1.1.4

Question 1

What are explanatory variables?

Answer 1

Explanatory variable = independent variable = predictor variable = X

Question 2

What are outcome variables?

Answer 2

Outcome variable = dependent variable = response variable = X

Question 3

What is the difference between the manipulation of variables in experimental and observational studies?

Answer 3

Experimental = explanatory variable is manipulated before response variable is measured
Observational = variables are observed as they naturally exist (not controlled)

Question 4

What does the group_by() function do?

Answer 4

Creates a grouping in the dataframe
Subsequent functions will be computed on each group

Question 5

Why do we combine the group_by() and summarise() functions?

Answer 5

To reduce a variable into a summary value for each group in a grouping variable

Question 6

How can we input group_by() and summarise() into R?

Answer 6

data %>%
group_by(grouping variable) %>%
summarise(
summary_value = …
)

Question 7

How do we map something on a plot e.g. the colour, to something in the data e.g. a variable?

Answer 7

ggplot(data = dataframe, aes(x = variable 1, col = group in variable)) +
geom_density()

Question 8

How can we split one plot up using facet_wrap to create separate graphs for each set of values/groups in a variable?

Answer 8

ggplot(data = dataframe, aes(x = variable)) +
geom_histogram() +
face_wrap(~variable/group)

Question 9

What is the most easily interpreted visualisation of the relationship between two numeric variables?

Answer 9

Scatterplot

Question 10

What is covariance, and what can it express?

Answer 10

A measure of how two numeric variables vary together
Can express the directional relationship

Question 11

What are two ways of using the cov() function to calculate covariance?

Answer 11

Use $ to pull out variables from the datset e.g.
cov(dataframe$variableX, dataframe$variableY)
or
Specify dataframe and use %>% + call cov() inside
data %>%
summarise(
mean_variableX = mean(variableX)
mean_variableY = mean(vaariableY),
cov_variableXY = cov(variableX, variableY)
)

Question 12

How can we summarise one categorical variable using the table function?

Answer 12

table(dataframe$categorical variable)
or
data %>%
count(categorical variable)

Question 13

What is a two-way table?

Answer 13

A table with each variable on either dimension

Question 14

How can we create a two-way table for two categorical variables in R?

Answer 14

table(dataframe$variable1, dataframe$variable2)

Question 15

How can we create a proportion table for two categorial variables in R?

Answer 15

dataframe %>%
select(variableX, variableY) %>%
table() %>%
prop.table() for total proportions
prop.table(margin = 1) for proportions of each row
prop.table(margin = 2) for proportions of each column

Question 16

How can we make a mosaic plot to visualise a contingency table of two categorical variables in R?

Answer 16

dataframe %>%
select(variable X, variable Y) %>%
table() %>%
or prop.table(margin = 1) %>%
or prop.table(margin = 2) %>%
plot()