Normal probabilities from context
Big idea: For a normal variable, a probability is an area under the bell curve.
P(X < a) is the area to the left of a.
[Diagram: math-normal-curve] - Available in full study mode
Example: if test scores are normal, P(score < 70) is the proportion of students scoring below 70.
AI SL method — use the GDC: Always write X ~ N(μ, σ²), then use your GDC: normalcdf(lower, upper, μ, σ).
There are no z-scores or z-tables in AI SL — you enter μ and σ straight into the calculator.
How to calculate common normal probabilities
Worked example
Let X~N(100,152).
Find P(X<120).
Step by step
- Write formula first: z=(x-mean)/sd
- Use table/calculator to get P(Z<1.33)
- P(Z<1.33)≈0.908
Final answer
P(X<120)≈0.908.
| Target | Calculator setup |
|---|---|
| P(X<a) | normalcdf(-1E99,a,mean,sd) |
| P(X>a) | normalcdf(a,1E99,mean,sd) |
| P(a<X<b) | normalcdf(a,b,mean,sd) |
Stop wasting time on topics you know
Our AI identifies your weak areas and focuses your study time where it matters. No more overstudying easy topics.
Try Smart Study Free7-day free trial • No card required
Inverse normal: finding cut-off values
When inverse is needed: If the probability is given and x is unknown, use inverse normal.
Worked example
Heights are X~N(170, 62).
Find the height exceeded by top 10%.
Step by step
- Top 10% means left area = 0.90
- Use invNorm(0.90,170,6)
- x≈177.7 cm
Final answer
Top 10% cut-off is about 177.7 cm.
Exam wording: "Exceeded by top 10%" means use cumulative 0.90, not 0.10.
Exam traps and past-paper habits
Common mistakes
- Confusing the upper tail with the lower tail.
- For invNorm, entering the tail area instead of the cumulative (left) area.
- Rounding too early in the working.
Safe method
- Sketch the curve and shade the required region first.
- For ''top p%'', use cumulative area 1 − p in invNorm.
- Keep full accuracy and round only the final answer to 3 s.f.
Exam Tips:
- Write X ~ N(μ, σ²) first.
- Mark the required tail with a quick sketch.
- State the final probability to 3 s.f. unless told otherwise.