NIOS Economics • Module 4

Measures of Dispersion

Chapter Summary & Core Concepts

Mastering Statistical Dispersion & Variability

Learn why central tendency alone is inadequate to describe statistical data. Explore absolute and relative measures including Range, Quartile Deviation, Mean Deviation, Standard Deviation, Coefficient of Variation, and the Lorenz Curve according to the NIOS Economics curriculum.

Section 1

1. Concept & Need for Measures of Dispersion

Fundamental Concept

Core Concept Definition

Dispersion is the extent to which values in a distribution differ from the average of the distribution. Measures of central tendency (like Mean or Median) provide a single representative value, but they fail to reveal how individual items scatter around this central point.

Textbook Example: Why Average Alone Is Not Enough

Consider student marks across three different test groups (Group X, Group Y, Group Z):

Student Group X Marks Group Y Marks Group Z Marks
1 50 45 05
2 50 50 45
3 50 55 100
Mean (X̄) 50 50 50

Conclusion: All three groups have the exact same mean (50). However, Group X has zero dispersion, Group Y has low dispersion, and Group Z has extreme scatter (5 to 100). Dispersion quantifies this variation to test how representative an average truly is.

Classification: Absolute vs. Relative Measures

Absolute Measures
  • Expressed in original units of the data (e.g., kg, ₹, cm, hours).
  • Includes Range, Quartile Deviation (QD), Mean Deviation (MD), and Standard Deviation (SD).
  • Limitation: Cannot be used to compare two distributions with different measurement units.
Relative Measures (Coefficients)
  • Unitless ratios or percentages derived by dividing absolute measures by central values.
  • Includes Coefficient of Range, Coefficient of QD, Coefficient of MD, and Coefficient of Variation (C.V.).
  • Advantage: Perfect for comparative study of different datasets.
Section 2

2. Range & Quartile Deviation (QD)

A Range (R) & Coefficient of Range

Range is the simplest measure of dispersion, defined as the difference between the largest value (L) and the smallest value (S) in a distribution.

Absolute Range
Range (R) = L - S
Coefficient of Range
Coeff. of Range = (L - S) / (L + S)

Note for Continuous Series: Range is the difference between the upper limit of the highest class and lower limit of the lowest class. Cannot be calculated for open-ended distributions!

B Quartile Deviation (QD) / Semi-Interquartile Range

Quartile deviation is based on lower quartile (Q₁) and upper quartile (Q₃). The difference Q₃ - Q₁ is the Interquartile Range. Half of this difference is Quartile Deviation.

Quartile Deviation (QD)
QD = (Q₃ - Q₁) / 2
Coefficient of QD
Coeff. of QD = (Q₃ - Q₁) / (Q₃ + Q₁)
Key Evaluation: QD focuses exclusively on the middle 50% of observations. It completely ignores 25% of items at the bottom and 25% at the top. Consequently, QD is unaffected by extreme values and can be calculated for open-ended frequency distributions.
Section 3

3. Mean Deviation (MD)

Mean Deviation (MD): The arithmetic average of the absolute deviations of various observations taken from a measure of central tendency (Mean, Median, or Mode, though Median is generally preferred).

Formulas for Mean Deviation

Absolute Mean Deviation
MD = Σf|D| / N

Where |D| = |X - Central Value| (ignoring + and - signs).

Coefficient of Mean Deviation
Coeff. of MD = MD / Central Value

Central Value = Mean, Median, or Mode used.

Mathematical Limitation of Mean Deviation: Mean deviation ignores algebraic plus (+) and minus (-) signs when taking deviations (|D|). This step is mathematically unsound and non-algebraic, making further algebraic processing impossible.
Section 4

4. Standard Deviation (SD) & Coefficient of Variation (CV)

Standard Deviation (σ): The most important and widely used measure of dispersion. It is defined as the positive square root of the arithmetic mean of squared deviations taken from the arithmetic mean.

Methods of Computing Standard Deviation (σ)

1. Actual Mean Method
σ = √[ Σf(m - X̄)² / N ]

Best when mean is an integer value.

2. Assumed Mean Method
σ = √[ (Σfd² / N) - (Σfd / N)² ]

Where d = m - A (deviation from assumed mean).

3. Step Deviation Method
σ = √[ (Σfd′² / N) - (Σfd′ / N)² ] × c

Where d′ = (m - A) / c, c = class width.

4. Direct Method
σ = √[ (ΣX² / N) - (X̄)² ]

Useful for small raw observations.

Karl Pearson's Coefficient of Variation (C.V.)

Comparative Metric

Used to compare variability, consistency, or stability between two or more distributions with different units or magnitudes.

C.V. = ( σ / X̄ ) × 100

Interpretation Rule: A higher C.V. indicates greater variability / less consistency. A lower C.V. indicates less variability / greater consistency and stability.

Section 5

5. Graphical Dispersion: The Lorenz Curve

Lorenz Curve: A cumulative percentage curve used to graphically measure economic dispersion and inequality in income, wealth, or health distribution across a population.

Line of Equal Distribution (45° Diagonal)

Connects point (0,0) to (100,100). If income were distributed perfectly equally, the curve would lie directly on this 45° line.

The Inequality Gap

The area or distance between the 45° equal line and the plotted Lorenz Curve represents the inequality gap. The farther the curve lies from the 45° line, the greater the dispersion/inequality.