The normal distribution

Welcome to the Normal Distribution!

Ever noticed how most people are of average height, and very few people are extremely tall or extremely short? Or how most apples in a bag weigh roughly the same, with only a few being tiny or massive? In Statistics, this "common" pattern is called the Normal Distribution. It is often nicknamed the "Bell Curve" because of its beautiful, symmetrical shape.

In this chapter, you’ll learn how to master this curve, use mathematical tables to find probabilities, and even use the Normal Distribution to make guesses about other types of data. Don't worry if it looks a bit "maths-heavy" at first—once you understand the symmetry, it’s like solving a satisfying puzzle!

1. What is the Normal Distribution?

The Normal Distribution is used for continuous random variables. Unlike discrete variables (like the number of students in a class), continuous variables can take any value (like the exact time it takes to run a race).

We describe a Normal Distribution using two main features (parameters):
1. The Mean (\( \mu \)): This is the center of the curve. It tells us where the peak is.
2. The Variance (\( \sigma^2 \)): This tells us how "spread out" the curve is. A small variance means a tall, skinny curve; a large variance means a short, fat curve.

We write this as: \( X \sim N(\mu, \sigma^2) \)

Important Property: Symmetry
The curve is perfectly symmetrical around the mean. This means:
- 50% of the data is above the mean.
- 50% of the data is below the mean.
- The total area under the curve is always 1 (representing a total probability of 100%).

Quick Takeaway: If you know one side of the curve, you know the other! If the probability of being "higher than X" is 0.1, the probability of being "lower than the equivalent point on the other side" is also 0.1.

2. The Standard Normal Distribution (\( Z \))

There are infinitely many Normal Distributions (different means, different variances). To make life easier, mathematicians created a "Universal Translator" called the Standard Normal Distribution, represented by the letter \( Z \).

For the \( Z \) distribution:
- The Mean (\( \mu \)) is always 0.
- The Variance (\( \sigma^2 \)) and Standard Deviation (\( \sigma \)) are always 1.
- We write this as: \( Z \sim N(0, 1) \).

The Standardization Formula:
To turn any value \( X \) into a \( Z \)-score, use this formula:
\( Z = \frac{X - \mu}{\sigma} \)

Analogy: Think of \( X \) as a currency (like Dollars or Euros) and \( Z \) as Gold. To compare them, we convert everything into Gold first!

Key Term: The \( Z \)-score tells you how many standard deviations a value is away from the mean.

3. Using the Normal Distribution Tables

In your exam, you are given a table that tells you the area (probability) to the left of a specific \( Z \)-value. This is often written as \( \Phi(z) \).

How to read the table:
1. Find the first two digits of your \( Z \)-score in the left column.
2. Find the third digit in the top row.
3. The number where they meet is your probability.

Symmetry Tricks to Remember:
- To find the area to the left of a positive \( z \): Just use \( \Phi(z) \).
- To find the area to the right of a positive \( z \): Use \( 1 - \Phi(z) \).
- To find the area to the left of a negative \( z \): Use \( 1 - \Phi(positive \ z) \).
- To find the area between two values \( a \) and \( b \): Find \( \Phi(b) - \Phi(a) \).

Common Mistake: Students often use the variance in the formula instead of the standard deviation. If the question says \( X \sim N(10, 16) \), remember that \( \sigma^2 = 16 \), so you must use \( \sigma = 4 \) in your calculation!

4. Solving Problems Step-by-Step

Most exam questions follow this flow:
Step 1: Write down what you know (\( \mu, \sigma, \) and the \( X \) value you are looking for).
Step 2: Standardize your \( X \) value using \( Z = \frac{X - \mu}{\sigma} \).
Step 3: Sketch a quick bell curve and shade the area you need. (This prevents silly mistakes!)
Step 4: Look up the \( Z \)-value in the table and apply symmetry rules if needed.

Reverse Problems:
Sometimes they give you the probability and ask for the value of \( X \).
1. Find the probability in the body of the table.
2. Find the corresponding \( Z \)-score.
3. Use the formula: \( X = \mu + Z\sigma \).

Don't worry if this seems tricky at first! Just remember: The table always gives the area to the left. If your shaded area is on the right, you'll need to do "1 minus" something.

5. Normal Approximation to the Binomial

Sometimes, using the Binomial Distribution (from the previous chapter) is too exhausting. If you are tossing a coin 1,000 times, you don't want to calculate 1,000 different probabilities! We can use the Normal Distribution as a shortcut.

When can you do this? (Syllabus Rules):
You can only use this approximation if:
1. \( np > 5 \)
2. \( nq > 5 \) (where \( q = 1 - p \))

Setting up the parameters:
- The mean is \( \mu = np \).
- The variance is \( \sigma^2 = npq \).

The Continuity Correction (The "Half-Unit" Rule):
The Binomial distribution is discrete (bars), but the Normal is continuous (a smooth line). To bridge the gap, we add or subtract 0.5.
- If you want \( P(X \ge 10) \), you actually look for \( P(X_{normal} > 9.5) \).
- If you want \( P(X \le 10) \), you actually look for \( P(X_{normal} < 10.5) \).
- If you want \( P(X = 10) \), you look for the area between 9.5 and 10.5.

Did you know? The continuity correction is like trying to wrap a round piece of paper over a square box—you need that extra 0.5 to make sure you cover the corners!

6. Summary & Quick Review

Key Takeaways:
- Notation: \( X \sim N(\mu, \sigma^2) \). Always check if you have variance or standard deviation!
- The Z-score: Measures how many "standard deviations" you are from the mean.
- Symmetry: The curve is identical on both sides. Total area = 1.
- Tables: They always show the area to the left of \( Z \).
- Continuity Correction: Only used when approximating Binomial data. Remember the +/- 0.5 rule.

Memory Aid: "Mean is the middle, Sigma is the spread. Standardize to Z, or you'll lose your head!"

Final Tip: Always sketch the curve! It takes 5 seconds and is the best way to ensure you are adding or subtracting from 1 correctly. You've got this!