Ch. 3 : Numerical Summaries of Center and Variation Flashcards
What to use when measuring Center?
To measure the “typical value” use: mean, median, mode
Mean
Add all data and divide by number of observations
Median
the midpoint of ranked values (from small to large)
- if have even observations, take two in middle and take average (that is new median)
- resistant to outliers
Mean vs Median
Symmetric: mean=median
Skewed: center is the median
(right: median<median)
What to use when measuring Spread?
Measured by: Standard Deviation, IQR, and Range
Standard Deviation
described by the square root of the variance
(The avg distance of a value from the mean)
-it represents a typical distance from the mean of observations
–deviation: take each data point , subtract the mean, and then square that difference
Finding IQR
Q3-Q1= iqr (the middle 50% of the data)
Range
Max-MIn
-poor measure of spread
Which Center and Spread are best?
Use Mean & Standard Deviation when distribution symmetric/unimodal
Use Median & IQR when distribution is skewed L or R
Review for Center
Use Mean for typical value of symmetric
Median for typical value of skewed
Review for Spread
Use SD for symmetric
Use IQR for skewed
Empirical Rule
rough guideline that helps understand how SD measures variability (68% to 95% to 99.7%)