Skewness: Definition and Interpretation
What you'll learn: You'll discover what skewness tells you about the asymmetry of your data's distribution.
What is Skewness?
Skewness measures asymmetry in a distribution—how lopsided your data looks when you visualize it. While measures like mean, median, and mode tell you where data centers, skewness reveals whether values bunch up on one side and stretch out toward the other.
Think of it like a tail on a hill: if one side has a long, gentle slope and the other drops off sharply, your distribution is skewed.
Types of Skewness
Positive (Right) Skewness:
The tail extends toward higher values (to the right). Most data clusters on the left, with a few unusually large values pulling the mean upward. Example: household income—most people earn moderate amounts, but a few billionaires create a long right tail.
In positively skewed data: mean > median > mode
Negative (Left) Skewness:
The tail extends toward lower values (to the left). Data clusters on the right, with a few unusually small values dragging the mean down. Example: test scores where most students score high, but a few very low scores stretch leftward.
In negatively skewed data: mean < median < mode
Symmetrical (No Skew):
Both sides mirror each other. The mean, median, and mode align near the center—like the normal distribution you learned about when understanding distribution shape.
Why It Matters
Skewness affects which central tendency measure best represents your data. In highly skewed distributions, the median often tells a more reliable story than the mean because it resists the pull of extreme values in the tail.
Key Takeaway: Skewness quantifies asymmetry in distributions—positive skew means a long right tail (mean pulled upward), negative skew means a long left tail (mean pulled downward), helping you choose appropriate summary statistics.