Question 229900: 1. Describe the effect of increasing the size of a sample on the margin of error of a 95% confidence interval. Address any potential changes to either piece of the formula.
2. Describe the effect of changing a confidence level from 95% to 90% on the margin of error. Address any potential changes to either piece of the formula.
3. Explain clearly and briefly what 95% confidence means.
4. Briefly explain the difference between statistical significance and practical significance.
5. Why is it not necessary to apply inferential methods to a census?
6. When is it ok to use statistics based on a normal distribution (e.g., z-test, t-test), even if the population from which the data is taken is not normally distributed. Why is it ok?
Answer by stanbon(75887)   (Show Source):
You can put this solution on YOUR website! Margin of Error = E = z[s/sqrt(n))
-------------------------------
1. Describe the effect of increasing the size of a sample on the margin of error of a 95% confidence interval. Address any potential changes to either piece of the formula.
---
As n increases E decreases
-----------------------------------
2. Describe the effect of changing a confidence level from 95% to 90% on the margin of error. Address any potential changes to either piece of the formula.
---
E decreases
-----------------------------------
3. Explain clearly and briefly what 95% confidence means.
We are 95% confident the population mean is between the
lower limit and the upper limit.
------------------------------------
4. Briefly explain the difference between statistical significance and practical significance.
---
Statistical significance implies a certain percent of confidence is
the decision to reject Ho.
Practical significance is a judgement call made by an indivisual on
the strength of evidence for or against some action.
--------------------------------
5. Why is it not necessary to apply inferential methods to a census?
Because not all individuals can be contacted.
--------------------------------
6. When is it ok to use statistics based on a normal distribution (e.g., z-test, t-test), even if the population from which the data is taken is not normally distributed. Why is it ok?
Check your text. I'll leave that to you.
Cheers,
Stan H.
|
|
|
