AP Statistics Exam Review Topic I: Describing Data FREE RESPONSE Use the given data set of test grades from a college statistics class for this question. 85 72 64 65 98 78 75 76 82 80 61 92 72 58 65 74 92 85 74 76 77 77 62 68 68 54 62 76 73 85 88 91 99 82 80 74 76 77 70 60 A. Construct two different graphs of these data B. Calculate the five-number summary and the mean and standard deviation of the data. C. Describe the distribution of the data, citing both the plots and the summary statistics found in questions 1 and 2. AP Statistics Exam Review Topic II: Normal Distribution FREE RESPONSE A set of 2,000 measurements had a symmetric, mound-shaped distribution. The mean is 5.3 and the standard deviation is 0.7. Determine an interval that contains approximately 1,360 date values. AP Statistics Exam Review Topic III: Bivariate Data FREE RESPONSE A survey was conducted recently in ten large American cities to determine whether there is any relationship between the average weekday hotel rates and average car rental rates. The following data was collected. Daily Hotel Rate (x) (in dollars) 149 187 171 122 115 147 128 212 168 181 A. B. C. D. E. Daily Car Rental Rate (y) (in dollars) 49 50 52 49 39 44 37 63 46 51 Construct a scatterplot for this data Use the scatterplot to determine if there is a linear relationship between the two variables If there is a linear relationship, numerically describe the strength of this relationship and construct a least squares regression model Find the residual associated with the point (168, 46). What percent of the variation in the car rental rates is explained by the regression of y on x? AP Statistics Exam Review Topic IV: Planning a Study FREE RESPONSE The loss of bone mass density (BMD) in men and women can be reduced by drug treatments. A pharmaceutical company has developed a new calcium supplement, which claims to reduce the loss of BMD. Volunteers who have been diagnosed with loss of bone mass density and who are currently not on medication or calcium supplements will be recruited to participate in a study. It is known that men and women experience different average losses of BMD. A. B. C. Explain how you would carry out a completely randomized experiment for the study. Describe an experimental design that would improve the design in part (A) by incorporating blocking. Can the experimental design in part (B) be carried out in a double-blind manner? Explain. AP Statistics Exam Review Topic V: Probability A. 12 B. 16 C. 24 D. 48 E. 144 A. 0.40 B. 0.42 C. 0.46 D. 0.05 E. 0.52 A. 1/8 B. ¼ C. 1/2 D. 2/3 E. 1/16 FREE RESPONSE A telecommunications company uses three different systems to produce the print shaft holder for its most popular fax machine. System A produces 50% of the shaft holders, System B produces 30% of the shaft holders, and System C produces 20% of the shaft holders. The percentages of the defective shaft holders produced by each system are respectively, 3%, 4% and 5%. A. B. If a shaft holder is selected at random, what is the probability that it is defective? If a shaft holder is selected at random and found to be defective, what is the probability that it was produced by System A? AP Statistics Exam Review Topic VI: Binomial Situations and Sampling Distributions A. 0.029 B. 0.020 C. 0.041 D. 0.032 E. 0.023 A. 0.9123 B. 0.9029 C. 0.9332 D. 0.0233 E. 0.1093 A. 6 B. 30 C. 20 D. 120 E. 60 FREE RESPONSE A breakfast cereal manufacturer has an incentive program for people to buy its product: in each box of cereal, one picture of a baseball superstar is included. Any one of five different pictures can be enclosed in the boxes and the total number of pictures of each superstar is the same. You are interested only in finding a picture of Babe Ruth. In each case, show the formula you use, the substitution into it and the answer. A. B. Find the probability that you get your first picture of Babe Ruth in the 2 nd box that you buy. Find the probability that you get your first picture of Babe Ruth in the 10 th box that you buy. AP Statistics Exam Review Topic VII: Confidence Intervals and Significance Testing FREE RESPONSE I. An educational group claims that teaching fraction concepts using math manipulatives results in higher student achievement and understanding of fractions than teaching fractions without the use of any math manipulatives. A teacher in a middle school taught a unit on fractions to two sixth grade classes, one using math manipulatives and the other without the use of any manipulatives. The table below shows the performance of these two classes on a unit test on fractions. With Manipulatives 85 75 83 87 80 79 88 94 87 82 Without Manipulatives 78 84 81 78 76 83 79 75 85 81 Test the claim that students who use manipulatives show higher achievement on a test of fractions. Give appropriate statistical evidence to support your answer. II. A county legislator is interested in polling her constituents to estimate the difference between the positions of men and women regarding a proposed bill to restrict cell phone use while driving. Her administrative assistants draw two samples, one consisting of 500 men and the other consisting of 500 women. The survey indicates that 230 men and 194 women favor legislation that would restrict the use of cell phones while driving. A. Construct a 95% confidence interval to estimate the true difference between the proportions of men and women who favor legislation that restricts cell phone use while driving. Explain how you arrived at your solution and explain your reasons for selecting the type of confidence interval that you chose. B. Write one or two statements to a non-statistician explaining what is meant by the 95% confidence interval found in part (A) C. How does a 99% confidence interval for the same data compare to the 95% confidence interval? Interpret your response in the context of this problem. AP Statistics Exam Review Topic VIII: Other Models of Inference FREE RESPONSE A medical research team is conducting a study to determine whether there is a relationships between aerobic walking and cholesterol levels. A random sample of 315 subjects is selected and represented in the table below. Test the claim that aerobic walking and cholesterol levels are related. Include appropriate statistical evidence to support your findings. Walkers Non-Walkers Low 51 23 Average 86 94 Elevated 31 30 SOLUTIONS Topic I: Describing Data A. B. C. It appears from both the graphical displays and the summary statistics indicated that the data are symmetric and unimodal. There is a slight skew to the right Topic II: Normal Distribution 1,360/ 2,000 is 68% of the data. Therefore, using the empirical rule, an interval within one standard deviation of the mean should have approximately 68% of the data: an approximate interval is (5.3 - 0.7, 5.3 + 0.7) = (4.6, 6.0) Topic III: Bivariate Data A. x= daily hotel rate in dollars y= daily car rental rate in dollars B. There is a moderate to strong positive linear relationship between daily hotel rates in dollars and the daily car rental rates in dollars. C. D. E. The correlation coefficient is r= 0.82. The least squares regression model is y-hat= 17.75 + 0.1914x. The residual associated with the point (168, 46) can be found by the finding the predicted value of y by substituting x= 168 into the regression model. y-hat= 17.75 + 0.1914 (168) y-hat= 49.9 Residual = observed minus the predicted Residual = 46- 49.9 Residual= -3.9 r 2 0.675 Topic IV: Planning a Study A. Throw the names of the volunteers in a hat or use some other random assignment method such as a random digit table and randomly assign volunteers to two different groups, a control group and a treatment group. The treatment group would receive the supplement. The control group would receive another type of supplement already in use. Compare the results of the two groups at the end. B. Identify group A as the new medicine group and group B as the control group. Since men and women experience different levels of BMD loss, separate men and women first. Then randomly assign members of each gender into Group A or Group B. Compare the results within each gender. Group A Group B Men Volunteers Group A Women Group B C. Compare Results Compare Results Yes, as long as the subjects do not know whether they are receiving the treatment or the placebo and the researcher administering and monitoring the experiment also does not know which one the subjects are receiving. Then the experiment would be double blind. Topic V: Probability 3% System A 97 % Defective Not Defective 50% 4% 30% Defective System B 96 % Not Defective 20% System C 5% 95 % A. Defective Not Defective P(Defective) = 0.5(0.03) 0.3(0.04) 0.2(0.05) 0.037 B. P(System A given that it is defective) = Topic VI: Binomial Situations and Sampling Distributions A. This is a geometric distribution. P (X n ) p (1 p )n 1 where the first success occurs on the nth trial. P (x 2) (0.2)(1 0.2)21 0.16 B. .05(0.03) 0.405 0.037 P (x 10) (0.2)(1 0.2)101 0.0268 Topic VII: Confidence Intervals and Significance Testing 1. The test required for this problem is a two sample difference of means t-test since the groups are independent and since σ is unknown. The test requires the two samples to be independently selected simple random samples and for the two populations to be normally distributed. It is not verified how these classes are selected, so we will assume that we have 2 independent SRSs. Checking for normality we will use normal quantile plots: 2. A. Because we have two independent random samples from two distinct populations (men and women) and we are looking to see the difference between the proportion of men and women who favor legislation which restrict cell phone use while driving, the choice of interval would be a two proportion confidence interval: pˆm pˆw z * pˆm (1 pˆm ) pˆw (1 pˆw ) nm nw Assumptions include two independent random samples from large populations. We will assume that there are at least 5000 men and 5000 women in the country. Our counts of successes and failures all have to be large ˆm enough. ( nm p 5,nm (1 pˆm ) 5,nw pˆw 5,nw 1 pˆw 5 ) All of the conditions are satisfied. 230 230 194 194 (1 ) (1 ) 230 194 500 500 500 500 1.96 500 500 500 500 (0.0109, 0.1331) B. We are 95% confident that the difference in population proportions is between 0.0109 and 0.1331. Because the interval does not contain 0, we are confident tht the men have a stronger preference for legislation that would restrict the use of cell phones while driving. C. A 99% confidence interval would change the value of z* in the formula above from 1.96 to 2.576 thus making the interval wider. The new interval would become (-0.0083, 0.15229). This wider interval would contain 0 and the conclusion reached concerning whether there is a difference in the proportion of men and women who favored legislation would be different in part A. i.e. there would appear to be no statistically significance difference between the two groups. Topic VIII: Other Models of Inference