Types of Normal Distribution Problems — How to Identify Them
🎯 The Most Common Mistake
Most errors in normal distribution problems come not from wrong calculation — but fromstarting to compute before understanding what is being asked..
The fix: Before every calculation, stop and ask: "What type of problem is this?"
Most errors in normal distribution problems come not from wrong calculation — but fromstarting to compute before understanding what is being asked..
The fix: Before every calculation, stop and ask: "What type of problem is this?"
The Four Problem Types
Every normal distribution question belongs to one of four types. Once you identify the type — half the work is done..
Type 1 — Find Probability from a Value
📋 Characteristics:
Baby weight is normally distributed: \(\mu = 3.3\) kg, \(\sigma = 0.4\) kg.
What is the probability a baby weighs less than 3.7 kg??
- Given: value \(X\) (or Z-score \(Z\))
- Find: probability (area under the curve)
- Key words: "what is the probability that…", "what percentage of…", "what fraction…"
- Convert \(X\) to \(Z\) (if not given): \(Z = \dfrac{X - \mu}{\sigma}\)
- Look up \(\Phi(z)\) in the table
- Adjust for left/right tail as required
Baby weight is normally distributed: \(\mu = 3.3\) kg, \(\sigma = 0.4\) kg.
What is the probability a baby weighs less than 3.7 kg??
Step 1: \(Z = \dfrac{3.7 - 3.3}{0.4} = \dfrac{0.4}{0.4} = 1\)
Step 2: From table: \(\Phi(1) = 0.8413\)
Step 3: "Less than" = left-tail area = \(\Phi(1)\) directly from table.
✅ Answer: \(P(X < 3.7) = 0.8413\), meaning about 84% of babies weigh less than 3.7 kg.
Step 2: From table: \(\Phi(1) = 0.8413\)
Step 3: "Less than" = left-tail area = \(\Phi(1)\) directly from table.
✅ Answer: \(P(X < 3.7) = 0.8413\), meaning about 84% of babies weigh less than 3.7 kg.
Type 2 — Probability Within an Interval
📋 Characteristics:
\(\mu = 3.3\), \(\sigma = 0.4\). What is the probability a baby weighs between 2.9 and 3.7 kg??
- Given: Two values (\(a\) and \(b\))
- Find: Probability the value lies between the two bounds
- Key words: "between … and …", "from … to …"
Formula:
\(P(a \le X \le b) = \Phi(z_b) - \Phi(z_a)\)
🔢 Solution steps: \(P(a \le X \le b) = \Phi(z_b) - \Phi(z_a)\)
- Convert both values to Z:\(Z\)
- Look up \(\Phi\) for each in the Z-table
- Subtract: larger minus smaller
\(\mu = 3.3\), \(\sigma = 0.4\). What is the probability a baby weighs between 2.9 and 3.7 kg??
Step 1 — Convert:
\(z_1 = \dfrac{2.9 - 3.3}{0.4} = -1\) \(z_2 = \dfrac{3.7 - 3.3}{0.4} = 1\)
Step 2 — Table:
\(\Phi(1) = 0.8413\) \(\Phi(-1) = 1 - \Phi(1) = 1 - 0.8413 = 0.1587\)
Step 3 — Subtract:
\(P(-1 \le Z \le 1) = 0.8413 - 0.1587 = 0.6826\)
✅ Answer: About 68% of babies weigh between 2.9 and 3.7 kg — matching the 68-95-99.7 rule exactly.
\(z_1 = \dfrac{2.9 - 3.3}{0.4} = -1\) \(z_2 = \dfrac{3.7 - 3.3}{0.4} = 1\)
Step 2 — Table:
\(\Phi(1) = 0.8413\) \(\Phi(-1) = 1 - \Phi(1) = 1 - 0.8413 = 0.1587\)
Step 3 — Subtract:
\(P(-1 \le Z \le 1) = 0.8413 - 0.1587 = 0.6826\)
✅ Answer: About 68% of babies weigh between 2.9 and 3.7 kg — matching the 68-95-99.7 rule exactly.
Type 3 — Inverse Problem (From Probability to X)
📋 Characteristics:
\(\mu = 3.3\), \(\sigma = 0.4\). What weight has 90% of babies below it?
- Given: probability (percentage, area)
- Find: value \(X\) corresponding to that probability
- Key words: "find the value such that…", "what score has 90% below it?", "find the nth percentile…"
Inverse formula:
\(X = \mu + Z \cdot \sigma\)
🔢 Solution steps (reverse of Type 1!): \(X = \mu + Z \cdot \sigma\)
- Find \(Z\) in the table corresponding to the given probability
- Substitute into the inverse formula: \(X = \mu + Z \cdot \sigma\)
\(\mu = 3.3\), \(\sigma = 0.4\). What weight has 90% of babies below it?
Step 1: Look up in table: \(\Phi(z) = 0.90\) → \(z \approx 1.28\)
Step 2: \(X = 3.3 + 1.28 \times 0.4 = 3.3 + 0.512 = 3.812\)
✅ Answer: 90% of babies weigh less than \(3.81\) kg (approximately).
Step 2: \(X = 3.3 + 1.28 \times 0.4 = 3.3 + 0.512 = 3.812\)
✅ Answer: 90% of babies weigh less than \(3.81\) kg (approximately).
⚠️ Common mistake in Type 3: Students sometimes substitute the probability (0.90) directly into the formula for \(Z\). This is wrong! First find the Z-value \(Z\) from the table, then substitute.
Type 4 — Comparing Groups
📋 Characteristics:
Sara scored 85 in biology (\(\mu = 75\), \(\sigma = 10\)) and 90 in chemistry (\(\mu = 88\), \(\sigma = 4\)).
In which subject did she perform better relative to the class??
- Given: Scores/values from different groups with different means and SDs
- Find: Who performed better? Who sts out more? Where is the achievement higher?
- Key words: "compare", "in which subject is she better?", "who succeeded more relative to…"
🔑 The rule:
Never compare raw scores — always convert to\(Z\) and compare Z-scores!
🔢 Solution steps: Never compare raw scores — always convert to\(Z\) and compare Z-scores!
- Calculate \(Z\) Z separately for each group
- Compare Z-scores — whoever has a higher \(Z\) Z performed better relative to their group
Sara scored 85 in biology (\(\mu = 75\), \(\sigma = 10\)) and 90 in chemistry (\(\mu = 88\), \(\sigma = 4\)).
In which subject did she perform better relative to the class??
Biology: \(z = \dfrac{85 - 75}{10} = 1\)
Chemistry: \(z = \dfrac{90 - 88}{4} = 0.5\)
✅ Conclusion: Although the raw chemistry score is higher (90 > 85), Sara sts out more in biology (\(z = 1\) compared to chemistry. \(z = 0.5\))!
Chemistry: \(z = \dfrac{90 - 88}{4} = 0.5\)
✅ Conclusion: Although the raw chemistry score is higher (90 > 85), Sara sts out more in biology (\(z = 1\) compared to chemistry. \(z = 0.5\))!
🗺️ How to Identify the Type? — Flow Chart
| Ask yourself | If yes → |
|---|---|
| Given value/score, find probability? | Type 1 |
| Find probability in an interval (between … and …)? | Type 2 |
| Given probability, find value? | Type 3 |
| Find Compare groups? | Type 4 |
Key Formulas When Working with the Z-Table
| What is required | Formula | Explanation |
|---|---|---|
| Left-tail area | \(P(Z \le z) = \Phi(z)\) | Read directly from table |
| Right-tail area | \(P(Z > z) = 1 - \Phi(z)\) | Complement to 1 |
| Interval | \(P(a \le Z \le b) = \Phi(b) - \Phi(a)\) | Subtract two areas |
| Symmetry | \(\Phi(-z) = 1 - \Phi(z)\) | Curve is symmetric about 0 |
| Single value | \(P(Z = z) = 0\) | In a continuous distribution — always 0! |
💡 Tip: The identification step prevents half the mistakes. Before every calculation, ask yourself: "What is given and what am I finding?" — that tells you the problem type.