
Tutorial description
Ultimately, data analysis is about understanding relationships among variables. Exploring data with multiple variables requires new, more complex tools, but enables a richer set of comparisons. In this tutorial, you will learn how to describe relationships between two numerical quantities. You will characterize these relationships graphically, in the form of summary statistics, and through simple linear regression models.
In this tutorial you’ll also take your skills with simple linear regression to the next level. By learning multiple and logistic regression techniques you will gain the skills to model and predict both numeric and categorical outcomes using multiple input variables. You’ll also learn how to fit, visualize, and interpret these models. Then you’ll apply your skills to learn about Italian restaurants in New York City!
Learning objectives
- Visualize, measure, and characterize bivariate relationships
- Fit, interpret, and assess simple linear regression models
- Measure and interpret model fit
- Identify and attend to the disparate impact that unusual data observations may have on a regression model
- Compute with
lm
objects in R
- Compute and visualize model predictions
- Visualize, fit, interpret, and assess a variety of multiple regression models, including those with interaction terms
- Visualize, fit, interpret, and assess logistic regression models
- Understand the relationship between R modeling syntax and geometric and mathematical specifications for models
Lessons
- Quantify the strength of a linear relationship
- Compute and interpret Pearson Product-Moment correlation
- Identify spurious correlations
- Interpret the meaning of coefficients in a regression model
- Understand the impact of units and scales
- Work with
lm
objects in R
- Make predictions from regression models
- Overlay a regression model on a scatterplot
- Assess the quality of fit of a regression model
- Interpret \(R^2\)
- Measure leverage and influence
- Identify and attend to outliers
- Visualize, fit, and interpret a parallel slopes model, which has one numeric and one categorical explanatory variable
- Describe a model in three different ways: mathematically, graphically, and through R syntax
- Assess and interpret model fit
- Compute residuals and predictions
- Fit and interpret interaction models
- Recognize Simpson’s paradox
- Visualize, fit, and interpret a multiple regression model with two numeric explanatory variables
- Visualize, fit, and interpret a parallel planes model with two numeric explanatory variables and a categorical variable
- Fit and interpret multiple regression models in higher dimensions
- Understand and identify multicollinearity
- Visualize, fit, and interpret logistic regression models
- Interpret coefficients on three different scales
- Make predictions from a logistic regression model
- Explore the relationship between price and the quality of food, service, and decor for Italian restaurants in New York City
Instructor

Benjamin S. Baumer
Smith College
Benjamin S. Baumer is an associate professor in the Statistical & Data Sciences program at Smith College. He has been a practicing data scientist since 2004, when he became the first full-time statistical analyst for the New York Mets. Ben is a co-author of The Sabermetric Revolution, Modern Data Science with R, and the second edition of Analyzing Baseball Data with R. He received his Ph.D. in Mathematics from the City University of New York in 2012, and is accredited as a professional statistician by the American Statistical Association. His research interests include sports analytics, data science, statistics and data science education, statistical computing, and network science.
Ben won the Waller Education Award from the ASA Section on Statistics and Data Science Education, and the Significant Contributor Award from the ASA Section on Statistics in Sports in 2019. He shared the 2016 Contemporary Baseball Analysis Award from the Society for American Baseball Research. Currently, Ben is the primary investigator on a three-year, nine-institution, $1.2 million award from the National Science Foundation for workforce development under the Data Science Corps program.