NHANES EDA

Let's examine our newly constructed dataset with a mind toward EDA. As in the last chapter, it's a good idea to look at both numerical summary measures and visualizations. These help with understanding data and are a good way to find data cleaning steps you may have missed. The nhanes_combined dataset has been pre-loaded for you.

Say we have access to NHANES patients and want to conduct a study on the effect of being told by a physician to reduce calories/fat in their diet on weight. This is our treatment; we're pretending that instead of this being a question asked of the patient, we randomly had physicians counsel some patients on their nutrition. However, we suspect that there may be a difference in weight based on the gender of the patient - a blocking factor!

This exercise is part of the course

Experimental Design in R

View Course

Exercise instructions

Fill in and run the dplyr code to find mean weight (bmxwt) in kg by our treatment (mcq365d). Is there anything interesting about the NA treated patients?
Fill in the ggplot2 code to look at a boxplot of the IQR of patients' weights by the treatment variable.

Hands-on interactive exercise

Have a go at this exercise by completing this sample code.

# Fill in the dplyr code
___ %>% 
  group_by(___) %>% 
  summarize(mean = mean(___, na.rm = TRUE))

# Fill in the ggplot2 code
___ %>% 
  ggplot(aes(as.factor(___), ___)) +
  geom_boxplot() +
  labs(x = "Treatment",
       y = "Weight")

Edit and Run Code