turn on suggestions

Auto-suggest helps you quickly narrow down your search results by suggesting possible matches as you type.

Showing results for

Find a Community

- Home
- /
- Analytics
- /
- Stat Procs
- /
- Question on the Binomial

Topic Options

- Subscribe to RSS Feed
- Mark Topic as New
- Mark Topic as Read
- Float this Topic for Current User
- Bookmark
- Subscribe
- Printer Friendly Page

- Mark as New
- Bookmark
- Subscribe
- Subscribe to RSS Feed
- Highlight
- Email to a Friend
- Report Inappropriate Content

06-14-2011 11:30 AM

I was using proc freq to calculate sample size for a binomial proportion and I got numbers that disagree with what I have used in the past for 95% confidence with 95% reliability. Using a sample size of 59 I get:

result Frequency Percent Cumulative

Frequency Cumulative

Percent

Fail 0 0.00 0 0.00

Pass 59 100.00 59 100.00

Binomial Proportion for result = Pass

Proportion (P) 1.0000

ASE 0.0000

95% Lower Conf Limit 1.0000

95% Upper Conf Limit 1.0000

Exact Conf Limits

95% Lower Conf Limit 0.9394

95% Upper Conf Limit 1.0000

I was just curious why SAS gives me a confidence just under 94% and not 95%

result Frequency Percent Cumulative

Frequency Cumulative

Percent

Fail 0 0.00 0 0.00

Pass 59 100.00 59 100.00

Binomial Proportion for result = Pass

Proportion (P) 1.0000

ASE 0.0000

95% Lower Conf Limit 1.0000

95% Upper Conf Limit 1.0000

Exact Conf Limits

95% Lower Conf Limit 0.9394

95% Upper Conf Limit 1.0000

I was just curious why SAS gives me a confidence just under 94% and not 95%

- Mark as New
- Bookmark
- Subscribe
- Subscribe to RSS Feed
- Highlight
- Email to a Friend
- Report Inappropriate Content

06-22-2011 03:46 PM

When you ask for EXACT you get what points actually in the Binomial distribution. For the sample size you have 0.9394 is as close as the distribution allows without going over.