← 📈 Descriptive Statistics & Sampling
Subtopic 2 of 3

6.2Variance and standard deviation of data

The idea

The variance and standard deviation of data measure its spread. Shifting all values changes the mean but not the spread; scaling changes both.

Think of it like this: Moving every student's marks up by 7 changes the average, not how spread out the class is. Doubling all marks doubles the spread too.

Why it exists

GATE asks which of these effects are true after "adding 7" or "doubling each entry".

How it works

  1. Variance = average squared deviation from the mean.
  2. Sample variance divides by n − 1 instead of n.
  3. Adding b: mean + b; variance and SD unchanged.
  4. Multiplying by a: mean × a, SD × |a|, variance × a².
  5. Standard deviation has the same unit as the data; variance has the unit squared.

Worked example: GATE 2012: add 7, or double each entry

Adding 7 raises the mean by 7 but leaves the standard deviation alone. Doubling doubles both. So "mean + 7" and "doubled mean" are true, while "SD + 7" and "SD unchanged by doubling" are false.

Now play with it

  • →Set a = 1 and b = 7: the variance row does not change.
  • →Set a = 2 and b = 0: the SD doubles and the variance quadruples.
listmeanvariancestd dev
X = 2, 4, 4, 4, 5, 5, 7, 9542
Y = 2X + 717164

E[aX + b] = a·E[X] + b, but Var(aX + b) = a²·Var(X) and SD(aX + b) = |a|·SD(X). Adding b moves everything and changes the mean, not the spread. Doubling doubles the mean and the standard deviation.

Check yourself

The SD of a list is 5. Every entry is multiplied by −3 and then 10 is added. The new SD is:

Remember: Shift: mean moves, SD does not. Scale by a: SD × |a|, variance × a².
Exam notes (the dense version)

For data x₁..xₙ: variance = (1/n) Σ(xᵢ − x̄)² (population) or divided by n − 1 for a sample estimate. Effects on a list: adding b adds b to the mean and leaves variance and SD unchanged; multiplying by a multiplies the mean by a, the SD by |a| and the variance by a².

Previous year questions: Descriptive Statistics & Sampling (2)

PYQs are grouped by topic, so this is the full set for Descriptive Statistics & Sampling; try the ones that match this subtopic. Open all 2 on the topic page.