Lesson 6 of 7 · 18 min
The normal approximation for p̂
Recognise that p̂ is approximately normal for large samples, check when the approximation is reasonable, and use it with technology to calculate probabilities and quantiles for sample proportions.
What you will be able to do
- Write the approximate distribution for a large sample.
- Check that and are large enough for the approximation to be reasonable.
- Calculate probabilities and quantiles for with technology, and interpret them in context.
Working out for a sample of 500 from the binomial distribution means adding up 141 binomial probabilities. There is a much quicker way. When you simulated for large samples, the display was a familiar shape: a symmetric bell.
Run a few thousand samples. How well does the normal curve fit? Now set and : what goes wrong?
Key idea
The normal approximation
For large random samples, the sample proportion is approximately normally distributed:
with mean and standard deviation .
The approximation is better for larger , and better when is near 0.5. A common rule of thumb is that it is reasonable when
that is, when you expect at least 10 members with the attribute and 10 without. (Some texts use 5; either way, the bigger these are, the better.)
Random samples of 525 are taken from a population with . Which is the approximate distribution of ?
. The shaded area is , found in the worked example below.
Worked example
A normal probability for p̂
Suppose 30 % of Queensland households have rooftop solar panels. A random sample of 525 households is surveyed.
(a) Explain why the distribution of is approximately normal, and state its mean and standard deviation.
(b) Find the probability that the sample proportion with solar panels is between 0.27 and 0.34, correct to three decimal places.
(c) Interpret your answer in terms of the number of households in the sample.
Step 1: Check the rule of thumb
and are both well above 10, so is approximately normal.
Step 2: Find the mean and standard deviation
and , so .
Step 3: Standardise the bounds
and .
Not needed if you enter and directly, but it shows the marker your method.
Step 4: Use technology
Step 5: Interpret as counts
and , so there is about a 91 % chance that between 142 and 178 of the 525 households have solar panels.
Try it yourself first, then check the working one step at a time.
Answer
since and are large. : about a 91 % chance that between 142 and 178 of the households have solar panels.
Method
Check and .
Write with the numbers in.
Use normal CD with and (or standardise first).
Round at the end, and interpret in context.
Watch out for
Entering the variance, 0.0004, as .
Using the counts 141.75 and 178.5 with the proportion's mean and standard deviation.
Directly on the calculator
Enter the mean and standard deviation of and the bounds as proportions.
- Casio: Normal CD with Lower , Upper , , .
- TI:
normalcdf(0.27, 0.34, 0.3, 0.02).
Use a very large bound such as (or ) for a one-sided probability.
By standardising first
Convert the bounds to -scores with :
Then find for the standard normal . Both methods give the same answer.
Suppose 20 % of Australian adults have never flown on a plane. For random samples of 400 adults, use the normal approximation to find , correct to four decimal places.
Common mistake
Variance in the σ box
Calculators ask for the standard deviation , not the variance. Entering instead of its square root makes the distribution far too narrow, and tail probabilities come out almost 0. If an answer looks absurdly small, check this first.
Worked example
Is the sample consistent with the claim?
An electricity retailer advertises that "80 % of our customers are satisfied with our service". A consumer group surveys a random sample of 400 of the retailer's customers, and 300 say they are satisfied.
Assuming the retailer's claim is true, find the probability of a sample proportion as low as or lower than the one observed. Use your answer to comment on the claim.
Step 1: Calculate the observed proportion
Step 2: Model p̂ assuming the claim
If , then and are large, so with .
Step 3: Standardise the observation
: the sample is 2.5 standard deviations below the claimed value.
Step 4: Find the probability
Step 5: Comment on the claim
If 80 % of customers really were satisfied, only about 6 in every 1000 random samples of 400 would give a proportion this low. The observed result would be very unusual, so the sample is strong evidence that fewer than 80 % of customers are satisfied.
This doesn't prove the claim false — unlikely samples do happen — but it makes the claim hard to believe, provided the survey was genuinely random and unbiased.
Try it yourself first, then check the working one step at a time.
Answer
and, if , . A result this low would be very unusual if the claim were true, so the data cast serious doubt on the claim (assuming the sample was random and unbiased).
Method
Assume the claimed and model with the normal approximation.
Find the probability of a result at least as extreme as the one observed.
A very small probability (say below 0.05) means the observation is unusual under the claim, which casts doubt on it.
Watch out for
Finding , which is 0 for a continuous model. Use a tail probability.
Concluding the claim is definitely false. A small probability is evidence, not proof.
Exam tip
Show the model
On Paper 2 a bare calculator answer risks the method marks. Write the model and its parameters — ", since and are large" — then the probability statement, then the answer, rounded as asked.
Note
A note on accuracy
is really discrete (it can only equal ), so the normal model is an approximation. For the large samples where you use it, the error is small. You are not expected to use a continuity correction in this course.
Summary
For large samples, .
Rule of thumb: the approximation is reasonable when and are both at least 10 (some texts say 5).
Calculate probabilities with technology using mean and standard deviation — or standardise and use .
A probability that is very small means the observed would be unusual if the claimed were true.
Finished the lesson?
Try the 2 quick checks above and the lesson completes itself, or mark it complete now.