Algebra 2 · Lesson 11.6 · Unit 11 · S-IC.4-6

How much a sample result could be wrong by, and what that permits you to say

Name:   Date:  

1 Why this lesson matters

Read the opening paragraph and the "The method" box on the lesson page. In your own words, restate each idea below.

A sample proportion \( \hat{p} \) estimates the population proportion \( p \).

The margin of error at 95 percent confidence is about \( \dfrac{1}{\sqrt{n}} \), a conservative approximation.

More precisely it is \( 1.96\sqrt{\dfrac{\hat{p}(1 - \hat{p})}{n}} \).

The confidence interval is \( \hat{p} \pm \) the margin of error.

Quadrupling the sample halves the margin, since it depends on \( \sqrt{n} \).

The margin covers sampling error only, never bias.

2 Worked example

Read the "Worked example" box. Then, without looking back, try writing your own answer below.

(a) A poll of 1,000 people finds 52 percent support. Find the margin of error and interpret it. (b) A survey of 200 people finds 120 prefer option A. Can a real majority be claimed? (c) A coin is flipped 20 times and lands heads 15 times. Judge whether it is fair. (d) Explain why the margin of error does not cover bias.

3 Sum it up

In two or three sentences, explain this lesson's main idea to someone who has not read it.