Chapter 3 Flashcards
What is an outlier?
An extreme value that lies outside the overall pattern of the data.
How do you calculate an outlier?
- Either greater than Q₃ + k (Q₃-Q₁)
- Or less than Q₁ - k(Q₃-Q₁) (k is often 1.5)
Sometimes outliers can be correct, so why remove them from the data?
Since it could be misleading to keep it in. We must remove anomalies from data
What does cleaning the data mean?
Removing anomalies from the data.
What do box plots show?
The important features of the data -
- Quartiles
- Max and min values (range)
- Outliers
How do you draw a cumulative frequency diagram?
x-axis = Class widths
y-axis = cumulative frequency
What type of data is represented in histograms?
Grouped continuous data.
What is the area of a bar in a histogram proportional to?
It is proportional to the frequency of each class.
How do you plot a histogram?
x-axis = class widths
y-axis = frequency density
Bars are together
How do you calculate frequency density?
Frequency density = frequency / class width
What do you compare data with?
- Measure of location (mean, medium, mode)
- Measure of spread (quartiles, standard deviation, variance, range)
