Quiz3

The data set is from a case-control study of smoking and Alzheimer’s disease. The data set has two variables of main interest:

• smoking a factor with four levels “None”, “<10”, “10-20”, and “>20” (cigarettes per day)
• disease a factor with three levels “Alzheimer”, “Other dementias”, and “Other diagnoses”.

library(tidyverse)

# Import data
data("alzheimer", package = "coin")

# create a table
tbl <- xtabs(~disease + smoking, alzheimer)
ftable(tbl)

# create a mosaic plot from the table
library(vcd)
mosaic(tbl, 
       shade = TRUE,
       legend = TRUE,
       labeling_args = list(set_varnames = c(disease = "")),
       set_labels = list(disease = c("Alzheimer", "Other\ndementias", "Other\ndiagnoses")))

Q1 Describe the largest group that has other dementias. Discuss it by number of cigarettes per day.

The largest group that has dementia are people who smoke 0 cigarettes each day.

Q2 Describe one group that has more cases than expected given independence (by chance). Discuss it by number of cigarettes per day.

People with dementia that smoke more than 20 cigarettes a day have more cases than expected.

Q3 Does smoking seem to matter in determining other dementias? Discuss your reason using the masaic chart above.

Yes, because smoking is known to increase the risks of having dementia. Using the chart, people are more likely to die of dementia if they smoke more than 20 cigarettes a day.

Q4 Create correlation plot for RailTrail.

Hint: The RailTrail data set is from the mosaicData package.

# import data
data(RailTrail, package="mosaicData")

# select numeric variables
df <- dplyr::select_if(RailTrail, is.numeric)

# calulate the correlations
r <- cor(df, use="complete.obs")
round(r,2)

library(ggplot2)
library(ggcorrplot)

# visualize the correlations
ggcorrplot(r, 
           hc.order = TRUE, 
           type = "lower",
           lab = TRUE)

Q5 List all four variables that have negative correlation with the number of trail users (volume).

The variables that have negative correlation are -0.04, -0.25, -0.37, and -0.23.

Q6 What season seems to be least popular for trail users?

Fall seems to be the least popular for trail users.

Q7 The correlation coefficient between hightemp and cloudcover is quite small. Would you be sure that the two variables are not related at all?

Hint: One word answer (e.g., yes or no) is NOT enough. Explain why.

Although you can’t be sure that there is no relationship between the two, we have to assume that there is a relationship since the ceofficient is above 0. There is a possibility that there is no reltionship because of how close to 0 the coefficient is, but there is no way to be 100% sure.

Q7.a Continued from Q7. Create scatter plot. After examing the scatter plot, would you conclude that the two variables are not related at all?

Hint: Discuss your reason by explaining your scatter plot.

library(ggplot2)
# Basic scatter plot
ggplot(RailTrail, aes(x=hightemp, y=cloudcover)) + geom_point()

I would conclude that they are not related based on the scatterplot, I say this because there is no pattern between the two which makes sense since the coefficient is that close to 0.

Q8 Hide the messages, the code and its results on the webpage.

Hint: Use message, echo and results in the chunk options. Refer to the RMarkdown Reference Guide.

Q9 Display the title and your name correctly at the top of the webpage.

Q10 Use the correct slug.