1Practice Final Exam B ____________________________________________________________________________________________ 1) Which of the following statements are true about hypothesis tests? I We always assume the null hypothesis is true when the P‐value is calculated. II Then the data supports the alternative hypothesis, the further the value of the test statistic from 0, then the stronger the evidence is in favor of the alternative hypothesis III Hypothesis tests are designed to determine probability that the null hypothesis is true or false. a) I only is true b) I and II only are true c) II only is true d) II and III only are true e) I, I and III are all true 2) Suppose you are planning an experiment in which you are going to select a random sample of 236 cars and measure their CO2 emission levels. As part of your report, you decide to give a 95% confidence interval for µ = average emission level. If instead, you report a 99% confidence interval, what would be the impact on your estimation of the parameter µ? I You would decrease the range of plausible values for µ in your report if you report a 99% confidence interval rather than a 95% confidence interval. II The chance you will select a sample whose calculated confidence interval would not contain the true value of the mean would decrease from 5% to 1%. III You increase the chance that µ would be a value in the confidence interval. a) I only is true b) II only is true c) I and II only are true d) II and III only are true e) All are true 3) A public health investigator claims there is an association between drinking habits and marital status of people in their thirties. To answer her question, she takes a SRS of 200 people in their thirties and asks them if they drink 1 or fewer drinks per day or 2 or more drinks per day. She also asks them if they are married, divorced or have never been married? What would be the most appropriate choice for analyzing the resulting data? a) Independent 2 samples t test b) Paired t test c) Chi‐square 2‐way table test d) Linear regression e) No test we’ve studied this semester can be used to analyze this data set. 24) Simpson's Paradox occurs when a) no baseline risk is given, so it is not known whether or not a high relative risk has practical importance. b) a confounding variable rather than the explanatory variable is responsible for a change in the response variable. c) the direction of the relationship between two variables changes when the categories of a confounding variable are taken into account. d) the results of a test are statistically significant but are really due to chance e) we make statistical inferences about population. 5) Data was collected to test a pop ulation mean with the hypotheses: H0: μ =22 versus HA: μ ≠ 22. From this same data, the following confidence intervals for µ were also calculated: 90% confidence interval for µ: (22.15, 25.45) 95% confidence interval for µ: (21.84, 25.76) 99% confidence interval for µ: (21.22, 26.38) What can we conclude about the P‐value for testing the hypotheses H0: μ =22 vs. HA: μ ≠22? a) P‐value > 0.10 b) 0.05 < p‐value < 0.10 c) 0.01 < p‐value < 0.05 d) p‐value < 0.01 e) We need a test statistic to determine the p‐value in this case. 6) The physician‐recommended dosage of a new medication is 14 mg. Actual administered doses vary slightly from dose to dose and are normally distributed with mea n μ. A representative of a medical review board wishes to see if there is any evidence that the mean dosage is more than recommended and so intends to test the following hypotheses :  14 against :  14. To do this, the physician selects 16 doses at random and determines the weight of each. The physician finds the sample mean to be  14.12 mg and the sample standard deviation to be s = 0.24 mg. Based on these data, a) we would reject H0 at significance level 0.10, but not at 0.05. b) we would reject H0 at significance level 0.05, but not at 0.025 c) we would reject H0 at significance level 0.025, but not at 0.01 d) we would reject H0 at significance level 0.01 e) we would fail to reject H0 at significance level 0.10 7) For a regression model, the standard error involved in estimating the population mean response of y at a particular value of x is _______ the standard error involved in predicting an individual value of y at that same particular value of x. a) equal to b) greater than or equal to c) less than or equal to d) greater than e) less than 3Sixteen patients with high blood pressure are randomly assigned to one of two groups. In the first group, just before their blood pressure was taken, doctors asked the patients questions about their medical history and sources of stress in their lives. In the second group, just before their blood pressure was taken patients were asked to count to 400. Scientists want to test the theory the average blood pressure of patients who’ve just talked to a doctor is different from the average blood pressure of patients who’ve just counted (but not talked to a doctor. Assume all conditions are met. Use this scenario and the output given below to answer the next 2 questions. Level Number Mean Std Dev Counting 8 103.9 3.8 Talking 8 106.7 3.9 t Test Talking‐Counting Assuming unequal variances Difference 2.85 t Ratio 1.506309Std Err Dif 1.89 DF 13.95487Upper CL Dif 6.90 Prob > |t| 0.1543Lower CL Dif ‐1.20 Prob > t 0.0771Confidence 0.95 Prob < t 0.9229 Matched Pairs Difference: Talking‐Counting Talking 106.725 t‐Ratio 2.799586Counting 104.125 DF 7Mean Difference 2.6 Prob > |t| 0.0265Std Error 0.92871 Prob > t 0.0133Upper 95% 4.79605 Prob < t 0.9867Lower 95% 0.40395 N 8 Correlation 0.73789 8) What is the correct P‐value to use to test the claim of the researchers? a) 0.1543 b) 0.0771 c) 0.9229 d) 0.0265 e) 0.0133 9) What is the conclusion when the significance level is α = 0.01? a) The data doe s not provide statistical evidence, at the .01 significance level, that the average blood pressure after talking to a doctor is different from the average BP after counting to 400. b) The data provides evidence at the .01 significance level that the average blood pressure after people talk to a doctor is higher than