1. Explain how the purpose of estimation differs from the purpose of a hypothesis test.
2. Explain why it would not be reasonable to use estimation after a hypothesis test for which the decision was "fail to reject Ho."
3. Explain how each of the following factors affects the width of a confidence interval:
a. Increasing the simple size.
b. Increasing the sample variability.
c. Increasing the level of confidence (the percentage of confidence.
7. A sample is obtained from an unknown population. The sample mean is m=34 with sample variance of s²=36.
a. Assuming n=4, use the data to make a 90% confidence interval estimate of the unknown population mean.
b. Assuming n=16, use the data to make a 90% confidence interval estimate of µ.
c. Assuming n=36, use the data to make a 90% confidence interval estimate of µ.
d. In general, how is the width of a confidence interval related to the size of the sample?
1. Explain why the expected value for an f-ratio is equal to 1.00 when there is no treatment effect.
2. Describe the similarities between an f-ratio and a t statistic.
3. Several factors influence the size of the f-ratio. For each of the following, indicate whether it would influence the numerator or the denominator of the f-ratio, and indicate whether the size of the f-ratio would increase or decrease.
a. Increase the differences between the sample means.
b. Increase the size of the sample variances.
4. Explain why you should use ANOVA instead of several t tests to evaluate mean differences when an experiment consists of three or more treatment conditions.
5. Posttests are done after an analysis of variance.
a. What is the purpose for posttests.
b. Explain why you would not do posttests if the analysis is comparing only two treatments.
c. Explain why you would not do posttests if the decision from the ANOVA was to fail to reject the null hypothesis.
This posting contains answers to following descriptive questions on statistics.