Introduction to the Normal Distribution

In the world of statistics, many things follow a predictable pattern. If you measure the heights of adults, the weights of apples, or the scores on an IB exam, you will notice that most data points cluster around the middle, with fewer and fewer values appearing as you move toward the extremes. This "bell-shaped" pattern is known as the Normal Distribution.

Understanding this distribution is like having a superpower for Paper 2. It allows us to calculate the probability of events happening and helps us compare different sets of data using a common "yardstick."

1. The Normal Distribution Curve (SL 4.9)

A continuous random variable \(X\) follows a normal distribution if its probability density function creates a specific bell-shaped curve. We write this as:

\(X \sim N(\mu, \sigma^2)\)

Where:
- \(\mu\) (mu) is the mean (the center of the curve).
- \(\sigma^2\) (sigma squared) is the variance.
- \(\sigma\) is the standard deviation (how spread out the data is).

Key Properties of the Normal Curve

Don't worry if the math looks complex; just remember these visual rules:

  • Symmetry: The curve is perfectly symmetrical around the mean, \(\mu\). This means 50% of the data is above the mean and 50% is below it.
  • Mean = Median = Mode: In a perfect normal distribution, these three measures of central tendency are all at the exact same peak.
  • Asymptotic: The "tails" of the curve get closer and closer to the horizontal axis but never actually touch it.
  • Total Area = 1: Since the area under the curve represents the total probability, it must equal \(1\).

The "Empirical Rule" (68-95-99.7 Rule)

Even without a calculator, you can estimate probabilities using these benchmarks:

- Approximately 68% of the data falls within \(1\) standard deviation of the mean: \((\mu \pm \sigma)\).
- Approximately 95% falls within \(2\) standard deviations: \((\mu \pm 2\sigma)\).
- Approximately 99.7% falls within \(3\) standard deviations: \((\mu \pm 3\sigma)\).

Quick Takeaway: If a value is more than 3 standard deviations away from the mean, it is very rare!

2. Calculations with the GDC (SL 4.9)

For IB Analysis and Approaches, you aren't expected to calculate normal probabilities by hand. You will use your Graphic Display Calculator (GDC) in Paper 2.

Finding Probabilities (Normal CDF)

When you want to find the probability that \(X\) falls between two values, \(P(a < X < b)\), you use the Normal Cumulative Distribution Function (Normal CDF).

Example: If IQ scores are \(X \sim N(100, 15^2)\), find the probability a person has an IQ between 85 and 115.
1. Go to the Distribution menu on your GDC.
2. Lower bound: \(85\)
3. Upper bound: \(115\)
4. \(\mu = 100, \sigma = 15\)
5. Result: \(\approx 0.683\)

Finding Values (Inverse Normal)

Sometimes you have the probability (the area) and you want to find the boundary value. This is the Inverse Normal function.

Example: "Find the score required to be in the top 10%."
- If you are in the top 10%, then 90% (0.90) of the data is below you.
- In your GDC, set the Area to \(0.90\) (always use the area to the left unless your calculator allows "Right" or "Center" tails), \(\mu\), and \(\sigma\).
- The calculator will give you the value \(k\) such that \(P(X < k) = 0.90\).

3. Standardization and Z-Values (HL 4.12 & SL 4.12)

Imagine trying to compare a student's score in a difficult Math test to their score in an easy History test. You can't just look at the raw percentages! To compare "apples to oranges," we standardize the data.

The Z-Score Formula

A z-score tells us how many standard deviations a value is away from the mean. The formula is:

\(z = \frac{x - \mu}{\sigma}\)

The Standard Normal Distribution

When we convert any normal distribution into z-scores, it becomes the Standard Normal Distribution, denoted by \(Z\):

\(Z \sim N(0, 1)\)

Here, the mean is always \(0\) and the standard deviation is always \(1\).

Why is this useful?
- If \(z = 1.5\), the value is 1.5 standard deviations above the mean.
- If \(z = -2\), the value is 2 standard deviations below the mean.

4. Finding Unknown Mean (\(\mu\)) or Standard Deviation (\(\sigma\))

A classic exam question will give you a probability but hide the mean or the standard deviation. To solve these, follow these steps:

Step 1: Use the Inverse Normal function on your GDC with the Standard Normal Distribution (\(\mu=0, \sigma=1\)) to find the z-score corresponding to the given probability.
Step 2: Plug that z-score, the given \(x\), and the known parameter into the standardization formula: \(z = \frac{x - \mu}{\sigma}\).
Step 3: Solve the equation for the missing variable.

Example: A variable \(X \sim N(\mu, 3^2)\) has \(P(X > 20) = 0.15\). Find \(\mu\).
1. If \(P(X > 20) = 0.15\), then the area to the left is \(0.85\).
2. Using GDC invNorm(0.85, 0, 1), we find \(z \approx 1.036\).
3. Substitute: \(1.036 = \frac{20 - \mu}{3}\).
4. \(3.108 = 20 - \mu \implies \mu = 16.892\).

Common Pitfalls to Avoid

  • Variance vs. Standard Deviation: In the notation \(N(\mu, \sigma^2)\), the second number is the variance. If you see \(N(10, 25)\), remember that \(\sigma = 5\). Always check if you need to square-root that number before putting it in your GDC!
  • Left vs. Right Tail: Most Inverse Normal calculations require the "Area" to be the probability to the left of the value. If a question says "the top 5%", use \(0.95\) as your area.
  • Rounding: Keep your z-scores to at least 4 decimal places during intermediate steps to ensure your final answer is accurate to 3 significant figures.

Key Takeaways

1. The Normal Distribution is symmetrical; half the data is above the mean, half is below.
2. Use Normal CDF for probabilities and Inverse Normal for values.
3. Standardizing with \(z = \frac{x - \mu}{\sigma}\) allows you to compare different data sets and find unknown parameters.
4. The Standard Normal Distribution \(Z\) always has \(\mu = 0\) and \(\sigma = 1\).