# Pop of differences are approx normal 1 for qqplot for

• Test Prep
• 6
• 100% (4) 4 out of 4 people found this document helpful

This preview shows page 5 - 6 out of 6 pages.

+1 pop of differences are approx normal; +1 for qqplot for normality; + 1 each difference is independent
(b) Perform the appropriate t hypothesis test at an alpha = 5% level regarding the research question.
ii. (3 points) Test statistic and p value (Show computations):
iii. (3 points) Draw a conclusion in the context of the question:
(c) (4 points) Suppose that instead of performing the t hypothesis test, the researcher wanted a [two-sided] confidence interval from a “bootstrap t” distribution. Some relevant quantiles are given for the bootstrap distribution below. Construct a 95% confidence interval using the bootstrap t distribution. Percent Below: 2.5% 5% 10% 90% 95% 97.5% Bootstrap t value: -2.64 -2.12 -1.54 1.26 1.67 2.06
5
Summary Statistics included for reference: Area Sample Mean Sample Median Sample SD Size Outter 10.40 10.23 0.90 10 Core 11.20 11.14 1.07 10 Core-Outter 0.80 0.71 0.99 10 (d) (6 points) Suppose that instead of either of the analyses above, the researcher ignored the study design and chose to do a two independent sample t test, with hypotheses: H 0 : μ core - μ out = 0 vs H 0 : μ core - μ out > 0 assuming equal population variances . Describe whether each of the following values would increase, decrease, or not change (You need to include some computations or clear explanation for your answer) compared to part b.