MATH 1040 - Introduction to Statistics

18.4 Finding the best sample size

Lesson 18.4 Finding the best sample size

Reading

Reading sections are from the Introductory Statistics Textbook

Lesson

How do you know what the sample size \(n\) should be? This is actually not hard. Take the Margin of Error and solve for \(n\):

\[E = z_c \frac{\sigma}{\sqrt{n}} \qquad \to \qquad n = \left(\frac{z_c \sigma}{E}\right)^2\]

So what we need is a critical value (a confidence level), a population standard deviation, and a desired Margin of Error. Then just plug them into the equation to get the desired sample size.

Recall the example problem from 18.3. We’ll modify the question a bit to show what we mean.

American mental abilities are often measured by an IQ test. The IQ distribution is normal with a mean of 100 and a population standard deviation of 15.

What sample size is needed if you want the margin of error to be 1.5 with a 90% confidence level?

\[n = \left(\frac{z_c\sigma}{E}\right)^2 = \left(\frac{1.645\cdot 15}{1.5}\right)^2 = \left(\frac{24.675}{1.5}\right)^2 = 16.45^2 = 270.60\]

The answer is that you need to sample 270.6 people to get a margin of error of 1.5. However, you can’t sample 0.6 people, so we need to round the answer up.

Thus, the sample size needs to be at least 271 people in order to get a margin of error at or below 1.5.

Practice

  1. A university is interested in estimating the average daily commute time for its students. The population standard deviation is known to be 8 minutes. What sample size is needed to get a margin of error less than 3.0 for a 95% confidence interval?
  2. A manufacturer claims that its LED light bulbs have an average lifespan of 1,200 hours. A consumer protection agency wants to verify this claim. The population standard deviation is known to be 100 hours. What sample size is needed to get a margin of error less than 35 for a 99% confidence interval?
  3. A national education researcher wants to estimate the average SAT Math score for high school seniors in a particular state. The population standard deviation is known to be 100 points. What sample size is needed to get a margin of error less than 25 for a 92% confidence interval?