Hang out with the TAs from STA 199! This is a casual conversation and a fun opportunity to meet the members of the STA 199 teaching team. The only rule is these can’t turn into office hours!
Tea with a TA counts as a statistics experience.
We will be looking at the Paris Paintings data set in today’s application exercise. We’ll primarily focus on the variables:
Height_in: Height (in inches)Width_in: Width (in inches)landsAll: If any type of landscape is mentioned (either lands_sc, lands_figs, or lands_ment)Click here for the complete codebook.
paintings <- read_csv("data/paris_paintings.csv")Create a scatterplot to visualize the relationship between Height and Width. Color the points based on the whether the painting has any landscape elements. Note: Be sure to make landsAll a factor, so R treats it as categorical variable.
Based on your scatterplot, does the relationship between Height and Width differ between paintings with landscape elements and those without? Briefly explain.
Fit a main effects model using width and whether the painting has landscape elements to predict the height.
Using your model from the previous exercise,
Width_in.landsAll.Write the equation of the model for
How do the slopes compare between the two models? How do the intercepts compare?
Now let’s consider an interaction term. Fit a linear model using width, whether the painting has landscape elements, and the interaction between the two variables to predict the height.
Write the equation of the model for
How do the slopes compare between the two models? How do the intercepts compare?
Now let’s see how well each model fits the data. Use the glance function to calculate \(R^2\) and Adj. \(R^2\) for the model fit in Ex. 3 and the one fit in Ex. 6.
Based on your answer to the previous exercise, which model is a better fit for the data? Briefly explain.
Interpret \(R^2\) for the model chosen in the previous exercise.