Re: margin of error of mean from 3-valued (0,100,200) mutiple-choice survey question
- From: jgsurvey2@xxxxxxxxx
- Date: Wed, 2 Jan 2008 13:36:12 -0800 (PST)
Hi Gus - thanks for the reply! Yes, I have access to individual coded
responses. Here's two typical samples:
200 38.0%
100 47.0%
0 15.0%
sample size: 1000
mean: 123
standard deviation: 69
200 33.0%
100 50.0%
0 17.0%
sample size: 500
mean: 116
standard deviation: 69
Given the samples above, what would be the best formula to compute:
- margin of error, with 95% confidence, for each mean?
- whether the difference between the two represents a statistically
significant change (at 95% confidence) ?
Some follow-up questions:
- does it matter that there are only three choices-- is standard
deviation meaningful in this case?
- does it matter that the distribution is not evenly distributed on
both ends, because there are usually more 200's than 0's?
.
- Follow-Ups:
- Prev by Date: Re: Is this enough information to make an inference?
- Next by Date: Re: I cannot understand the sentence in a probability book.
- Previous by thread: I cannot understand the sentence in a probability book.
- Next by thread: Re: margin of error of mean from 3-valued (0,100,200) mutiple-choice survey question
- Index(es):
Relevant Pages
|
