Mean Deviation, Variance and Standard Deviation
Lesson 11 of STA301 examines the most powerful and mathematically rigorous measures of dispersion: Mean Deviation, Variance, and Standard Deviation. Mean Deviation (MD), or average deviation, represents the arithmetic average of absolute deviations of observations from a central measure (either the mean or the median): MD = sum|X - X-bar| / n. Because deviations from the mean sum algebraically to zero, absolute value signs are introduced to prevent positive and negative deviations from cancelling out. Its relative measure is the Coefficient of Mean Deviation, computed as MD divided by the average employed. To overcome the algebraic limitations of absolute values, modern mathematical statistics relies on squaring deviations. The Variance represents the arithmetic mean of squared deviations from the arithmetic mean. For a population, it is denoted by sigma^2 = sum(X - mu)^2 / N, and for a sample, s^2 = sum(X - X-bar)^2 / (n - 1) when applying Bessel's correction for unbiased estimation. For computational efficiency, the short-cut formula s^2 = [sum(X^2) - (sum X)^2 / n] / (n - 1) is derived for ungrouped data, and s^2 = [sum(f*X^2) - (sum f*X)^2 / n] / (n - 1) for grouped data. The Standard Deviation represents the positive square root of variance: s = sqrt(s^2). Expressed in the original units of measurement, standard deviation serves as the foundational parameter in probability modeling, inferential estimation, and hypothesis testing. Important properties are demonstrated: standard deviation is invariant under changes of origin (adding or subtracting a constant does not alter s), but changes proportionally under changes of scale (multiplying by a constant scales s by |c|). Finally, Karl Pearson's Coefficient of Variation (CV = (s / X-bar) * 100%) is introduced as the standard relative measure for comparing variability, stability, and consistency between competing distributions. A smaller CV indicates greater uniformity and consistency. In Virtual University midterm examinations, variance formulas, standard deviation properties, and CV comparisons appear regularly in both objective and subjective portions. Mastering these concepts through sta301 handouts and solved past papers guarantees success.
No questions available yet.
There are currently no examination questions published for Mean Deviation, Variance and Standard Deviation. Explore other curriculum units or build a custom quiz with the practice builder.