Sei sulla pagina 1di 2

CH 1 - Exploring Data 1.

1 - Displaying Distributions with Graphs Distribution o Center: mean, median, mode o Spread: range, quartiles o Shape: symmetry, skewness o Outliers?

Eva was very bored at work at Irvings TestMagic and decided to measure the weather in SF for two weeks at 2:00pm every day. She recorded the temperature (*F) below and then graphed it. 41 52 53 63 65 68 72 72 73 75 75 78 83 85

Qualitative/Categorical Variables Weather Very Cold Frequency (f) 1 1. Bar Graph


8 6 4 2 0 Very cold Cold Warm Hot Very hot

Cold 2

Warm 3 2. Pie Chart

Hot 6

Very Hot 2

Very cold Cold Warm Hot Very hot

Quantitative Variables Weather (*F) Frequency (f) Relative Freq. (rf) Cumulative Freq. (cf) Rel. Cum. Freq. (rcf) 1. Histogram
8 6 4 2 0 0 40 50 2 60

41-50 1 1/14=7.1% 1 1/14=7.1%

51-60 61-70 71-80 81-90 2 3 6 2 2/14=14.3% 3/14=21.4% 6/14=42.9% 2/14=14.3% 3 6 12 14 3/14=21.4% 6/14=42.9% 12/14=85.7% 14/14=100 2. Ogive
15 10 5 0

70 4 80

6 90 (*F) 4 5 6 7 8

0 41-50 51-60 2 61-70 71-80 4 81-90 (*F) 6

3. Dotplots

4. Stem-leaf plot
1 23 358 2223558 35 Stem = tens Leaf = ones

____________________________
41-50 51-60 61-70 71-80 81-90 (*F)

CH 1 - Exploring Data 1.2 - Describing Distributions with Numbers Central Tendency o Mean: AKA average Used best with quantitative data Affected by extreme outliers o Median: AKA midpoint Used best when there is an extreme outlier More resistant to extreme outliers o Mode: AKA most often Used best with categorical variables Spread o Range: Maximum value Minimum value o Quartiles: Q1 = Median of Minimum value to the overall Median Q2 = overall Median Q3 = Median of Maximum value to the overall Median Interquartile Range (IQR) = Q3 Q1 o Five-Number Summary: Minimum Q1 Median Q2 o Boxplot:

Maximum

Best for side by side comparison, if 2 box plots on 1 graph Less detailed than histograms or stemplots Modified boxplot: Draw regular boxplot with 5 Number Summary Add extreme outliers as points outside the whiskers Standard Deviation: average of the summed squares away from the mean Measures spread about the mean IF mean is chosen as measurement of center S=0 OR S>0; cannot be negative Can be skewed by extreme outliers

Potrebbero piacerti anche