SDA: Distributions/Skew Flashcards
What is normal distribution?
Symmetrical dispersion
Mode=Median=Mean
What is bimodal distribution?
Data with 2 Modes
NB: must draw histograms to know if data is bimodal as may be hidden in data tables
Positive skew
Mean>Median>Mode
Highest point furthest left
Negative skew
Mode>Median>Mean
Highest point furthest right
Inter-quartile range
Upper quartile minus lower quartile
i.e. Q3-Q1
What is the ‘fence’ calculation?
1.5xdQ
Any values that are 1.5 times dQ above or below Q1 and Q3 are outliers
What are the three ways to measure spread?
Variance
Standard Deviation
Coefficent of variation
What is the notation for the inter-quartile range?
dQ
Which measurement of spread is preferred? Variance or Standard Deviation?
Standard deviation because it’s measured in data units rather than data units squared as with variance
What is the advantage of the Coefficient of Variation measurement of spread?
Allows comparisons of spread for different variables