
ANOVA, single, and multiple factor experiments50 xpExploratory Data Analysis (EDA) Lending Club100 xpHow does loan purpose affect amount funded?100 xpWhich loan purpose mean is different?100 xpMultiple Factor Experiments100 xpModel validation50 xpPre-modeling EDA100 xpPost-modeling validation plots + variance100 xpKruskal-Wallis rank sum test100 xpA/B testing50 xpWhich post-A/B test test?50 xpSample size for A/B test100 xpBasic A/B test100 xpA/B tests vs. multivariable experiments100 xp - 3
Randomized Complete and Balanced Incomplete Block Designs
Use the NHANES data to build a RCBD and BIBD experiment, including model validation and design tips to make sure the BIBD is valid.
Intro to NHANES and sampling50 xpNHANES dataset construction100 xpNHANES EDA100 xpNHANES Data Cleaning100 xpResampling NHANES data100 xpRandomized Complete Block Designs (RCBD)50 xpWhich is NOT a good blocking factor?50 xpDrawing RCBDs with Agricolae100 xpNHANES RCBD100 xpRCBD Model Validation100 xpBalanced Incomplete Block Designs (BIBD)50 xpIs a BIBD even possible?50 xpDrawing BIBDs with agricolae100 xpBIBD - cat's kidney function100 xpNHANES BIBD100 xp - 4
Latin Squares, Graeco-Latin Squares, and Factorial Experiments
Evaluate the NYC SAT scores data and deal with its missing values, then evaluate Latin Square, Graeco-Latin Square, and Factorial experiments.
Latin squares50 xpNYC SAT Scores EDA100 xpDealing with Missing Test Scores100 xpDrawing Latin Squares with agricolae100 xpLatin Square with NYC SAT Scores100 xpGraeco-Latin squares50 xpNYC SAT Scores Data Viz100 xpDrawing Graeco-Latin Squares with agricolae100 xpGraeco-Latin Square with NYC SAT Scores100 xpFactorial experiments50 xpNYC SAT Scores Factorial EDA100 xpFactorial Experiment with NYC SAT Scores100 xpEvaluating the NYC SAT Scores Factorial Model100 xpWhat's next in experimental design?50 xp
Datasets
sample of Lending Club dataNHANES Body MeasuresNHANES DemographicsNHANES final combined datasetNHANES Medical ConditionsNYC SAT ScoresCollaborators
Prerequisites
Hypothesis Testing in RJoanne Xiong
See More
Data Literacy & Essentials Content Developer
Joanne is a Data Literacy & Essentials Content Developer at DataCamp. She holds a Master’s degree in Statistical Science from the University of Oxford. Her passion is applying statistics and data science to a broad range of fields, ranging from clinical statistics to quantitative psychology.
What do other learners have to say?
Join over 12 million learners and start Experimental Design in R today!
Create Your Free Account
or
By continuing, you accept our Terms of Use, our Privacy Policy and that your data is stored in the USA.