Chapter 11

Statistics

Partition Values, Mean Deviation, Quartile Deviation, Standard Deviation, and Box Plots.

1. Mean Deviation (MD)

Mean deviation measures the average absolute difference between data points and a central value (either the Mean or the Median).

Measure Formula
MD from Mean (\(\bar{x}\)) \(MD_{\bar{x}} = \frac{\sum f|x - \bar{x}|}{N}\)
MD from Median (\(Md\)) \(MD_{Md} = \frac{\sum f|x - Md|}{N}\)
Coefficient of MD \(\frac{MD}{A}\) (where \(A\) is the Mean or Median used)

Example: MD from Mean

Data: 2, 4, 6 (\(N=3\))
Calculation: \(\bar{x} = \frac{2+4+6}{3} = 4\)

\(x\) \(|x - \bar{x}|\) i.e., \(|x - 4|\)
22
40
62
Total (\(N=3\)) \(\sum |x - \bar{x}| = 4\)

Result: \(MD_{\bar{x}} = \frac{4}{3} = \mathbf{1.33}\)
Coefficient: \(\frac{1.33}{4} = \mathbf{0.33}\)

Example: MD from Median

Data: 2, 4, 6 (\(N=3\))
Calculation: \(Md = 4\) (The middle value)

\(x\) \(|x - Md|\) i.e., \(|x - 4|\)
22
40
62
Total (\(N=3\)) \(\sum |x - Md| = 4\)

Result: \(MD_{Md} = \frac{4}{3} = \mathbf{1.33}\)
Coefficient: \(\frac{1.33}{4} = \mathbf{0.33}\)

2. Quartiles and Quartile Deviation (QD)

Quartiles divide data into four equal parts. Quartile Deviation measures the dispersion of the middle 50% of the data.

Quartile Position: \(Q_k = \text{Value of } \left[ \frac{k(N+1)}{4} \right]^{th} \text{ term}\)

Quartile Deviation: \(QD = \frac{Q_3 - Q_1}{2}\)

Coeff. of QD: \(\frac{Q_3 - Q_1}{Q_3 + Q_1}\)

Example: Calculating QD

Data: 10, 20, 30, 40, 50 (\(N=5\))

Step Calculation Result
Find \(Q_1\) Position = \(\frac{1(5+1)}{4} = 1.5^{th}\) term (\(\frac{10+20}{2}\)) 15
Find \(Q_3\) Position = \(\frac{3(5+1)}{4} = 4.5^{th}\) term (\(\frac{40+50}{2}\)) 45
Calculate QD \(QD = \frac{45 - 15}{2}\) 15
Coefficient \(\frac{45 - 15}{45 + 15} = \frac{30}{60}\) 0.5

3. Standard Deviation (\(\sigma\))

Standard deviation shows how much the data points deviate from the mean. It is the most reliable measure of dispersion.

Series Type Formula
Individual \(\sigma = \sqrt{\frac{\sum x^2}{N} - (\bar{x})^2}\)
Discrete \(\sigma = \sqrt{\frac{\sum fx^2}{N} - \left(\frac{\sum fx}{N}\right)^2}\)

Example: Discrete Series

Data: \(x = \{2, 4\}\), \(f = \{1, 3\}\) (\(N=4\))

\(x\) \(f\) \(fx\) \(fx^2\)
2 1 2 4
4 3 12 48
Total \(N = 4\) \(\sum fx = 14\) \(\sum fx^2 = 52\)

Step 1: Mean (\(\bar{x}\)) = \(\frac{14}{4} = 3.5\)

Step 2: \(\sigma = \sqrt{\frac{52}{4} - (3.5)^2} = \sqrt{13 - 12.25} = \sqrt{0.75} = \mathbf{0.866}\)

4. Box and Whisker Plot

This graphical representation is built using the 5-Number Summary derived from the partition values.

Feature Definition Data Example (10, 20, 30, 40, 50)
Minimum The lowest value in the dataset. 10 (Start of lower whisker)
\(Q_1\) The First Quartile (25th percentile). 15 (Start of the box)
Median (\(Md\)) The middle value (50th percentile). 30 (Line inside the box)
\(Q_3\) The Third Quartile (75th percentile). 45 (End of the box)
Maximum The highest value in the dataset. 50 (End of upper whisker)

Note: The "Box" represents the Interquartile Range (IQR), encompassing the middle 50% of the data from \(Q_1\) to \(Q_3\).