Skip to content
Tutorials Center
My learning

Topics

Lesson 25 of 37 · Probability Distributions

The Normal Distribution

Meet the bell curve: mean, standard deviation, the 68-95-99.7 rule, and how areas under the curve give probabilities for continuous data.

Heights, measurement errors, test scores and the daily variation of many natural quantities cluster around a typical value and thin out symmetrically on both sides. The mathematical model for this is the normal distribution, with its famous bell-shaped curve.

Unlike the binomial, a normal variable is continuous: it can take any value in a range, so the probability of one exact value is zero. Probabilities are areas under the curve over an interval.

Two parameters

A normal distribution is fully determined by:

  • the mean μ, the centre of the curve, and
  • the standard deviation σ, which controls the width.

We write X ~ N(μ, σ²) (some courses write the variance σ², others the standard deviation; check which your syllabus uses). The curve is symmetric about μ, with mean, median and mode all equal. The total area under it is 1.

The density formula is:

Text
f(x) = 1 / (σ √(2π)) × e^( -(x - μ)² / (2σ²) )

You will rarely evaluate it by hand. What matters is the shape and the areas.

Python Plot normal curves with different mean and spread

Press Run (or Ctrl + Enter) to see the result here.

Program input · what you'd type while the program runs

Type the answers before you press Run, one per line. Each time the program reads a line (with input()), it gets the next line from this box.

A larger σ gives a lower, wider curve; changing μ slides the curve left or right without changing its shape.

The 68-95-99.7 rule

For any normal distribution:

Interval Approximate share of data
μ ± 1σ 68%
μ ± 2σ 95%
μ ± 3σ 99.7%

More precisely these are 68.27%, 95.45% and 99.73%.

Example

Adult heights in some population are roughly normal with mean 170 cm and standard deviation 8 cm. Then about 68% are between 162 and 178 cm, about 95% between 154 and 186 cm, and nearly everyone (99.7%) between 146 and 194 cm. A height of 190 cm is 2.5 standard deviations above the mean: unusual, but not impossible.

Computing areas with code

The area to the left of x for a normal curve is the cumulative distribution function (CDF). Python's standard library has the error function erf, which gives the CDF of the standard normal:

Text
Φ(z) = 0.5 × (1 + erf(z / √2))
Python Check the 68-95-99.7 rule

Press Run (or Ctrl + Enter) to see the result here.

Program input · what you'd type while the program runs

Type the answers before you press Run, one per line. Each time the program reads a line (with input()), it gets the next line from this box.

Checking whether data looks normal

A histogram that is roughly symmetric and bell-shaped suggests normality. Real data is never exactly normal, but many sets are close enough for the model to be useful. Samples from a skewed distribution (like incomes) are not.

Python Simulate data and compare with the rule

Press Run (or Ctrl + Enter) to see the result here.

Program input · what you'd type while the program runs

Type the answers before you press Run, one per line. Each time the program reads a line (with input()), it gets the next line from this box.

This data is simulated from a normal distribution, so it will fit the rule. Real measurements will match it only approximately.

Why the normal distribution is everywhere

Averages and sums of many small independent effects tend to look normal, whatever the shape of the individual effects. This is the central limit theorem, treated in sampling distributions. It is the reason the normal curve underlies so much of statistics.

Next, learn to turn any normal value into a standard score with z-scores.

Check your understanding

Quick check · 4 questions

of 4 answered You scored / 4 · perfect!

  1. 1. Scores are normal with mean 60 and standard deviation 10. About what share of scores lie between 50 and 70?

    50 to 70 is within one standard deviation of the mean, which is about 68%.
  2. 2. For a continuous normal variable, what is P(X = 60) exactly?

    A single point has no width, so no area. Probabilities come from intervals.
  3. 3. Increasing σ while keeping μ fixed makes the curve...

    More spread means the same total area is distributed over a wider range, so the peak is lower.
  4. 4. Which statement is true of every normal distribution?

    The curve is symmetric about μ, so mean, median and mode coincide.

Summary

  • A normal distribution N(μ, σ²) is symmetric and bell-shaped, set by its mean and spread.
  • Probabilities are areas under the curve; single points have probability 0.
  • About 68%, 95% and 99.7% of values fall within 1, 2 and 3 standard deviations of the mean.
  • The CDF can be computed with erf.

Next: Z-Scores and the Standard Normal Distribution