Homework: pg. 713 #35, 37, 38 35.) A.) The sample is not an SRS so may not be representative of the population, Normality is reasonable due to large sample size. Independence is met because the population of pregnant women in Guatemala is more than 1600. B.) Ho:µ=9.5, Ha:µ≠9.5; z=2.21; P-value=0.0272; We reject the null. If the mean were 9.5, the probability of getting a sample mean of 9.57 is 2.72%. The mean calcium level in healthy, pregnant Guatemalan women likely differs from 9.5. C.) (9.508, 9.632) We are 95% confident that the mean blood calcium for these women is between these two numbers.

37.) A. Yes. We would not reject the null so it’s reasonable so it would be in our confidence interval. No. We would reject the null at the 10% level so the null is not reasonable and won’t be in our confidence interval. 38.) A. No. 13 is in the 90% confidence interval so the null cannot be rejected. **B. No. The sample mean is less than one standard error away from the mean. Yes. 10 is not in the 90% confidence interval so the null can be rejected at the 10% level. **D. It depends on which side; µ<10, the answer is no. µ>10, the answer is yes.

Use and Abuse of Significance Tests Section 11.3

#1: Choosing a significance level How much evidence is needed to convince people? What are the consequences of rejecting the Null Hypothesis?

#2: Statistical Significance vs. Practical Importance Just because something is statistically significant, does not mean it is important and vice versa. Example: travel time to school

#3: Statistical Inference is not valid for all sets of data You need to know how the data is collected Example: Music/work productivity

#4: Multiple Analyses Don’t collect a lot of data and then see what is significant or not. Chances are something will be significant. Example: wanting to raise ACT scores and measuring all different things at once

Pg. 721 #43,44,46,47 Section 11.3