2.7 EXPLORATORY DATA ANALYSIS - EDA

Exploring the data

       pg. 104 -  Comparison of EDA and Traditional Descriptive Statistics

 BOX PLOTS  - loss of detail, but very useful for visual comparisons

        ARRANGE Data in Order         (Find  fences to break into 4 = pieces)

             5 Scores  Minimum

                            Q1

                             Median

                             Q3

                             Maximum
 

Number of public high school graduates, in thousands, for 15 states 1980 & 1990
 
State 1980 1990
Fl 87.3 Min  88.8 Min 
PA 146.5 Q1  110.5 Q1 
MI 124.3 Median  93.8 Median 
OH 144.2 Q3  114.5 Q3 
AL 45.2 Max  40.5 Max 
WY 6.1 5.8
NJ 94.6 69.8
ME 15.4 13.8
LA 46.3 36.1
IL 135.6 108.1
WI 69.3 52.0
NY 204.1 143.3
AZ 38.6 32.1
TN 49.8 46.1
SD 10.7 7.7
 
 
 
 
 
 
 

 Histogram Vs Box Plot       bottom pg. 105

      Outliers    Discussion in Exercise 12
 

      Review  Pg. 110,  Cumulative Review pg. 113