Return to Lessons

 

Confidence Intervals for Proportions

 

Suppose you survey 200 people and 48 say they smoke.  Then the proportion in the sample of 200 people who smoke is 48/200 = 0.24.

This gives us a point estimate of the true proportion of people who smoke.  The idea is to construct an interval for which there is, say, a 95% chance

that the interval contains the true population proportion.  This would be the 95% confidence interval for the proportion.

 

Note:  We use p for the true population proportion and q for 1 - p.

 

Note:  This is valid for large values of n.  The usual criteria is np > 5 and nq > 5, a condition that is easily satisfied in the examples we study.

You can either evaluate it on a calculator or get it from the last line of the t-table.

 

Example 1:  You survey 173 students and 123 say they drive to campus daily.  Find the 90% confidence interval for the

proportion of students who drive to campus daily.

 

There is a 90% chance that between 65.4% and 76.8% of students drive to campus daily.

 

Note that you get a fairly wide range for the confidence interval.  It turns out that to get a narrow range you need large samples.

 

Sample size

 

Example 2.  Suppose you wish to know with 99% confidence the proportion of voters who will vote for candidate Jones in an upcoming election

with an error of no more than 2 percentage points.  That is, E = 0.02.

 

(a)  Determine the sample size needed supposing you have no idea what proportion will vote for Jones.

                                     

In order to sure n is big enough always round up.  Survey 4,148 voters.

 

(b)  Determine the sample size needed supposing an earlier polled showed 60% will vote for Jones.

Question:  Why would you want to limit the sample size?

Answer:  Gathering data is expensive!

 

Return to Lessons