Tenth Grade Mathematics: Introduction to Range, Mean Deviation, and Variance
While measures of central tendency—the mean, median, and mode—describe the central cluster of a dataset, they provide zero information about how widely the data is scattered. Two distinct classes of students might both have an identical average test score of 70 percent; yet in one class, every student scored between 68 and 72, while in the other class, scores ranged wildly from 20 to 100. To differentiate between these distributions, statisticians employ Measures of Dispersion. In this introductory chapter, we study the Range, Mean Deviation, Variance, and Standard Deviation.
The Need for Measures of Dispersion
Dispersion measures the degree of scatter, variation, or spread of individual observations around their central average:
Demonstration of Identical Central Tendencies with Different Dispersion:
Dataset A: 48, 49, 50, 51, 52 ===> Mean = 50, Median = 50 (Tightly clustered)
Dataset B: 10, 30, 50, 70, 90 ===> Mean = 50, Median = 50 (Widely dispersed)
Central tendency alone cannot distinguish Dataset A from Dataset B. Measures of dispersion quantify the extent to which Dataset B deviates from its central value.
Measure 1: The Range
The Range is the simplest measure of dispersion, defined as the difference between the maximum and minimum observations in the dataset:
Formula for Range:
Range = Maximum Value - Minimum Value
= X_max - X_min
While instantaneous to compute, the range suffers from severe limitations: it depends entirely on the two extreme outlier values and ignores all intermediate data.
Measure 2: The Mean Deviation (MD)
To account for all observations, we measure how far each observation xi lies from a central value (usually the mean x̄ or median M). Because positive and negative differences cancel out (Sum(xi - x̄) is always zero!), we take the absolute value of each deviation |xi - x̄|:
Theorem (Mean Deviation about the Mean):
For ungrouped data:
Sum( |xi - x̄| )
MD(Mean) = -----------------
N
For grouped data with frequencies fi:
Sum( fi * |xi - x̄| )
MD(Mean) = ---------------------
Sum(fi)
Mean Deviation about the Median:
Substitute the median M in place of x̄:
Sum( fi * |xi - M| )
MD(Median) = -------------------
Sum(fi)
Measure 3: Variance and Standard Deviation (SD)
While absolute values prevent algebraic cancellation, the absolute value function |x| is not smooth and possesses no derivative at zero. In advanced statistics and calculus, deviations are made positive by squaring them rather than taking absolute values.
Formal Definition of Variance (sigma^2):
Variance is defined as the arithmetic mean of the squares of the deviations
from the arithmetic mean:
Sum( (xi - x̄)^2 )
sigma^2 = --------------------- (Ungrouped Data)
N
Sum( fi * (xi - x̄)^2 )
sigma^2 = ----------------------- (Grouped Data)
Sum(fi)
Formal Definition of Standard Deviation (sigma):
The Standard Deviation is the positive square root of the variance:
+-----------------------
| Sum( fi * (xi - x̄)^2 )
sigma = sqrt( | ----------------------- )
\| Sum(fi)
Because squaring numbers squares the units (e.g., measuring height in cm yields variance in cm^2), taking the square root restores the original physical units, making the Standard Deviation the gold standard of statistical dispersion across all modern science.
Chapter Exercises
Exercise 1: Find the range, mean, and mean deviation about the mean for the following dataset: 6, 7, 10, 12, 13, 4, 8, 12.
Exercise 2: Calculate the variance and standard deviation for the discrete dataset: 2, 4, 6, 8, 10.
Exercise 3: Explain why the sum of the deviations of observations from their arithmetic mean without taking absolute values or squaring is always identically equal to zero.
Solutions and Explanations
Solution 1: The given dataset is: 4, 6, 7, 8, 10, 12, 12, 13 (arranged in ascending order). The number of observations is N = 8. Minimum value = 4, Maximum value = 13. Range = 13 - 4 = 9. Calculate the mean: Sum(xi) = 4 + 6 + 7 + 8 + 10 + 12 + 12 + 13 = 72. Mean x̄ = 72 / 8 = 9. Now compute absolute deviations |xi - x̄| from the mean 9: |4 - 9| = 5; |6 - 9| = 3; |7 - 9| = 2; |8 - 9| = 1; |10 - 9| = 1; |12 - 9| = 3; |12 - 9| = 3; |13 - 9| = 4. Sum of absolute deviations = 5 + 3 + 2 + 1 + 1 + 3 + 3 + 4 = 22. Mean Deviation about the mean: MD = Sum(|xi - x̄|) / N = 22 / 8 = 2.75.
Solution 2: The dataset is 2, 4, 6, 8, 10 with N = 5. First compute the mean: Sum(xi) = 2 + 4 + 6 + 8 + 10 = 30. Mean x̄ = 30 / 5 = 6. Now compute the squared deviations (xi - x̄)^2: (2 - 6)^2 = (-4)^2 = 16; (4 - 6)^2 = (-2)^2 = 4; (6 - 6)^2 = 0^2 = 0; (8 - 6)^2 = 2^2 = 4; (10 - 6)^2 = 4^2 = 16. Sum of squared deviations: Sum((xi - x̄)^2) = 16 + 4 + 0 + 4 + 16 = 40. The variance is: sigma^2 = Sum((xi - x̄)^2) / N = 40 / 5 = 8. The standard deviation is the positive square root of the variance: sigma = sqrt(8) = 2 * sqrt(2) is approximately 2.828. Thus, variance is 8 and standard deviation is approximately 2.83.
Solution 3: Let x1, x2, ..., xN be N observations with mean x̄ = (1/N) * Sum(xi). The sum of the raw deviations from the mean is: Sum(xi - x̄) = (x1 - x̄) + (x2 - x̄) + ... + (xN - x̄). Expanding this expression: = (x1 + x2 + ... + xN) - (x̄ + x̄ + ... + x̄). The first bracket is Sum(xi). In the second bracket, x̄ is added N times, which equals N * x̄. Thus: Sum(xi - x̄) = Sum(xi) - N * x̄. From the definition of the mean, multiplying both sides by N gives N * x̄ = Sum(xi). Substituting this into the deviation sum: Sum(xi - x̄) = Sum(xi) - Sum(xi) = 0. Because positive deviations from values above the mean exactly balance negative deviations from values below the mean, their algebraic sum is always identically zero.