Kurtosis measures the "tailedness" of the distribution — that is, how much data are in the tails (extreme ends) compared to the center. It shows how peaked or flat a distribution is relative to a normal (bell-shaped) curve.

• High kurtosis (Leptokurtic): Tall peak and heavy tails → more extreme values (outliers).
• Medium kurtosis (Mesokurtic): Like a normal distribution.
• Low kurtosis (Platykurtic): Flat top and light tails → fewer outliers.

Simple Example:
Imagine two datasets:

• A group of students mostly scoring around 50–60 (values close to average) — this might show low kurtosis if few extreme scores exist.
• Another class where most scores are around the mean but some students have very high or very low scores — this shows high kurtosis because tails are heavier.

What it tells you:
Kurtosis helps understand whether a dataset has many outliers or data tightly clustered around the mean.