2. Data Visualisations

Question 1

How do you draw a boxplot?

Answer 1

• vertical axis represents data values
• Draw a box from lower quartile to median
• draw box from median to upper quartile
• decide which points are circles and plot with cicles (LQ - 1.5IQR and UP + 1.5IQR)
• draw whiskers to the minimum and maximum non-outlier datapoints

Question 2

How is data organised in histograms?

Answer 2

In bins, intervals of data

Question 3

What is the frequency f of bin on a histogram?

Answer 3

the number of observations in that bin

Question 4

what is the relative frequency of a bin on a histogram?

Answer 4

f/n, the proportion of observations of the observations in that bin

Question 5

what is the frequency density of bins on a histogram?

Answer 5

the relative frequency divided by the size of the bin

Question 6

what is the area of a bar on a histogram?

Answer 6

relative frequency

Question 7

can you work out mean and variance of data on a histogram?

Answer 7

no as it not exact, but you can approximate mean and variance

Question 8

how do you approximate the mean on a histogram?

Answer 8

x(hat) = 1/n sum(frequency * midpoint, number of bins)

Question 9

how do you approximate the variance on a histogram?

Answer 9

Var = 1/n-1 sum(frequency + (midpoint - mean)^2 , number of bins)

Question 10

when would you use a scatterplot?

Answer 10

when more than one piece of data is collected from each subject, and you want to compare the data to find a relationship between variables.

Question 11

how would you work out correlation between variables?

Answer 11

Sxy=SxSy
where Sxy is the sample covariance