← Back to FractionRush

Normal Distribution S1 Statistics

Grade 11 · Statistics 1 · Cambridge A-Level 9709 · Age 16–17

Welcome to Normal Distribution!

The normal distribution is the most important continuous probability distribution in statistics. It describes how many real-world quantities — heights, weights, exam scores, measurement errors — are naturally distributed. In Cambridge A-Level Statistics 1 (9709), you learn to use the standard normal distribution, standardise values, find probabilities from tables, work backwards using inverse normal, and approximate binomial distributions. This is essential for all further statistics work.

X~N(μ,σ²)  |  Z = (X−μ)/σ where Z~N(0,1)  |  P(X<a) = Φ(z)  |  Φ(−z) = 1 − Φ(z)

Learning Objectives

  • Understand the shape and properties of the normal distribution N(μ,σ²)
  • Recognise that the normal curve is symmetric, bell-shaped and the total area = 1
  • Apply the 68–95–99.7 rule to estimate probabilities
  • Standardise X to Z = (X−μ)/σ where Z~N(0,1)
  • Find probabilities P(X<a), P(X>b), P(a<X<b) using tables
  • Use the symmetry rule Φ(−z) = 1 − Φ(z)
  • Apply inverse normal to find x given a probability
  • Find unknown μ or σ by setting up equations from given probabilities
  • Apply the normal approximation to binomial B(n,p) when np>5 and nq>5
  • Use continuity corrections correctly when approximating discrete with continuous

Bell Curve

X~N(μ,σ²): symmetric, mean=median=mode=μ, total area = 1

68-95-99.7 Rule

68% within 1σ, 95% within 2σ, 99.7% within 3σ of mean

Standardising

Z = (X−μ)/σ transforms any N(μ,σ²) to N(0,1)

Probabilities

P(X<a) = Φ(z); use symmetry Φ(−z) = 1−Φ(z)

Inverse Normal

Given p, find x: z = Φ²(p), then x = μ + zσ

Unknown μ, σ

Set up simultaneous equations from two probability conditions

Normal Approx

B(n,p) ≈ N(np, npq) when np > 5 and nq > 5

Continuity Correction

P(X=k) becomes P(k−0.5<Y<k+0.5) for approximation

Learn 1 — The Normal Distribution

What is the Normal Distribution?

The normal distribution is a continuous probability distribution described by a symmetric bell-shaped curve. It is completely determined by two parameters: the mean μ and the variance σ². We write X~N(μ,σ²).

Notation alert: We write N(μ, σ²) — the second parameter is the variance, not the standard deviation. This is one of the most common mistakes in exams. For example, X~N(50, 16) means μ=50 and σ²=16, so σ=4.

Key Properties of the Normal Distribution

Bell-shaped and symmetric about the mean μ
Mean = Median = Mode = μ (all three measures of centre are equal)
• The curve extends from −∞ to +∞ but the tails approach zero quickly
Total area under the curve = 1 (since it is a probability distribution)
• The spread is controlled by σ: larger σ means a wider, flatter curve
• The curve has points of inflection at x = μ ± σ

The 68–95–99.7 Rule (Empirical Rule)

For any X~N(μ,σ²), the following approximate probabilities hold:

P(μ−σ < X < μ+σ) ≈ 0.6827   (about 68%)
P(μ−2σ < X < μ+2σ) ≈ 0.9545   (about 95%)
P(μ−3σ < X < μ+3σ) ≈ 0.9973   (about 99.7%)
Example: X~N(100, 225) so μ=100, σ=15.
• P(85 < X < 115) ≈ 68% (within 1 standard deviation)
• P(70 < X < 130) ≈ 95% (within 2 standard deviations)
• P(55 < X < 145) ≈ 99.7% (within 3 standard deviations)

Continuous Distribution

Because the normal distribution is continuous, the probability of X taking any exact value is zero: P(X = a) = 0. We can only find probabilities over intervals:

P(a < X < b) = P(a ≤ X ≤ b) — inequalities < and ≤ are equivalent for continuous distributions

Why N(μ, σ²) matters

Many real-world quantities are approximately normally distributed:
• Heights of adults in a population
• Scores on a well-designed standardised test
• Measurement errors in scientific experiments
• Diameters of manufactured components (quality control)
• Daily returns on financial assets (approximately)

Shape of the Curve — Sketch Guide

When sketching a normal curve in an exam:
1. Draw a smooth bell curve symmetric about μ
2. Label the x-axis with μ at the centre, μ−σ and μ+σ either side
3. The peak of the curve is directly above μ
4. Shade the region you are finding the probability of
5. Always write the area = probability you are calculating
Comparing two distributions:
X~N(50, 9) has μ=50, σ=3. Curve centred at 50, narrow.
Y~N(50, 25) has μ=50, σ=5. Curve centred at 50, wider and flatter.
Both are symmetric about x=50 but Y has more spread.

Learn 2 — Standardising to Z~N(0,1)

The Standard Normal Distribution

The standard normal distribution is the special case with mean 0 and variance 1: Z~N(0,1). Tables of probabilities are published for Z~N(0,1), so we convert any normal distribution to this standard form before looking up probabilities.

The Standardisation Formula

Z = (X − μ) / σ    where Z~N(0,1)

This transforms a value x from X~N(μ,σ²) into a z-score: how many standard deviations x is above or below the mean. A positive z means x is above the mean; a negative z means x is below the mean.

Step-by-Step Standardisation

Example: X~N(70, 64). Find the z-score for x = 82.
(Note: σ² = 64, so σ = 8)

Step 1: Identify μ = 70 and σ = 8 (take square root of variance!)
Step 2: Apply Z = (X − μ)/σ = (82 − 70)/8 = 12/8 = 1.5
Interpretation: x = 82 is 1.5 standard deviations above the mean.
Example: X~N(50, 100). Find the z-score for x = 35.
σ = √100 = 10
Z = (35 − 50)/10 = −15/10 = −1.5
x = 35 is 1.5 standard deviations below the mean.

Back-Transforming: X from Z

If you know the z-score and want to find the original x-value, rearrange the formula:

X = μ + Zσ
Example: X~N(40, 25). Z = 1.8. Find x.
σ = √25 = 5
x = 40 + 1.8 × 5 = 40 + 9 = 49
Example: X~N(100, 16). Z = −2.1. Find x.
σ = 4
x = 100 + (−2.1)(4) = 100 − 8.4 = 91.6

Why Standardise?

Normal distribution tables are printed for Z~N(0,1) only. There are infinitely many possible (μ, σ²) combinations, so we cannot tabulate them all. Instead:
1. Standardise your value to a z-score using Z = (X−μ)/σ
2. Look up Φ(z) = P(Z < z) in the standard normal table
3. Convert back to a statement about X if needed

Properties of Z~N(0,1)

P(Z < 0) = 0.5  |  P(Z < z) = Φ(z)  |  P(Z > z) = 1 − Φ(z)
Φ(−z) = 1 − Φ(z)   (symmetry rule)

Finding σ from the Standardisation Formula

Example: X~N(60, σ²). If x = 66 corresponds to z = 1.5, find σ.
Z = (X−μ)/σ ⇒ 1.5 = (66−60)/σ ⇒ σ = 6/1.5 = 4
Always extract σ (standard deviation) from σ² (variance) before standardising. Plugging in σ² instead of σ is one of the most common exam errors!

Learn 3 — Finding Probabilities

The Cumulative Distribution Function Φ(z)

For Z~N(0,1), Φ(z) = P(Z < z). The normal distribution tables give values of Φ(z) for z ≥ 0. We combine these with the symmetry rule to find all other probabilities.

Case 1: P(X < a)

P(X < a) = P(Z < z) = Φ(z)   where z = (a−μ)/σ
Example: X~N(50, 16), find P(X < 56).
[Diagram: bell curve, shade left of 56, label μ=50]
z = (56−50)/4 = 6/4 = 1.5
P(X < 56) = Φ(1.5) = 0.9332

Case 2: P(X > a)

P(X > a) = 1 − P(X < a) = 1 − Φ(z)
Example: X~N(50, 16), find P(X > 56).
[Diagram: bell curve, shade right of 56]
z = (56−50)/4 = 1.5
P(X > 56) = 1 − Φ(1.5) = 1 − 0.9332 = 0.0668

Case 3: P(X < a) when a < μ (negative z)

P(X < a) = Φ(z) where z < 0. Use symmetry: Φ(−z) = 1 − Φ(z)
Example: X~N(50, 16), find P(X < 44).
[Diagram: bell curve, shade left of 44, note 44 is to left of μ=50]
z = (44−50)/4 = −1.5
P(X < 44) = Φ(−1.5) = 1 − Φ(1.5) = 1 − 0.9332 = 0.0668
Note: by symmetry, P(X < 44) = P(X > 56) since 44 and 56 are equidistant from μ=50.

Case 4: P(a < X < b)

P(a < X < b) = Φ(z⊂2;) − Φ(z⊂1;)   where z⊂1; = (a−μ)/σ, z⊂2; = (b−μ)/σ
Example: X~N(50, 16), find P(44 < X < 58).
[Diagram: bell curve, shade between 44 and 58]
z⊂1; = (44−50)/4 = −1.5 ⇒ Φ(−1.5) = 1 − 0.9332 = 0.0668
z⊂2; = (58−50)/4 = 2.0 ⇒ Φ(2.0) = 0.9772
P(44 < X < 58) = 0.9772 − 0.0668 = 0.9104

The Symmetry Rule — Essential!

Φ(−z) = 1 − Φ(z)   for any z > 0
Tables only give Φ(z) for z ≥ 0. For negative z values, always use the symmetry rule.

More examples:
Φ(−0.8) = 1 − Φ(0.8) = 1 − 0.7881 = 0.2119
Φ(−2.33) = 1 − Φ(2.33) = 1 − 0.9901 = 0.0099
P(Z > −1.2) = 1 − Φ(−1.2) = 1 − (1−Φ(1.2)) = Φ(1.2) = 0.8849

Summary Table of Cases

Probability requiredz-score(s)Formula
P(X < a), z > 0z = (a−μ)/σΦ(z)
P(X < a), z < 0z = (a−μ)/σ1 − Φ(|z|)
P(X > a)z = (a−μ)/σ1 − Φ(z) or Φ(−z)
P(a < X < b)z⊂1;, z⊂2;Φ(z⊂2;) − Φ(z⊂1;)
Always sketch a diagram! Drawing the bell curve, marking μ, marking the boundary value(s), and shading the required region before calculating massively reduces errors. It helps you see whether to add or subtract, and whether to use 1−Φ(z).

Learn 4 — Inverse Normal

What is Inverse Normal?

In inverse normal problems, you are given a probability and asked to find the value of x. This is the reverse of finding P(X < a).

P(X < x) = p ⇒ z = Φ²(p) ⇒ x = μ + zσ

Type 1: Find x Given P(X < x) = p

Example: X~N(60, 25). Find x such that P(X < x) = 0.9382.
σ = √25 = 5

Step 1: Look up 0.9382 in normal tables to find z: z = 1.54
Step 2: Back-transform: x = μ + zσ = 60 + 1.54 × 5 = 60 + 7.7 = 67.7
Check: P(X < 67.7) = Φ((67.7−60)/5) = Φ(1.54) = 0.9382 ✓

Type 2: Find x Given P(X > x) = p

Example: X~N(40, 9). Find x such that P(X > x) = 0.0668.
P(X > x) = 0.0668 ⇒ P(X < x) = 1 − 0.0668 = 0.9332
Look up 0.9332: z = 1.5
x = 40 + 1.5 × 3 = 40 + 4.5 = 44.5

Type 3: Find x When z is Negative

Example: X~N(80, 36). Find x such that P(X < x) = 0.1587.
P(X < x) = 0.1587 — this is less than 0.5, so x < μ and z < 0.
By symmetry: P(X < x) = 0.1587 means 1 − Φ(z) = 0.1587, so Φ(z) = 0.8413 ⇒ z = 1.0
Since x < μ, z = −1.0
x = 80 + (−1.0)(6) = 80 − 6 = 74

Finding Unknown μ (Mean Unknown)

Example: X~N(μ, 25). P(X < 70) = 0.9772. Find μ.
z = (70−μ)/5
From tables: Φ(2.0) = 0.9772, so z = 2.0
(70−μ)/5 = 2.0 ⇒ 70−μ = 10 ⇒ μ = 60

Finding Unknown σ (SD Unknown)

Example: X~N(50, σ²). P(X > 60) = 0.0228. Find σ.
P(X > 60) = 0.0228 ⇒ P(X < 60) = 0.9772 ⇒ z = 2.0
(60−50)/σ = 2.0 ⇒ 10 = 2σ ⇒ σ = 5

Finding Both μ and σ Simultaneously

Example: X~N(μ, σ²). P(X < 30) = 0.2119 and P(X < 55) = 0.9332. Find μ and σ.

Step 1: Convert each probability to a z-score.
P(X < 30) = 0.2119 = Φ(−0.8) ⇒ z = −0.8
P(X < 55) = 0.9332 = Φ(1.5) ⇒ z = 1.5

Step 2: Set up equations from Z = (X−μ)/σ:
−0.8 = (30−μ)/σ ⇒ 30 − μ = −0.8σ   ...(i)
1.5 = (55−μ)/σ ⇒ 55 − μ = 1.5σ   ...(ii)

Step 3: Subtract (i) from (ii):
25 = 2.3σ ⇒ σ = 25/2.3 ≈ 10.87
From (i): μ = 30 + 0.8(10.87) ≈ 30 + 8.70 = μ ≈ 38.70
For simultaneous equation problems: always start by converting probabilities to z-scores using the symmetry rule if needed, then write one equation per condition, subtract to eliminate μ and find σ first, then substitute back.

Learn 5 — Normal Approximation to Binomial

When to Use the Approximation

When X~B(n,p) and n is large, the binomial becomes hard to compute directly. If n is large enough, X is approximately normal:

If X~B(n,p) with np > 5 and nq > 5 (where q = 1−p),
then X ≈ Y~N(np, npq)
Mean of approximation: μ = np
Variance of approximation: σ² = npq = np(1−p)

Example: X~B(100, 0.4). Check: np = 40 > 5 ✓ and nq = 60 > 5 ✓
So X ≈ Y~N(40, 24)

Continuity Correction

The binomial is discrete but the normal is continuous. We must apply a continuity correction of 0.5 to account for this gap.

Binomial X (discrete) → Normal Y (continuous) with correction:
P(X = k)  →  P(k−0.5 < Y < k+0.5)
P(X ≤ k)  →  P(Y < k+0.5)
P(X < k)  →  P(Y < k−0.5)
P(X ≥ k)  →  P(Y > k−0.5)
P(X > k)  →  P(Y > k+0.5)

Memory Trick for Continuity Corrections

Think of each integer k as occupying the interval [k−0.5, k+0.5). The correction extends the boundary by 0.5 in the direction that includes more values:
• ≤ k includes k, so extend up to k+0.5
• < k excludes k, so stop at k−0.5
• ≥ k includes k, so start from k−0.5
• > k excludes k, so start from k+0.5

Full Worked Example

Example: X~B(200, 0.3). Using a suitable approximation, find P(X ≤ 55).

Step 1: Check conditions.
np = 200 × 0.3 = 60 > 5 ✓   nq = 200 × 0.7 = 140 > 5 ✓

Step 2: Find parameters of normal approximation.
Y~N(np, npq) = N(60, 200 × 0.3 × 0.7) = N(60, 42)
σ = √42 ≈ 6.481

Step 3: Apply continuity correction.
P(X ≤ 55) → P(Y < 55.5)

Step 4: Standardise.
z = (55.5 − 60)/6.481 = −4.5/6.481 ≈ −0.694

Step 5: Find probability.
P(Y < 55.5) = Φ(−0.694) = 1 − Φ(0.694) ≈ 1 − 0.7560 = 0.2440

Another Example: P(X = k)

Example: X~B(150, 0.4). Using normal approximation, find P(X = 65).
np = 60 > 5 ✓; nq = 90 > 5 ✓ → Y~N(60, 36), σ = 6
P(X = 65) → P(64.5 < Y < 65.5)
z⊂1; = (64.5−60)/6 = 0.75 ⇒ Φ(0.75) = 0.7734
z⊂2; = (65.5−60)/6 = 0.917 ⇒ Φ(0.917) = 0.8204
P = 0.8204 − 0.7734 = 0.0470
Never forget: Always state the conditions np > 5 and nq > 5 before using the approximation. Examiners award marks for this. Also always state the continuity correction you used — write it explicitly.

Worked Examples

Example 1 — Find P(X < a)

X~N(45, 100). Find P(X < 58).

Step 1: Identify parameters. μ = 45, σ² = 100, so σ = 10.
Step 2: Standardise. z = (58 − 45)/10 = 13/10 = 1.3
Step 3: Look up table. P(X < 58) = Φ(1.3) = 0.9032 M1 A1
Diagram: bell curve centred at 45, shade entire left region up to 58. Since 58 > 45, z is positive, Φ(z) > 0.5.

Example 2 — Find P(X > b)

X~N(30, 64). Find P(X > 22).

Step 1: μ = 30, σ = 8
Step 2: z = (22 − 30)/8 = −8/8 = −1
Step 3: P(X > 22) = 1 − P(X < 22) = 1 − Φ(−1) = 1 − (1 − Φ(1)) = Φ(1) = 0.8413 M1 A1
Note: P(X > 22) = P(X > value below mean) is greater than 0.5 — makes sense since most of the distribution is above 22.

Example 3 — Find P(a < X < b)

X~N(200, 400). Find P(190 < X < 216).

Step 1: μ = 200, σ = 20
Step 2: z⊂1; = (190 − 200)/20 = −0.5    z⊂2; = (216 − 200)/20 = 0.8
Step 3: P(X < 190) = Φ(−0.5) = 1 − Φ(0.5) = 1 − 0.6915 = 0.3085
Step 4: P(X < 216) = Φ(0.8) = 0.7881
Step 5: P(190 < X < 216) = 0.7881 − 0.3085 = 0.4796 M1 M1 A1

Example 4 — Inverse Normal: Find x

X~N(70, 25). Find x such that P(X < x) = 0.8944.

Step 1: μ = 70, σ = 5
Step 2: Look up 0.8944 in tables. Φ(1.25) = 0.8944, so z = 1.25
Step 3: x = μ + zσ = 70 + 1.25 × 5 = 70 + 6.25 = 76.25 M1 A1

Example 5 — Find μ Given a Probability

X~N(μ, 36). P(X < 80) = 0.9772. Find μ.

Step 1: σ = 6. Look up 0.9772: z = 2.0
Step 2: Set up equation: (80 − μ)/6 = 2.0
Step 3: 80 − μ = 12 ⇒ μ = 68 M1 A1

Example 6 — Find σ Given a Probability

X~N(50, σ²). P(X > 65) = 0.0668. Find σ.

Step 1: P(X > 65) = 0.0668 ⇒ P(X < 65) = 1 − 0.0668 = 0.9332
Step 2: Φ(1.5) = 0.9332, so z = 1.5
Step 3: (65 − 50)/σ = 1.5 ⇒ 15/σ = 1.5 ⇒ σ = 10 M1 A1

Example 7 — Find Both μ and σ

X~N(μ, σ²). P(X < 20) = 0.1151 and P(X < 40) = 0.8849. Find μ and σ.

Step 1: P(X<20) = 0.1151 = Φ(−1.2) ⇒ z⊂1; = −1.2
Step 2: P(X<40) = 0.8849 = Φ(1.2) ⇒ z⊂2; = 1.2
Step 3: Equations: (20−μ)/σ = −1.2 and (40−μ)/σ = 1.2
Step 4: Add equations: 60/(2σ) not helpful. Instead note by symmetry: μ = (20+40)/2 = 30
Step 5: (40−30)/σ = 1.2 ⇒ 10/σ = 1.2 ⇒ σ = 8.33 M1 A1 A1

Example 8 — Normal Approximation to Binomial

X~B(120, 0.35). Using normal approximation, find P(X ≥ 50).

Step 1: np = 42 > 5 ✓ nq = 78 > 5 ✓ → Y~N(42, 120×0.35×0.65) = N(42, 27.3), σ = √27.3 ≈ 5.225
Step 2: Continuity correction: P(X ≥ 50) → P(Y > 49.5)
Step 3: z = (49.5 − 42)/5.225 = 7.5/5.225 ≈ 1.436
Step 4: P(Y > 49.5) = 1 − Φ(1.436) ≈ 1 − 0.9246 = 0.0754 M1 M1 A1

Common Mistakes

Mistake 1 — Using σ² Instead of σ in the Formula

WRONG: X~N(50, 16). z = (58−50)/16 = 0.5   ✗
RIGHT: σ = √16 = 4. z = (58−50)/4 = 2.0   ✓

The formula Z = (X−μ)/σ uses σ, not σ². Always take the square root of the variance first.

Mistake 2 — Applying Symmetry Rule Incorrectly

WRONG: Φ(−z) = Φ(z)   ✗
RIGHT: Φ(−z) = 1 − Φ(z)   ✓

For example, Φ(−1.5) = 1 − Φ(1.5) = 1 − 0.9332 = 0.0668. The symmetry rule means the left tail equals the right tail.

Mistake 3 — Forgetting the Continuity Correction

WRONG: B(100,0.4) approx, P(X ≤ 45) → P(Y < 45) with z = (45−40)/√24   ✗
RIGHT: P(X ≤ 45) → P(Y < 45.5) — always add 0.5 for ≤   ✓

Without continuity correction, the approximation is less accurate and you lose accuracy marks.

Mistake 4 — Wrong Direction of Continuity Correction

WRONG: P(X ≥ 50) → P(Y > 50.5)   ✗
RIGHT: P(X ≥ 50) → P(Y > 49.5) — ≥ includes 50, so start below   ✓

Mnemonic: If the inequality includes the endpoint (le/ge), move 0.5 towards it. If it excludes (lt/gt), move 0.5 away from it.

Mistake 5 — Writing N(μ, σ) Instead of N(μ, σ²)

WRONG: X~N(50, 4) when you mean mean=50, sd=4   ✗
RIGHT: X~N(50, 16) since σ² = 16 when σ = 4   ✓

Cambridge uses N(μ, σ²) notation. The second parameter is always the variance.

Mistake 6 — Subtracting in the Wrong Order for P(a < X < b)

WRONG: P(a<X<b) = Φ(z⊂1;) − Φ(z⊂2;) where z⊂2; > z⊂1;   ✗
RIGHT: P(a<X<b) = Φ(z⊂2;) − Φ(z⊂1;) — larger minus smaller   ✓

Standardise both boundaries, then subtract the smaller Φ value from the larger.

Mistake 7 — Not Stating Conditions for Normal Approximation

WRONG: Just using normal approximation without checking np > 5 and nq > 5   ✗
RIGHT: State np = ... > 5 ✓ and nq = ... > 5 ✓ before proceeding   ✓

Examiners award a method mark for checking conditions. Always verify both np > 5 and n(1−p) > 5.

Mistake 8 — Sign Error When Finding Negative z for Inverse Normal

WRONG: P(X < x) = 0.1587 ⇒ z = 1.0 ⇒ x = μ + 1.0σ (ignoring that x < μ)   ✗
RIGHT: P(X < x) = 0.1587 < 0.5 so x < μ, hence z = −1.0 ⇒ x = μ − σ   ✓

Whenever the given probability is less than 0.5, the z-value is negative and x is below the mean.

Key Formulas

Core Standardisation and Probability Formulas

FormulaDescriptionWhen to use
Z = (X−μ)/σStandardise X~N(μ,σ²) to Z~N(0,1)Before every probability calculation
X = μ + ZσBack-transform z-score to original unitsInverse normal problems
P(X<a) = Φ(z)CDF of standard normalz > 0; look up in tables
P(X>a) = 1−Φ(z)Upper tail probabilityRight-hand probabilities
Φ(−z) = 1−Φ(z)Symmetry of N(0,1)Negative z-scores
P(a<X<b) = Φ(z⊂2;)−Φ(z⊂1;)Interval probabilityBoth boundaries given

Normal Approximation to Binomial

ItemFormula / Rule
Conditionsnp > 5 AND nq > 5 (where q = 1−p)
Approximating distributionY~N(np, npq)
Meanμ = np
Varianceσ² = npq = np(1−p)

Continuity Correction Table

Binomial (discrete)Normal approximation (continuous)
P(X = k)P(k−0.5 < Y < k+0.5)
P(X ≤ k)P(Y < k+0.5)
P(X < k)P(Y < k−0.5)
P(X ≥ k)P(Y > k−0.5)
P(X > k)P(Y > k+0.5)
P(a ≤ X ≤ b)P(a−0.5 < Y < b+0.5)

68–95–99.7 Rule

P(μ−σ < X < μ+σ) ≈ 0.6827
P(μ−2σ < X < μ+2σ) ≈ 0.9545
P(μ−3σ < X < μ+3σ) ≈ 0.9973

Key Φ Values to Memorise

zΦ(z)1−Φ(z)
1.000.84130.1587
1.280.89970.1003
1.6450.95000.0500
1.960.97500.0250
2.000.97720.0228
2.3260.99000.0100
2.5760.99500.0050
3.000.99870.0013

Proof Bank

Proof 1 — Justification of the 68–95–99.7 Rule

For Z~N(0,1), we want to show P(−1 < Z < 1) ≈ 0.6827.

By definition, P(−1 < Z < 1) = Φ(1) − Φ(−1) = Φ(1) − (1−Φ(1)) = 2Φ(1) − 1.

From standard normal tables: Φ(1.00) = 0.8413.

Therefore P(−1 < Z < 1) = 2(0.8413) − 1 = 1.6826 − 1 = 0.6826 ≈ 68.27%


Similarly: P(−2 < Z < 2) = 2Φ(2)−1 = 2(0.9772)−1 = 0.9544 ≈ 95.44%

P(−3 < Z < 3) = 2Φ(3)−1 = 2(0.9987)−1 = 0.9974 ≈ 99.74%


For a general X~N(μ,σ²), we standardise: P(μ−σ < X < μ+σ) = P(−1 < Z < 1) ≈ 0.6827, and similarly for 2σ and 3σ bounds. These results hold for any normal distribution regardless of μ and σ.

Proof 2 — Why the Continuity Correction is Needed

The core problem: A discrete distribution (binomial) assigns positive probability to integer values only. A continuous distribution (normal) assigns zero probability to any single point. We must bridge this gap.


Geometric justification: Imagine a bar chart of the binomial distribution. Each integer k is represented by a bar from k−0.5 to k+0.5 with height P(X=k). The area of bar k equals P(X=k).


When we approximate with a continuous normal curve, P(X ≤ k) for the binomial is the total area of bars up to and including bar k — but bar k extends from k−0.5 to k+0.5. So the continuous analogue should include up to k+0.5:

P(X ≤ k) ≈ P(Y ≤ k + 0.5)

Without this correction, we would be cutting bar k in half, systematically underestimating the probability by approximately P(X=k)/2.


How accurate is the approximation? The error in the normal approximation (after continuity correction) is of order 1/√n, which is small when n is large. This is why we require np > 5 and nq > 5 to ensure n is large enough for the error to be acceptably small.


Formal statement (Berry–Esseen theorem, simplified): If X~B(n,p) and Y~N(np, npq), then for all x:
|P(X ≤ k) − P(Y ≤ k+0.5)| ≤ C/(n√(pq))
for a constant C, so the error shrinks as n grows.

Proof 3 — The Symmetry Rule Φ(−z) = 1−Φ(z)

The standard normal distribution has probability density function f(z) = (1/√(2π)) e^(−z²/2), which is symmetric about z = 0, i.e. f(−z) = f(z).


By definition, Φ(z) = P(Z ≤ z) = area under curve to the left of z.

By symmetry: area to left of −z = area to right of z (mirror image).

Area to right of z = 1 − Φ(z).

Therefore: Φ(−z) = 1 − Φ(z).   □

Normal Distribution Visualiser

Enter the distribution parameters and boundaries to compute and visualise a probability. Uses JavaScript approximation of Φ(z).

Enter values and press Calculate.

How to Use

• Enter μ (mean) and σ (standard deviation — NOT variance!)
• Enter lower and upper bounds for the shaded region
• For P(X < a): set lower bound to a very large negative number (e.g. −999) and upper = a
• For P(X > b): set lower = b and upper = 999
• For P(a < X < b): set lower = a and upper = b
• The probability is computed numerically using an accurate series approximation of Φ(z)
Try: μ=0, σ=1, lower=−1, upper=1 → you should get ≈ 0.6827 (the 68% rule!)

Exercise 1 — Features of the Normal Distribution (10 questions)

Self-marking. Enter numeric answers where asked; enter 1 for Yes and 0 for No for yes/no questions.

Exercise 2 — Standardising and Z-Scores (10 questions)

Self-marking. Give z-scores to 2 decimal places where needed.

Exercise 3 — Finding Probabilities (10 questions)

Self-marking. Give answers to 4 decimal places. Use standard normal table values.

Exercise 4 — Inverse Normal: Find x (10 questions)

Self-marking. Give answers to 2 decimal places.

Exercise 5 — Normal Approximation to Binomial (10 questions)

Self-marking. Always apply continuity correction. Give answers to 4 decimal places.

Practice — 30 Mixed Questions

Mixed normal distribution questions. Confetti on 100%!

Challenge — 15 Harder Questions

Harder and multi-step problems requiring deeper understanding.

Exam Style Questions — 8 Cambridge S1 Questions

Cambridge-style structured questions. Show full working. Marks shown in brackets.

Question 1 [5 marks]

The heights of students in a school are modelled by X~N(165, 81).
(a) Find P(X < 174).   [2]
(b) Find P(153 < X < 174).   [3]

σ = √81 = 9
(a) z = (174−165)/9 = 1.0; P(X<174) = Φ(1.0) = 0.8413 [M1 A1]
(b) z⊂1; = (153−165)/9 = −4/3 ≈ −1.333; P(X<153) = 1−Φ(1.333) ≈ 1−0.9088 = 0.0912
P(153<X<174) = 0.8413 − 0.0912 = 0.7501 [M1 M1 A1]

Question 2 [4 marks]

X~N(μ, 49). Given that P(X > 85) = 0.0401, find μ.

σ = 7; P(X>85) = 0.0401 ⇒ P(X<85) = 0.9599
Φ(1.75) = 0.9599, so z = 1.75 [M1]
(85−μ)/7 = 1.75 ⇒ 85−μ = 12.25 ⇒ μ = 72.75 [M1 A1 A1]

Question 3 [5 marks]

X~N(40, σ²). Given that P(X < 33) = 0.1587, find σ and hence find P(X > 46).

P(X<33) = 0.1587 = Φ(−1.0), so z = −1.0 [M1]
(33−40)/σ = −1.0 ⇒ −7/σ = −1 ⇒ σ = 7 [A1]
P(X>46): z = (46−40)/7 = 6/7 ≈ 0.857 [M1]
P(X>46) = 1 − Φ(0.857) ≈ 1 − 0.8042 = 0.1958 [M1 A1]

Question 4 [6 marks]

X~N(μ, σ²). Given P(X < 28) = 0.2266 and P(X < 42) = 0.9332. Find μ and σ.

P(X<28) = 0.2266 = Φ(−0.75), so (28−μ)/σ = −0.75  ...(i) [M1]
P(X<42) = 0.9332 = Φ(1.5), so (42−μ)/σ = 1.5  ...(ii) [M1]
From (i): μ − 0.75σ = 28  ...(i')
From (ii): μ + 1.5σ = 42  ...(ii')
(ii')−(i'): 2.25σ = 14 ⇒ σ = 14/2.25 ≈ 6.22 [M1 A1]
μ = 28 + 0.75(6.22) ≈ 28 + 4.67 = μ ≈ 32.67 [M1 A1]

Question 5 [5 marks]

The mass of apples is normally distributed with mean 140 g and standard deviation 18 g.
(a) Find the probability that a randomly chosen apple has mass greater than 158 g.   [2]
(b) Find the mass m such that 90% of apples have mass less than m.   [3]

(a) z = (158−140)/18 = 1.0; P(X>158) = 1−Φ(1.0) = 1−0.8413 = 0.1587 [M1 A1]
(b) P(X<m) = 0.90 ⇒ z = 1.282 [M1]
m = 140 + 1.282×18 = 140 + 23.08 ≈ 163.1 g [M1 A1]

Question 6 [6 marks]

X~B(200, 0.45). Using a suitable approximation, find:
(a) P(X ≤ 80),   (b) P(85 ≤ X ≤ 95).   [3, 3]

np = 90, nq = 110 both > 5 ✓. Y~N(90, 200×0.45×0.55) = N(90, 49.5), σ = √49.5 ≈ 7.036 [M1]
(a) P(X≤80) → P(Y<80.5); z = (80.5−90)/7.036 ≈ −1.350 [M1]
P = 1−Φ(1.350) ≈ 1−0.9115 = 0.0885 [A1]
(b) P(85≤X≤95) → P(84.5<Y<95.5) [M1]
z⊂1;=(84.5−90)/7.036 ≈ −0.782; z⊂2;=(95.5−90)/7.036 ≈ 0.782
P = 2Φ(0.782)−1 = 2(0.7829)−1 = 0.5658 [M1 A1]

Question 7 [5 marks]

X~N(50, σ²) and P(45 < X < 55) = 0.6826. Find σ, and hence P(X > 58).

P(45<X<55) = 0.6826 ≈ 0.6827 = P(μ−σ < X < μ+σ) for a normal distribution [M1]
This means μ+σ = 55 and μ−σ = 45 ⇒ μ=50 ✓ and σ = 5 [A1]
P(X>58): z = (58−50)/5 = 1.6 [M1]
P(X>58) = 1−Φ(1.6) = 1−0.9452 = 0.0548 [M1 A1]

Question 8 [6 marks]

The lifetimes of batteries are normally distributed. 10% last less than 8 hours and 5% last more than 20 hours. Find the mean and standard deviation of battery lifetime.

P(X<8) = 0.10 ⇒ z = −1.282 ⇒ (8−μ)/σ = −1.282  ...(i) [M1]
P(X>20) = 0.05 ⇒ P(X<20)=0.95 ⇒ z=1.645 ⇒ (20−μ)/σ=1.645  ...(ii) [M1]
(i): μ − 1.282σ = 8
(ii): μ + 1.645σ = 20
Subtract: 2.927σ = 12 ⇒ σ = 12/2.927 ≈ 4.10 hours [M1 A1]
μ = 8 + 1.282(4.10) ≈ 8 + 5.26 = μ ≈ 13.26 hours [M1 A1]

Past Paper Questions

Representative Cambridge 9709 S1 past paper questions on Normal Distribution. Attempt fully before revealing.

Past Paper Q1 — (Nov 2018 style) [7 marks]

The random variable X~N(μ, σ²). It is given that P(X > 23.5) = 0.2 and P(X < 14.5) = 0.1.
(a) Find μ and σ.   [5]
(b) Find P(17 < X < 22).   [2]

(a) P(X>23.5) = 0.2 ⇒ P(X<23.5) = 0.8 ⇒ z = 0.842 ⇒ (23.5−μ)/σ = 0.842 ...(i)
P(X<14.5) = 0.1 ⇒ z = −1.282 ⇒ (14.5−μ)/σ = −1.282 ...(ii)
(i)−(ii): 9/σ = 2.124 ⇒ σ = 9/2.124 ≈ 4.238
μ = 23.5 − 0.842(4.238) ≈ 23.5 − 3.57 = μ ≈ 19.93
(b) z⊂1; = (17−19.93)/4.238 ≈ −0.692; z⊂2; = (22−19.93)/4.238 ≈ 0.489
P = Φ(0.489) − Φ(−0.692) = 0.6876 − (1−0.7554) = 0.6876−0.2446 = 0.4430

Past Paper Q2 — (Jun 2019 style) [5 marks]

The masses of oranges are normally distributed with mean 190 g and standard deviation 15 g. An orange is selected at random.
(a) Find the probability the orange has mass between 175 g and 208 g.   [3]
(b) Find the value of m such that P(X > m) = 0.25.   [2]

(a) z⊂1; = (175−190)/15 = −1; P(X<175) = 1−0.8413 = 0.1587
z⊂2; = (208−190)/15 = 1.2; P(X<208) = Φ(1.2) = 0.8849
P(175<X<208) = 0.8849−0.1587 = 0.7262
(b) P(X>m) = 0.25 ⇒ P(X<m) = 0.75 ⇒ z = 0.674
m = 190 + 0.674(15) = 190 + 10.11 = 200.11 g

Past Paper Q3 — (Nov 2020 style) [7 marks]

X~B(180, 0.3). Using a normal approximation, find P(X ≤ 60) and P(50 < X < 65).

np = 54 > 5 ✓; nq = 126 > 5 ✓ → Y~N(54, 180×0.3×0.7) = N(54, 37.8), σ≈6.148
P(X≤60) → P(Y<60.5): z=(60.5−54)/6.148≈1.057; P=Φ(1.057)≈0.8547 ≈ 0.855
P(50<X<65) → P(50.5<Y<64.5):
z⊂1;=(50.5−54)/6.148≈−0.569; z⊂2;=(64.5−54)/6.148≈1.708
P = Φ(1.708)−Φ(−0.569) = 0.9562−(1−0.7153) = 0.9562−0.2847 = 0.6715

Past Paper Q4 — (Jun 2021 style) [6 marks]

The times taken by students to complete a test follow X~N(45, σ²). Given that 20% of students take more than 55 minutes, find σ. Hence find the probability a student completes the test in less than 35 minutes.

P(X>55) = 0.20 ⇒ P(X<55) = 0.80 ⇒ z = 0.842
(55−45)/σ = 0.842 ⇒ 10/σ = 0.842 ⇒ σ = 11.876 ≈ 11.88
P(X<35): z = (35−45)/11.88 = −10/11.88 ≈ −0.842
P(X<35) = Φ(−0.842) = 1−Φ(0.842) = 1−0.8 = 0.2
(By symmetry: P(X<35) = P(X>55) = 0.20, since 35 and 55 are symmetric about μ=45.)

Past Paper Q5 — (Nov 2022 style) [8 marks]

The lengths of steel rods produced are normally distributed with mean μ and standard deviation 2.5 cm. A rod is acceptable if its length is between 95 cm and 103 cm.
(a) Given μ = 99, find the probability that a randomly chosen rod is acceptable.   [3]
(b) The manufacturer wants at least 98% of rods to be acceptable. If σ = 2.5, find the range of values of μ.   [5]

(a) z⊂1; = (95−99)/2.5 = −1.6; z⊂2; = (103−99)/2.5 = 1.6
P(95<X<103) = 2Φ(1.6)−1 = 2(0.9452)−1 = 0.8904
(b) P(95<X<103) ≥ 0.98
The interval [95,103] has width 8 cm and σ = 2.5. For symmetry, μ should ideally be 99. We need P(X<95) ≤ 0.01 AND P(X>103) ≤ 0.01.
P(X<95) ≤ 0.01: (95−μ)/2.5 ≤ −2.326 ⇒ 95−μ ≤ −5.815 ⇒ μ ≥ 100.815
P(X>103) ≤ 0.01: (103−μ)/2.5 ≥ 2.326 ⇒ 103−μ ≥ 5.815 ⇒ μ ≤ 97.185
No single μ satisfies both — so to maximise coverage, μ = 99 gives 89% — the manufacturer must reduce σ or set μ = 99 and accept <98%.