You are on page 1of 52

uestion 1: What is the value m in the table below, if the mean and the variance of the random variable

X are μx = 25 and σ2x = 80?
X 10 20 30 35 m Probability .1 .5 .1 .2 .1

m = 30 m = 35 m = 38 m = 40 m = 45 Question 2: Two distributions D1 and D2 are displayed on the same graph. If the distribution D1 is right skewed, the distribution D2 is left skewed, and the mean of D1 is lower than the mean of D2, which of the following statements is not true? The median of D1 is lower than the median of D2 The mean of D1 is lower than the median of D2. The mean of D2 is lower than the median of D1. The median of D1 is lower than the mean of D2. The mean of D2 is lower than the median of D2. Question 3: A large sample of size n is used to estimate the confidence interval for a proportion p. After further evaluation, the standard deviation of the sample distribution is considered too large. What size sample do we need to use for a new standard error equal to one tenth of the original standard error? 10n 50n 100n 141n 200n Question 4: A box contains 3 yellow, 2 red, 4 green and 3 black marbles. Two marbles are taken one after the other at random from the box. What is the probability that both marbles are red?

1/50 1/60 1/66 1/24 1/18 Question 5: A large automobile manufacturer states that approximately 4% of all their cars made in 2010 have a defective component. Two samples of 150 cars (sample A) and 250 cars (sample B) were taken to test for the defective component. Which of the following statements must be true: mean of sample A < mean of sample B mean of the sample A > mean of sample B standard deviation of sample A = standard deviation of sample B standard deviation of sample A < standard deviation of sample B standard deviation of sample A > standard deviation of sample B

Question 6: You roll a die until you get a 5 or a 6. What is the variance of this distribution? 6 3 1/3 8 12 Question 7: A back to back stem and leaf plot compares the heights of the players of two basketball teams. All heights in the plot below are in inches.
Team A Team B 6 89 1444589 7 2366689

1124

8 256

The means of the two distributions are the same. The medians of the two distributions are the same. The ranges of the two distributions are the same. The distributions have the same number of observations. None of the statements above is correct. Question 8: A car manufacturer wishes to estimate the difference in early failures between cars sold in a warm climate and cars sold in a cold climate. They take a random sample of 500 cars from each group. The means obtained are 4.1 failures for cars running in a warm climate and 3.8 failures for cars running in a cold climate. The standard deviation for both populations is 0.4 failures. Find the 95% confidence interval for the difference in the population means. 0.3 +/- 0.0333 0.3 +/- 0.0350 0.3 +/- 0.0400 0.3 +/- 0.0452 0.3 +/- 0.0496 Question 9: A nutritional consulting company is trying to find what percentage of the population of a town is overweight. The marketing department of the company contacts by telephone 600 people from a list of the entire town's population. All 100 people give answers to the survey. Which of the following is the most significant source of bias in this survey? Size of sample. Undercoverage. Voluntary response bias Nonresponse. Response bias. Question 10: Which of the following are true statements: I. All bell-shaped distributions are symmetric.

II. Bar charts are useful to describe quantitative data. III. Cumulative frequency plots are useful to describe quantitative data. I only. I and II only. II and III only. I and III only. I, II and III. Question 11: A random sample of 400 passengers of an airline is polled after their flights. Of the passengers, 300 say they will fly again with the same airline. Which of the following is a 90% confidence interval for the proportion of passengers that will fly again with the same airline? 0.75 +/- 0.066 0.75 +/- 0.005 0.75 +/- 0.045 0.75 +/- 0.036 0.75 +/- 0.15 Question 12: The mean number of points per game scored by basketball players during a high school championship is 9.4, and the standard deviation is 1.5. Assuming that the number of points are normally distributed, what number of points per game will place a player in the top 15% players taking part in the basketball championship? 9.10 points per game 10.57 points per game 10.95 points per game 12.35 points per game 13.96 points per game Question 13: A residual plot: displays residuals of the response variable versus the independent variable. displays residuals of the independent variable versus the response variable displays residuals of the independent variable versus residuals of the response variable.

displays the independent variable versus the response variable. displays the response variable versus the dependent variable. Question 14: For any A and B random variables, which of the following statements must be true: I: μA+B = μA + μB II σ2A-B = σ2A + σ2B III σ2A+B = σ2A - σ2B I only. II only. I and II. II and II. I and II and II. Question 15: A random sample of 1000 balances in the retirement accounts of exempt employees of a company has a sample mean of μ1 = $100,000 and a standard deviation σ1 = $12,000. A random sample of 4000 balances in the retirement accounts of hourly employees of the same company has a sample mean of μ2 = $80,000 and a standard deviation σ2 = $14,000. If X is the sampling distribution of the differences in account balances of the two categories of employees, what is σx? $439.3 $2657.6 $5490.9 $11,065 $13,000 Question 16: Four children are asked to pick their favorite ice cream flavor out of 8 different flavors, and each of them is equally likely to pick any of the eight ice cream flavors. What is the probability that each child orders a different ice cream type? 5/72 2/5 7/64 105/256

.000 employees. 1.000 employees and corporation C has 125. circuit B creates an error with a mean of 80mV and a standard deviation of 10mV and circuit C creates an error with a mean of 50mV and a standard deviation of 10mV.972 33 +/.422 Find the 96% confidence interval for the slope of the least squares regression line.45 .645 Question 18: An electronics company designs and manufactures DC voltage power supplies. of coefficient Constant 33 Rainfall 1. Each survey will be conducted with a simple random sample of 500 employees. corporation B has 60. Corporation A has 40.972 1. B and C..45 +/.8 .1. Variable Coefficient s. The output voltages of the power supplies have accuracy errors that are caused by three independent internal circuits: A.1 33 +/.5mV 32.. No irrigation was used during this time.45/128 Question 17: The rainfall of a county is measured for 14 years in a row to ascertain the local corn crop.45 +/.5mV 40mV 55mV Question 19: An opinion survey will be conducted at three corporations.056 1. Which corporation will have its employees .45 +/. Past measurements have shown that circuit A creates an error with a mean of 100mV and a standard deviation of 20mV. The results of each survey will be used to estimate the opinions of employees at each corporation.000 employees.e.. What is the standard deviation of the error of the power supplies caused by all three circuits? 12mV 24.

Neither corporation will have a more accurate estimate.5Ω and the standard deviation was .5+/-. What is the standard deviation of the heights given that the probability of a height above 180cm is . Corporation C.9cm 8.7cm 9.5+/-.155 201. Corporation B.5cm 11.133 201.5+/-.093 201. The following table gives the two-way classification of their destination choices.125 201.5+/-. Corporations A and B.9cm 10.212 Question 21: The heights of 100 students are normally distributed with a mean of 172cm. Find the 98% confidence interval for this problem.5+/-.4Ω.9cm Question 22: Six hundreds travelers have purchased airline tickets through the same travel agency.opinions estimated more accurately by the surveys? Corporation A. Question 20: The quality department of an electronics manufacturer randomly selected 100 resistors.25? 7. What is the joint relative conditional frequency for female travelers to Asia if the marginal row totals are fixed? Male Female Totals Europe Asia 189 195 49 62 384 111 . 201. The mean resistance of the resistors was 201.

which of the following statements is correct: The sampling distribution is small relative to the population.3. what is the mean weight of the women? a) 120lb b) 125lb c) 130lb d) 132lb e) 135lb Question 25: A real estate agent finds home buyers and closes the sales for 70% of his clients that sell thier houses. 30% wear glasses. The mean of the sampling distribution is equal to . The sampling distribution is a skewed distribution.45 mean = 7 and standard deviation = 2. None of the above. What is the mean number of sales for his next 10 clients and what is the standard deviation of this distribution? mean = 7 and standard deviation = 1. Question 24: The mean of the weights of a group of 100 men and women is 160lb.South America 55 Totals 50 105 600 293 307 a) 53% b) 56% c) 58% d) 61% e) 63% Question 23: Out of the 500 students of a school.5 . If the number of men in the group is 60 and the mean weight of the men is 180lb. The sampling distribution is normal. If we use a simple random sample of 25 students.

as it is shown in the distribution below.5 mean = 7 and standard deviation = 1.45 mean = 10 and standard deviation = 2. The boxplot displays time in minutes.33 Question 26: The height and the weight of 18 students were measured and a scatterplot of the measures is shown below.mean = 10 and standard deviation = 1. If two pairs of measurements need to be removed from the set of 18. Which of the following statements must be true? . which of the choices shown below decreases the coefficient of correlation the most? a) S2 and S3 b) S2 and S5 c) S1 and S3 d) S2 and S4 e) S1 and S2 Question 27: It takes different times for different workers to perform the same specific task.

46 .5. apples and oranges. Which of the following is the probability that the fruit is an apple? .5 . b) The distribution is skewed to the left.32 . Ha: μ < 8. Ha: μ < 8.5 H0: μ = 8.5 H0: μ ≠ 8. two other baskets contain three apples and one peach each. Three of the baskets contain two apples and one orange each.5 H0: μ = 8. d) distribution median < distribution mean e) distribution median = distribution mean Question 28: Six fruit baskets contain peaches.5.a) The 25th percentile is greater than 70 minutes. c) The interquartile range is higher than 20 minutes. If we want to conduct an experiment to determine if the patients are losing less weight than advertised.5. Ha: μ ≠ 8.5lb after one month of treatment.58 Question 29: A pharmaceutical company claims that its weight loss drug allows women to lose 8. Ha: μ > 8. which of the following hypotheses should be used? H0: μ = 8.5. and the last basket contains two peaches and two oranges. You select a basket at random and then select a fruit at random from the basket.4 .5 .

151 z = -1. Which of the following is an appropriate test outcome? z = 0.214 z = -1. with a standard deviation of 28 minutes.034 p = .214 z = -0.214 z = -0. Ha: μ > 8.151 Question 31: A tutoring company tests the results of their intensive training for a standardized test.274 p = .274 p = . A random sample of 40 trains was taken and the average time required to travel from New York to Buffalo was 188 minutes.000 miles.552 p = .000 miles.0099 . What is the p-value for this test? .5 Question 30: A car manufacturer claims that 90% of their cars do not experience engine failures before reaching a mileage of 150.034 p = .0355 . A sample of 65 cars is investigated and 9 of the cars had an engine failure before 150.H0: μ ≠ 8. Six students randomly selected have taken the test before and after having been trained by the company. The following table gives the test scores of the 6 students: Before 88 79 77 83 69 After 78 54 83 73 71 The appropriate test for this situation is: a matched pair t-test a chi-square goodness of fit test a two-sample z-test a one-sample t-test a one-sample z-test Question 32: The Department of Transportation of the State of New York claimed that it takes an average of 200 minutes to travel by train from New York to Buffalo.5.

I only $439.0496 Response bias.036 10.1167 m= 40 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 the mean of D2 is lower than the median of D1.3 105/256 1..3 +/.2881 . I and III only.75 +/..972 Not answered Not answered Not answered Not answered Not answered Not answered Not answered Not answered Not answered Not answered Not answered Not answered Not answered Not answered Not answered Not answered Not answered .0.1294 .0.45 +/. 100n 1/66 standard deviation of sample A > standard deviation of sample B 6 None of the statements above is correct.95 points per game displays residuals of the response variable versus the independent variable. 0. 0.

Categorical variables are the same as quantitative variables.151 a matched pair t-test . Quantitative variables can be continuous variables. .0099 Not answered Not answered Not answered Not answered Not answered Not answered Not answered Not answered Not answered Not answered Not answered Not answered Not answered Not answered roblem 1 Which of the following statements are true? (Check one) I.093 11. .45 S2 and S4 The distribution is skewed to the left. Categorical variables are the same as qualitative variables. 130lb mean = 7 and standard deviation = 1. 201. II.58 H0: mu = 8.18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 24. Ha: mu < 8.5mV Neither corporation will have a more accurate estimate.5 z = -1.5+/-. III.5.034 p = .9cm 56% The sampling distribution is a skewed distribution.

And quantitative variables are numeric variables. THH. are not quantitative variables. They are: HHH.125 (B) 0. Categorical variables are just another name for qualitative variables. They are: HTT. HTH. three have exactly one head. HHT.(A) I only (B) II only (C) III only (D) I and II (E) I and III Solution The correct answer is (E). there are a total of eight possible outcomes. Categorical variables. TTH.333 (D) 0. See also: Probability Problem 3 . so they can be continuous variables. What is the probability that it lands on heads exactly one time? (A) 0.500 Solution The correct answer is (D). and TTT.375 (E) 0. Of the eight possible outcomes.250 (C) 0. THT. HTT. Therefore. If you toss a coin three times.375. THT. See also: Variables Problem 2 A coin is tossed three times. and TTH. however. the probability that three flips of a coin will produce exactly one head is 3/8 or 0.

there was a 100 percent chance that the sample would include 100 purchasers of each brand of car. Similarly. for example. The sampling method in this problem used random sampling and gave each buyer an equal chance of being selected. The fact that each buyer in the sample was randomly sampled is a necessary condition for a simple random sample. so this cannot be a simple random sample. A simple random sample requires that every sample of size n (in this problem. n is equal to 400) have an equal chance of being selected. because each buyer in the sample was randomly sampled. the fact that population consisted of buyers of different car brands is irrelevant. and 2. The analyst selects a sample of 400 car buyers. 101 Honda buyers. 99 Ford buyers. (C) Yes. and 100 GM buyers.000 new car buyers. See also: Survey Sampling Methods Problem 4 Which of the following statements is true? . (D) No. (B) Yes.500 Honda buyers. but it is not sufficient. by randomly sampling 100 buyers of each brand. 2. but the sampling method was actually stratified random sampling. but it is not sufficient.500 Ford buyers. (E) No. There was zero percent chance that the sample would include. Is this an example of a simple random sample? (A) Yes. The fact that car buyers of every brand were equally represented in the sample is irrelevant to whether the sampling method was simple random sampling. because each buyer in the sample had an equal chance of being sampled. 100 Toyota buyers. Solution The correct answer is (D). sampling from a list of 10. Thus. In this problem.500 GM buyers. because every possible 400-buyer sample did not have an equal chance of being chosen.An auto analyst is conducting a satisfaction survey. because the population consisted of purchasers of four different brands of car. The list includes 2.500 Toyota buyers. 2. the fact that each buyer in the sample had an equal chance of being selected is characteristic of a simple random sample. because car buyers of every brand were equally represented in the sample. all possible samples of size 400 did not have an equal chance of being selected. Similarly.

I. The confidence interval is a type of point estimate. The bigger the margin of error. II. a sample mean is an example of a point estimate. 7}.67 (E) None of the above Solution The correct answer is (B). the smaller the confidence interval. First. (A) I only (B) II only (C) III only (D) IV only (E) None of the above. A population mean is not an example of a point estimate. so the confidence interval gets bigger as the margin of error gets bigger. we need to compute the sample mean. x=(1+3+5+7)/4=4 . 3. A population mean is an example of a point estimate. IV.58 (C) 6 (D) 6. The confidence interval is equal to the sample statistic plus or minus the margin of error. 5. Solution The correct answer is (E). not a type of point estimate. not a population parameter. III. The center of a confidence interval is a population parameter. What is the standard deviation? (A) 2 (B) 2. The center of a confidence interval is a sample statistic. See also: Estimation Problems Problem 5 A sample consists of four observations: {1. A confidence interval is a type of interval estimate.

4 )2 + ( 7 . We know the following:     There are 52 cards in the deck.1 ) ] s = sqrt { [ ( 1 .1 ) } s = sqrt { [ ( -3 )2 + ( -1 )2 + ( 1 )2 + ( 3 )2 ] / 3 } s = sqrt { [ 9 + 1 + 1 + 9 ] / 3 } = sqrt (20 / 3) = sqrt ( 6.P(S ∩ A) P(S ∪ A) = 13/52 + 4/52 . Solution The correct answer is C. so P(S) = 13/52. Therefore.58 See also: Measures of Variability Problem 6 A card is drawn randomly from a deck of ordinary playing cards.67 ) = 2. and let A = the event that the card is an ace. we plug all of the known values into the formula for the standard deviation of a sample. as shown below: s = sqrt [ Σ ( xi .4 )2 ] / ( 4 . so P(A) = 4/52. Let S = the event that the card is a spade. There is 1 ace that is also a spade.Then. so P(S ∩ A) = 1/52. What is the probability that you will win the game? (A) 1/13 (B) 13/52 (C) 4/13 (D) 17/52 (E) None of the above.1/52 = 16/52 = 4/13 See also: Rules of Probability .4 )2 + ( 3 . There are 4 aces. There are 13 spades.4 )2 + ( 5 . You win $10 if the card is a spade or an ace. based on the rule of addition: P(S ∪ A) = P(S) + P(A) .x )2 / ( n .

II. See also: Standard Error Problem 8 Nine hundred (900) high school freshmen were randomly selected for a national survey. the mean grade-point average (GPA) was 2. The standard error can be computed from a knowledge of sample attributes . Solution .4. I.sample size and sample statistics. What is the margin of error. The standard deviation is computed solely from sample attributes.Problem 7 Which of the following statements is true. The standard error is a measure of central tendency.025 (C) 0. The standard error is computed solely from sample attributes. assuming a 95% confidence level? (A) 0.013 (B) 0. The standard error is a measure of variability. it requires a knowledge of one or more population parameters. III.7. and the standard deviation was 0.500 (D) 1. The standard deviation cannot be computed solely from sample attributes. not a measure of central tendency. (A) I only (B) II only (C) III only (D) I and II (E) I and III Solution The correct answer is (A).960 (E) None of the above. Among survey participants.

or the standard normal distribution table.g. To compute the margin of error. For this problem. it will be the z score having a cumulative probability equal to 0.975 Find the critical z score.. The test has a mean score of 100 and a standard deviation of 15. Using the Normal Distribution Calculator. we find the cumulative probability associated with the z-score. we find the standard error of the mean.0. we need to find the critical value and the standard error of the mean. the sampling distribution will be roughly normal in shape. ME = Critical value x Standard error = 1. a handheld graphing calculator. Since the sample size is large. (On the actual AP Statistics exam. we can express the critical value as a z score.05 Find the critical probability (p*): p* = 1 .) Next. we find that the critical value is 1.96 * 0.05/2 = 0.0. Stat Trek's free normal distribution calculator).4 / 30 = 0.96. using the following equation: SEx = s / sqrt( n ) = 0.    Compute alpha (α): α = 1 . using an online calculator (e.20. you may need to use a graphing calculator or a normal table.013 And finally.α/2 = 1 .025 See also: Margin of Error Problem 9 A national achievement test is administered annually to 3rd graders. To find the critical value. Therefore. If Jane's z-score is 1. we take the following steps.(confidence level / 100) = 1 .4 / sqrt( 900 ) = 0. since Stat Trek's analytical tools will not be available. what was her score on the test? (A) 82 (B) 88 (C) 100 (D) 112 (E) 118 Solution .95 = 0. we compute the margin of error (ME). Then.The correct answer is (B).013 = 0.975.

we know z = (X . From the z-score equation. The average height of a group of boys could be a non-integer. but it does not vary and it does not result from a random process.20 * 15) + 100 = 18 + 100 = 118 See also: Measures of Position Problem 10 Which of the following is a discrete random variable? I.μ) / σ where z is the z-score. The annual number of sweepstakes winners from New York City. we get X = ( z * σ) + 100 = ( 1. and σ is the standard deviation. The average height of a randomly selected group of boys. (A) I only (B) II only (C) III only (D) I and II (E) II and III Solution The correct answer is B. μ is the mean of the population. See also: Random Variables Problem 11 . The number of presidential elections in the 20th century. And the number of presidential elections in the 20th century is an integer. II. so it is a discrete random variable. so it is not a random variable. Solving for Jane's test score (X). III. The annual number of sweepstakes winners is an integer value and it results from a random process. X is the value of the element.The correct answer is (E). so it is not a discrete variable.

(E) None of the above.86 (B) 180 + 3.Which of the following statements are true? (Check one) I. Therefore.000. The best method for investigating causal relationships is an experiment . A sample survey is an example of an experimental study. See also: Data Collection Methods Problem 12 Suppose we want to estimate the average weight of an adult male in Dekalb County. III. and the standard deviation of the sample is 30 pounds. (A) 180 + 1. The best method for investigating causal relationships is an observational study. Solution The correct answer is (E). What is the 95% confidence interval. it is an observational study. the researcher does not assign treatments to survey respondents. In a sample survey. rather. manpower) than an experiment.000 men from a population of 1. (A) I only (B) II only (C) III only (D) All of the above.000 men and weigh them. We find that the average man in our sample weighs 180 pounds. . An observational study requires fewer resources than an experiment. a sample survey is not an experimental study. which makes it easier for the researcher to identify causal effects of treatment variables. money.0 (C) 180 + 5. We draw a random sample of 1.because an experiment features randomized assignment of subjects to treatment groups. Georgia. Randomization "evens out" the effects of extraneous variables.88 (D) 180 + 30 (E) None of the above. An observational study may or may not require fewer resources (time. II.not an observational study .

0. Previously.  Find the margin of error. we work through the four steps below.1 = 1000 . a handheld graphing calculator. The key steps are shown below. To specify the confidence interval. To express the critical value as a t score (t*).) Note: We might also have expressed the critical value as a z score. In this case.95 = 1.975 Find the degrees of freedom (df): df = n .  Find standard error.96.(confidence level / 100) = 0. The standard error (SE) of the mean is: SE = s / sqrt( n ) = 30 / sqrt(1000) = 30/31.  Compute margin of error (ME): ME = critical value * standard error = 1.  Identify a sample statistic. Since we are trying to estimate the mean weight in the population. The critical value is a factor used to compute the margin of error. you may need to use a graphing calculator or a t table. Stat Trek's free t Distribution Calculator).1 = 999 The critical value is the t score having 999 degrees of freedom and a cumulative probability equal to 0.05/2 = 0.Solution The answer is (A). since Stat Trek's analytical tools will not be available. o o o o Compute alpha (α): α = 1 . We are working with a 95% confidence level. (On the actual AP Statistics exam. a z score analysis produces the same result . Using an online calculator (e.975 is 1.  Select a confidence level.95  Find critical value. follow these steps.96.α/2 = 1 .86 . we find that the t score associated with a cumulative probability of 0. or a t distribution table. we described how to compute the margin of error. the confidence level is defined for us in the problem..975. Because the sample size is large.96 * 0. we choose the mean weight in our sample (180) as the sample statistic.05 Find the critical probability (p*): p* = 1 .g.62 = 0.a critical value equal to 1.

See also: Confidence Intervals Problem 13 The stemplot below shows the number of hot dogs eaten by contestants in a recent hot dog eating contest. Specify the confidence interval. The range of the confidence interval is defined by the sample statistic + margin of error. The median is 46. (A) I only (B) II only (C) III only (D) I and II (E) I. The mean is 47. And the uncertainty is denoted by the confidence level. 80 1 70 60 4 7 50 2 2 6 40 0 2 5 7 9 9 30 5 7 9 20 7 9 10 1 Which of the following statements are true? I. we can be 95% confident that the population means falls within the interval 180 + 1. Therefore.86. and III Solution . The range is 70. II. II. III.

They asked whether the paper should increase its coverage of local news.10.in the middle of the data set.50 + 0.15 4 or more ??? What is the probability that 4 or more adults reside at a randomly selected home? (A) 0.000 subscribers. The sum of all the probabilities is equal to 1. See also: Probability Distributions Problem 15 A major metropolitan newspaper selected a simple random sample of 1.50 3 0. The biggest value is 81. so the range is equal to 81 -11 or 70.45 and 47 . and the smallest value is 11.600 readers from their list of 100. Number of adults. x Probability.15) or 0. See also: Stemplots Problem 14 The number of adults living in homes on a randomly selected city block is described by the following probability distribution.25 + 0. The range is equal to the biggest value minus the smallest value. The mean is equal to the average of all the values . so the median is equal to 46.15 (C) 0. Therefore.10 (B) 0.The correct answer is (D).56. Their average is (45 + 47)/2 or 46.50 (E) There is not enough information to answer this question. P(x) 1 0.25 2 0. Forty . we have an even number of values .45. the probability that four or more adults reside in a home is equal to 1 . The median is equal to the middle value in the data set. Solution The correct answer is A. Here.(0.25 (D) 0.

What is the 99% confidence interval for the proportion of readers who would like more coverage of local news? (A) 0. Elsewhere on this site. The approach that we used to solve this problem is valid when the following conditions are met.30 to 0. Since we are trying to estimate a population proportion.35 to 0.50 (B) 0.  Find the margin of error.plenty of successes and failures. we have 0. so we will use a simple "approximate" formula for the standard error. Then. Suppose we classify a "more local news" response as a success. The confidence level is defined for us in the problem statement. Since the above requirements are satisfied. we can use an "approximate" formula for the standard deviation or the standard error.41 Solution The answer is (D).37 to 0.  The sample should include at least 10 successes and 10 failures. and 0.  Identify a sample statistic.32 to 0.60 * 1600 = 960 failures . and any other response as a failure.40 * 1600 = 640 successes.40) as the sample statistic. We are working with a 99% confidence level. we choose the sample proportion (0. .48 (C) 0. we can use the following four-step approach to construct a confidence interval.43 (E) 0.45 (D) 0.percent of the sample wanted more local news. The key steps are shown below.  If the population size is much larger than the sample size.39 to 0. This condition is satisfied. This condition is satisfied.  The sampling method must be simple random sampling. we show how to compute the margin of errorwhen the sampling distribution is approximately normal. the problem statement says that we used simple random sampling.  Select a confidence level.

p) / n ] = sqrt [ (0. you may need to use a graphing calculator or a normal distribution table.24/1600 ] = 0.995 The critical value is the z score having a cumulative probability equal to 0. (On the actual AP Statistics exam. The range of the confidence interval is defined by the sample statistic + margin of error.995 is 2.01 Find the critical probability (p*): p* = 1 .(99/100) = 0. we are 99% confident that the true population proportion is in the range defined by 0.α/2 = 1 . And since the population is more than 10 times larger than the sample. we cannot compute the standard deviation.37 to 0.4 + 0.03. a handheld graphing calculator. we can use the following formula to compute the standard error (SE) of the proportion: SE = sqrt [ p(1 .012  Find critical value. The critical value is a factor used to compute the margin of error.0. Stat Trek's free Normal Distribution Calculator).03  Specify the confidence interval. the 99% confidence interval is 0. we can express the critical value as a z score by following these steps.58 * 0. See also: Estimating a Proportion Problem 16 .. since Stat Trek's analytical tools will not be available. Since we do not know the population proportion.995. And the uncertainty is denoted by the confidence level.)  Compute margin of error (ME): ME = critical value * standard error = 2. Find standard deviation or standard error.58. we find that the z score associated with a cumulative probability of 0.6) / 1600 ] = sqrt [ 0.4)*(0. Because the sampling distribution is approximately normal and the sample size is large.012 = 0. or a normal distribution table.01/2 = 0. instead. Therefore.43.g. Using an online calculator (e. That is.(confidence level / 100) = 1 . we compute the standard error. o o o Compute alpha (α): α = 1 .

we cannot compute the standard deviation of the sample mean. the confidence level is defined for us in the problem. What is the 99% confidence interval for the students' IQ score? (A) 115 + 0. the average IQ score is 115 with a standard deviation of 10. We are working with a 99% confidence level.Suppose a simple random sample of 150 students is drawn from a population of 3000 college students. Since we are trying to estimate a population mean. Among sampled students. The key steps are shown below. Solution The correct answer is (C).  The sampling distribution should be approximately normally distributed. Elsewhere on this site. we can use the following four-step approach to construct a confidence interval.82 (C) 115 + 2. we know from the central limit theorem that the sampling distribution of the mean will be normal or nearly normal. so this condition is satisfied. Since the above requirements are satisfied. This condition is satisfied.6 (E) None of the above. the problem statement says that we used simple random sampling. we choose the sample mean (115) as the sample statistic. we show how to compute the margin of errorwhen the sampling distribution is approximately normal.  Find the margin of error.  Identify a sample statistic. The approach that we used to solve this problem is valid when the following conditions are met. In this analysis. Because the sample size is large. . Since we do not know the standard deviation of the population.  The sampling method must be simple random sampling.  Find standard deviation or standard error.  Select a confidence level.1 (D) 115 + 2.01 (B) 115 + 0.

Therefore. (On the actual AP Statistics exam.) Note: We might also have expressed the critical value as a z score.g.instead.01/2 = 0.α/2 = 1 .1. a z score analysis produces a similar result .82  Find critical value.1. And the uncertainty is denoted by the confidence level.99/100 = 0. To find the critical value.995 is 2. or a t distribution table.1 = 150 .1  Specify the confidence interval.82 = 2.61. o o o o Compute alpha (α): α = 1 .01 Find the critical probability (p*): p* = 1 . The critical value is a factor used to compute the margin of error.25 = 0.58. SE = s / sqrt( n ) = 10 / sqrt(150) = 10 / 12. Because the sample size is much smaller than the population size. we are 99% confident that the true population mean is in the range defined by 115 + 2.  Compute margin of error (ME): ME = critical value * standard error = 2.995 Find the degrees of freedom (df): df = n . since Stat Trek's analytical tools will not be available.. Using an online calculator (e.a critical value equal to 2. we find that the t score associated with a cumulative probability of 0. we express the critical value as a t score rather than a z score. a handheld graphing calculator. you may need to use a graphing calculator or a t table.0. we take these steps.61 * 0. The range of the confidence interval is defined by the sample statistic + margin of error.9 to 117. Stat Trek's free t Distribution Calculator).1 = 149 The critical value is the t score having 149 degrees of freedom and a cumulative probability equal to 0. See also: Estimating the Population Mean .995. Because the standard deviation of the population is unknown. we can use the "approximate" formula for the standard error. Because the sample size is fairly large. That is. the 99% confidence interval is 112. we compute the standard error (SE).(confidence level / 100) = 1 .

so the distribution is skewed left. Most of the observations are on the high end of the scale. The interquartile range is about 8. And the median is indicated by the vertical line running through the middle of the box. III. The median is about 10.Problem 17 Consider the boxplot below. Number of adults. which is roughly centered over 15. which is 18 minus 10 or 8. (A) I only (B) II only (C) III only (D) I and II (E) II and III Solution The correct answer is (B). 2 4 6 8 10 12 14 16 18 Which of the following statements are true? I. See also: Boxplots Problem 18 The number of adults living in homes on a randomly selected city block is described by the following probability distribution. II. The distribution is skewed right. The interquartile range is indicated by the length of the box. x 1 2 3 4 . So the median is about 15.

E(x) ]2 * P(xi) σ2 = (1 .81) * 0.2.15 + (4 .79 (D) 0.50 + 3*0.89 (E) 2.2.Probability. use a convenience sample.50) + (0. See also: Attributes of Random Variables Problem 19 Which of the following statements are true? I. then.01 * 0. Computations are shown below. σ2 = Σ [ xi . the standard deviation is equal to the square root of the variance.10 Solution The correct answer is D. P(x) 0. E(X) = Σ [ xi * P(xi) ] E(X) = 1*0.89.79 And finally.15 + 4*0.0050 + 0. find the variance. The solution has three parts.1)2 * 0.1215 + 0.50 (B) 0. II.2.10 What is the standard deviation of the probability distribution? (A) 0. beginning with the expected value.3025 + 0.10 Now that we know the expected value.62 (C) 0. we find the variance. .1)2 * 0.10) = 0. then.10 = 2.2.21 * 0.1)2 * 0. Random sampling is a good way to reduce response bias. so the standard deviation is sqrt(0.15 0.10 σ2 = (1.79) or 0. To guard against bias from undercoverage.25) + (0. find the standard deviation.3610 = 0.50 0. First.25 + (2 .1)2 * 0.25 0.61 * 0.50 + (3 . find the expected value.25 + 2*0.15) + (3.

use a mail-in survey. it sometimes causes undercoverage bias.d)2 1 2 3 4 5 6 7 8 9 95 89 76 92 91 53 67 88 75 90 85 73 90 90 53 68 90 78 5 4 3 2 1 0 -1 -2 -3 16 9 4 1 0 1 4 9 16 Student English Math Difference. Solution The correct answer is (E). Random sampling provides strong protection against bias from undercoverage bias and voluntary response bias. See also: Survey Sampling Bias Problem 20 Twenty-two students were randomly selected from a population of 1000 students. And finally. d (d . Test results are summarized below. A convenience sample does not protect against undercoverage bias. In fact.d)2 12 13 14 15 16 17 18 19 20 85 87 85 85 68 81 84 71 46 83 83 83 82 65 79 83 60 47 2 4 2 3 3 2 1 11 -1 1 9 1 4 4 1 0 100 4 . The sampling method was simple random sampling. Increasing the sample size tends to reduce survey bias. Increasing sample size does not affect survey bias. None of the statements is true.III. To guard against nonresponse bias. IV. Student English Math Difference. but it is not effective against response bias. using a mail-in survey does not prevent nonresponse bias. mail-in surveys are quite vulnerable to nonresponse bias. in fact. d (d . All of the students were given a standardized English test and a standardized math test. (A) I only (B) II only (C) III only (D) IV only (E) None of the above.

 The sampling method must be simple random sampling. In this analysis. We are working with a 90% confidence level.0 (E) 1 + 3.0 (C) 1 + 1. we use the sample mean difference (d = 1) as the sample statistic. Since we are trying to estimate a population mean difference in math and English test scores.d)2 = 270 d=1 What is the 90% confidence interval for the mean difference between student scores on the math and English tests? Assume that the mean differences are approximately normally distributed. The approach that we used to solve this problem is valid when the following conditions are met. This condition is satisfied. we can use the following four-step approach to construct a confidence interval. The problem statement says that the differences were normally distributed. the confidence level is defined for us in the problem.6 Solution The answer is (C). so this condition is satisfied.10 11 85 90 89 95 -4 -5 25 36 21 22 75 80 77 83 -2 -3 9 16 Σ(d . . The key steps are shown below. we show how to compute the margin of errorwhen the sampling distribution is approximately normal.3 (D) 1 + 2. (A) 1 + 0.  Identify a sample statistic.8 (B) 1 + 1.  Select a confidence level. Since the above requirements are satisfied. Elsewhere on this site. the problem statement says that we used simple random sampling.  The sampling distribution should be approximately normally distributed.  Find the margin of error.

10/2 = 0. the 90% confidence interval is -0. The range of the confidence interval is defined by the sample statistic + margin of error. The critical value is a factor used to compute the margin of error. Because the standard deviation of the population is unknown.586/4.95 Find the degrees of freedom (df): df = n .3.. since Stat Trek's analytical tools will not be available.69 = 0.10 Find the critical probability (p*): p* = 1 .g. we express the critical value as a t score rather than a z score.586 / [ sqrt(22) ] = 3. we compute the standard error (SE). Since the sample size is fairly small.586 SE = sd / sqrt( n ) = 3. Therefore. Find standard deviation or standard error.d)2 / (n . we are 90% confident that the true population proportion is in the range defined by 1 + 1.)  Compute margin of error (ME): ME = critical value * standard error = 1. . we can use the approximation equation for the standard error.3  Specify the confidence interval. Since the sample size is much smaller than the population size.1 = 21 The critical value is the t score having 21 degrees of freedom and a cumulative probability equal to 0.1) ] = sqrt[ 270/(22-1) ] = sqrt(12.(confidence level / 100) = 1 . we take these steps. instead. And the uncertainty is denoted by the confidence level. we cannot compute the standard deviation of the sample mean. o o o o Compute alpha (α): α = 1 . we find that the t score associated with a cumulative probability of 0.3.857) = 3. Since we do not know the standard deviation of the population.90/100 = 0.765  Find critical value. sd = sqrt [ (Σ(di .α/2 = 1 . we choose the t score.95.1 = 22 .0. That is. a handheld graphing calculator.72. you may need to use a graphing calculator or a t table.95 is 1. Using an online calculator (e. (On the actual AP Statistics exam.765 = 1.72 * 0. To find the critical value. or a t distribution table. Stat Trek's free t Distribution Calculator).3 to 2.

This line intersects the curve over the X axis at a height of about 71 inches. so Q1 is 77. To find Q3. Q1 is the height for which the cumulative percentage is 25%. Since the interquartile range is Q3 minus Q1. the cumulative frequency plot shows height (in inches) of college basketball players. follow the grid line to the right from the Y axis at 75%. See also: Cumulative Frequency Plots . This means that 25% of the basketball players are at most 71 inches tall.See also: Mean Difference Between Matched Data Pairs Problem 21 Below. To find Q1 from the cumulative frequency plot. This line intersects the curve over the X axis at a height of about 77 inches. the interquartile range is 77 . follow the grid line to the right from the Y axis at 25%. What is the interquartile range? (A) 3 inches (B) 6 inches (C) 25 inches (D) 50 inches (E) None of the above Solution The correct answer is (B). so Q1 is 71. The interquartile range is the middle range of the distribution. defined by Q3 minus Q1.71 or 6 inches. This means that 75% of the basketball players are at most 77 inches tall.

based on the information given. which is 5. Therefore. Suppose you bought a package of cards and counted the players from each group. The solution requires us to recognize that Variable Z is a combination of twoindependent random variables. and the variance of Y is equal to 9. 60% are veterans. Let Z = X . See also: Combinations of Random Variables Problem 23 Acme Toy Company sells baseball cards in packages of 100. and 10% are All-Stars.00 (C) 7. Solution The correct answer is B.Y. AllStars. the standard deviation is equal to the square root of 25. and 10%.Y) = Var(X) + Var(Y) = 16 + 9 = 25 The standard deviation of Z is equal to the square root of the variance.0 (E) It is not possible to answer this question. Var(Z) = Var(X .Problem 22 Suppose X and Y are independent random variables. Three types of players are represented in each package -. the variance of Z is equal to the variance of X plus the variance of Y. veterans. 60%.00 (D) 25. and All-Stars. What is the standard deviation of Z? (A) 2. The variance of X is equal to 16. veterans. .65 (B) 5. The company claims that 30% of the cards are rookies.rookies. What method would you use to test Acme's claim that 30% of the production run are rookies. As such. Cards from each group are randomly assigned to packages.

The t-tests are not appropriate. The probability of making a Type II error. (A) I only (B) II only (C) III only (D) All of the above (E) None of the above . Since we only have one sample in this baseball card problem. In this problem. The other chi-square options (the independence test and the homogeneity test) involve comparing data from two samples. The effect size of the hypothesis test. since they are used with quantitative data. The power of the hypothesis test. If she doubles her sample size. III. See also: Chi-Square Goodness of Fit Test Problem 24 Suppose a researcher conducts an experiment to test a hypothesis. II. and this problem involves categorical data.the type of baseball card. we are dealing with a categorical variable -. which of the following will increase? I. And we want to determine whether the distribution in our package is consistent with the production distribution claimed by Acme Toy Company. they are not appropriate.(A) Chi-square goodness of fit test (B) Chi-square test for homogeneity (C) Chi-square test for independence (D) One-sample t test (E) Matched pairs t-test Solution The answer is (A). The chi-square goodness of fit test is used to find out whether an observed pattern of categorical data is consistent with a specified distribution. So the chisquare goodness of fit test is appropropriate.

more likely to reject the null hypothesis when it is. Thus. See also: Power of a Hypothesis Test Problem 25 College High school 7 366 1234 6889 28 3 0 1 2 3 4 5 6 7 0035 12446 189 01 . not bigger. as sample size increases. Increasing sample size makes the hypothesis test more sensitive . in fact. false. And the probability of making a Type II error gets smaller. The effect size is not affected by sample size.Solution The answer is (A). it increases the power of the test.

See also: Comparing Distributions Problem 26 Suppose a die is tossed 5 times.028 (B) 0. II. (A) I only (B) II only (C) I and III only (D) II and III only (E) I. The mean is greater than the median in both groups. II. so the mean is greater than the median in both groups. the median is equal to 24. III. Solution .4 for college students. the fewest books read by a college student was seven. and III Solution The correct answer is (E). All of the college students read books during the year.333 (E) There is not enough information to answer this question. One college student read seven books.167 (D) 0.The back-to-back stemplot on the right shows the number of books read in a year by a random sample of college and high school students. Which of the following statements are true? I. In both groups.161 (C) 0. The college median is equal to the high school median.3 for high school students and 30. What is the probability of getting exactly 2 fours? (A) 0. And the mean number of books read per year is 25.

Therefore.P)n . (A) I only (B) II only (C) III only (D) All of the above. Solution The correct answer is (B). II. and the probability of success on a single trial is 1/6 or about 0. which of the following statements is always true? . Blinding controls for the effects of confounding.The answer is (B). III. By randomly assigning subjects to treatment levels. 5. each factor has two or more treatment levels. the number of successes is equal to 2. 0. 0.833)3 = 10 * (0. See also: Experiments Problem 28 In hypothesis testing. which of the following statements are true? I. the binomial probability is: b(2.161 See also: Binomial Distribution Problem 27 With respect to experimental design. but it does not guard against confounding. randomization spreads potential effects of lurking variables roughly evenly across treatment levels. (E) None of the above.167.167) = nCx * Px * (1 . 5.167)2 * (0. that factor would be confounded with every other factor in the experiment. each subject in the experiment would get the same treatment on that factor.167)2 * (0.833)3 b(2. Each experimental factor has one treatment level.x = 5C2 * (0. If a factor had only one treatment level. Randomization controls for effects of lurking variables. Blinding ensures that subjects in control and treatment conditions experience the placebo effect equally. As a result. And finally. This is a binomial experiment in which the number of trials is equal to 5.167) = 0.

but it can also be smaller than the significance level. The correlation between car weight and annual maintenance cost is 0. Heavier cars tend to be less reliable.20. The P-value is the probability of observing a sample statistic as extreme as the test statistic. V. III. See also: How to Test Hypotheses Problem 29 A national consumer magazine reported the following correlations. IV. It can be greater than the significance level. The P-value is a probability. The P-value is greater than the significance level. It is not computed from the significance level.30. II. and it is not a test statistic. The P-value is the parameter in the null hypothesis. Which of the following statements are true? I. The P-value is computed from the significance level. Heavier cars tend to cost more to maintain. III.   The correlation between car weight and car reliability is -0.I. (A) I only (B) II only (C) III only (D) IV only (E) V only Solution The answer is (E). The P-value is a test statistic. (A) I only (B) II only (C) III only . it is not the parameter in the null hypothesis. II. Car weight is related more strongly to reliability than to maintenance cost.

See also: Correlation Problem 30 Bob is a high school basketball player. this problem can be solved using the negative binomial formula or the geometric formula. The correlation between car weight and maintenance cost is positive. beginning with the negative binomial formula.1681 Solution The answer is (B). Therefore. which is a special case of a negative binomial distribution.30. This means that reliability tends to decrease as car weight increases. This means that maintenance costs tend to increase as car weight increases. He is a 70% free throw shooter.0720 (E) 0. Therefore. The strength of a relationship between two variables is indicated by the absolute value of the correlation coefficient.(D) I and II (E) I. and III Solution The correct answer is (E).0024 (B) 0. The correlation between car weight and reliability has an absolute value of 0. What is the probability that Bob makes his first free throw on his fifth shot? (A) 0.0081 (D) 0.0057 (C) 0.20.70. This is an example of a geometric distribution. II. The correlation between car weight and reliability is negative. . That means his probability of making a free throw is 0. the relationship between car weight and reliability is stronger than the relationship between car weight and maintenance cost. The correlation between car weight and maintenance cost has an absolute value of 0. We demonstrate each approach below.

r. 0. The solution to this problem involves finding the P-value.. 0.7 * 0.02 (C) 0.r b*(5.25.04 (D) 0. take the following steps.e.70. the probability of making 60 or fewer bulls-eyes. P) = x-1Cr-1 * Pr * Qx .00567 Now. .00567 Notice that each approach yields the same answer. 3.7) = 4C0 * 0. the number of trials (x) is 5. the null hypothesis is P = 0.7) = 0. If the 300 shots result in 60 or fewer bulls-eyes (i.08 (E) 0.71 * 0. In other words.1 g(5. P) = P * Qx . To find the P-value. 0. assuming that 25% of the archer's shots are normally bulls-eyes. we demonstate a solution based on the geometric formula.34 b*(5.The probability of success (P) is 0. See also: Binomial and Geometric Distributions Problem 31 An archer claims that 25% of her shots will be in the center of the target (i. 1. What is the probability that the sports writer will reject the archer's claim. a bulls-eye).7) = 0. A sports writer plans to test this claim by sampling 300 shots. b*(x. and the number of successes (r) is 1. We enter these values into the negative binomial formula.16 Solution The answer is (B). g(x. when it is actually true? (A) 0.e.. the writer will reject the archer's claim. 20% bulls-eyes).01 (B) 0.34 = 0.

025 = -2 where P is the hypothesized value of population proportion in the null hypothesis.0.16.) Note: This problem can also be treated as a binomial experiment. because the P-value is 0. you may need to use a graphing calculator or a normal distribution table.0. Should she reject the 8% claim? Assume that the significance level is 0. Using an online calculator (e. σ = sqrt[ P * ( 1 . because the P-value is 0. Therefore.028. It produces a P-value with a cumulative probability of 0. (On the actual AP Statistics exam. the binomial approach is computationally demanding. or a normal distribution table.20 . since Stat Trek's analytical tools will not be available. assuming that the null hypothesis is true. many statistics texts emphasize the approach presented above.. The binomial experiment is actually the more exact analysis.75) / 300] = sqrt(0. we showed how to analyze a binomial experiment. A quality control inspector randomly selects 300 new cars and finds that 33 have a defect.  The P-value is the probability that a z-score will be less than -2.05. and p is the proportion of bulls-eyes observed in the sample. which is a normal approximation of the binomial. and n is the sample size.g.000625) = 0.016. we find that the z score of -2. (B) Yes. because the P-value is 0.025 where P is the hypothesized value of population proportion in the null hypothesis.023.P ) / n ] = sqrt [(0. z = (p . Calculate the standard deviation (σ).0245. Previously.0 has a cumulative probability of 0.25)/0.P) / σ = (0.  Compute the z-score test statistic (z). a handheld graphing calculator. (A) Yes. Without a computer. . Stat Trek's free Normal Distribution Calculator). See also: Hypothesis Test for a Proportion Problem 32 The Acme Car Company claims that at most 8% of its new cars have a manufacturing defect.25 * 0. (C) No.

We use the Normal Distribution Calculator to find P(z > 1. p is the sample proportion. The solution to this problem takes four steps: (1) state the hypotheses.028) is less than the significance level (0.028.  Interpret results. Solution The answer is (B). We work through those steps below:  State the hypotheses. shown in the next section. (3) analyze sample data.05.P) / σ = (.0157 = 1. we calculate the standard deviation (σ) and compute the z-score test statistic (z).P ) / n ] = sqrt [(0. The test method. because the P-value is 0. and n is the sample size. Thus.0002453) = 0. we cannot accept the null hypothesis.08 * 0.028.0157 z = (p . the significance level is 0. Note: If you use this approach on an exam.91) = 0. you may also want to mention why this approach is appropriate. the P-value = 0.(D) No. Since the P-value (0.11 .91.  Analyze sample data.08)/0.08 Alternative hypothesis: P > 0. Using sample data. (2) formulate an analysis plan.  Formulate an analysis plan.05). For this analysis.28. and (4) interpret results. the P-value is the probability that the z-score is greater than 1. is a one-sample z-test.91 where P is the hypothesized value of population proportion in the null hypothesis. Since we have a one-tailed test. Specifically. The null hypothesis will be rejected only if the sample proportion is too big. the approach is appropriate because the sampling method was simple random .. The first step is to state the null hypothesis and an alternative hypothesis. Null hypothesis: P <= 0. (E) There is not enough information to reach a conclusion. σ = sqrt[ P * ( 1 .92) / 300] = sqrt(0.08 Note that these hypotheses constitute a one-tailed test.

The mean test score was 850 with a standard deviation of 100. See also: Residual Analysis in Regression Problem 34 Molly earned a score of 940 on a national achievement test. the sample included at least 10 successes and 10 failures. See also: Hypothesis Test for a Proportion Problem 33 In the context of regression analysis. the model is nonlinear. What proportion of students had a higher score than Molly? (Assume that test scores are normally distributed.) (A) 0. A random pattern in the residual plot indicates that linear regression is appropriate. The sum of the residuals is always zero. and the population size was at least 10 times the sample size. a non-random pattern suggests that a non-linear model may be appropriate. III. When the sum of the residuals is greater than zero.sampling. (A) I only (B) II only (C) III only (D) I and II (E) I. II. and III Solution The correct answer is (B). which of the following statements are true? I. A random pattern of residuals suggests that a linear model is appropriate. And influential points often increase the correlation coefficient.18 . whether the regression model is linear or nonlinear. II. Researchers use residual plots to decide whether data fit a linear or a nonlinear model. Influential points always reduce the correlation coefficient.10 (B) 0.

using an online calculator (e.8159. (A) I only (B) II only (C) III only . Stat Trek's free normal distribution calculator). we assume that test scores are normally distributed.90  Then.g. or the standard normal distribution table.82 (E) 0.. we find P(Z < 0.41 percent of the students tested had a higher score than Molly. See also: Standard Normal Distribution Problem 35 Which of the following statements are true? I.90) = 1 . subjects within each pair receive the same treatment. In this way. the P(Z > 0.1841.  First. A randomized block design controls for the placebo effect. In this case. using the z-score transformation equation.8159 = 0. Thus. we use the normal distribution as a model for measurement.P(Z < 0. As part of the solution to this problem.90) = 1 . In a matched pairs design.  Therefore.μ) / σ = (940 .0. the solution involves three steps.90) = 0. we transform Molly's test score into a z-score. we estimate that 18. II. A completely randomized design offers no control for lurking variables.850) / 100 = 0.50 (D) 0. we find the cumulative probability associated with the z-score.(C) 0. III. a handheld graphing calculator. Given an assumption of normality. z = (X .90 Solution The correct answer is B.

In a matched pairs design. strong evidence that Woods plays better on par 3 holes. (C) The P-value is between 0. On the par 3 holes. (E) The P-value is > 0. See also: Experimental Design Problem 36 A sports writer hypothesized that Tiger Woods plays better on par 3 holes than on par 4 holes. Solution The correct answer is (E). Solution The answer is (E). By itself.10. (3) analyze sample data.001 and 0. In a completely randomized design. The solution to this problem takes four steps: (1) state the hypotheses. and (4) interpret results.05.001. Randomization provides some control for lurking variables. How would you interpret this result? (A) The P-value is < 0. He reviewed Woods' performance in a random sample of golf tournaments.05 and 0.(D) All of the above. We work through those steps below: . Woods made a birdie in 20 out of 80 attempts.01 and 0. he made a birdie in 40 out of 200 attempts.10. the experimenter must include a placebo in one of the treatment levels. little or no support for the notion that Woods plays better on par 3 holes. (B) The P-value is between 0. very strong evidence that Woods plays better on par 3 holes. (2) formulate an analysis plan. moderate evidence that Woods plays better on par 3 holes. some evidence that Woods plays better on par 3 holes. subjects within each pair are assigned to different treatment levels. On the par 4 holes. (D) The P-value is between 0. subjects are randomly assigned to treatment conditions.01. arandomized block design does not control for the placebo effect. (E) None of the above. To control for the placebo effect.

and n4 is the number of par 4 holes. the P-value is the probability that the z-score is greater than 0. Null hypothesis: P3 <= P4 Alternative hypothesis: P3 > P4 Note that these hypotheses constitute a one-tailed test. p = (p3 * n3 + p4 * n4) / (n3 + n4) = [(0. the P-value = 0.20 * 200)] / (80 + 200) = 50/280 = 0. we compute the z-score test statistic (z).25 * 80) + (0. The first step is to state the null hypothesis and an alternative hypothesis.92. In short. the samples were independent.92 where p3 is the sample proportion of birdies on par 3. The null hypothesis will be rejected if the proportion of birdies on par 3 holes (p3) is sufficiently greater than the proportion of birdies on par 4 holes (p4).   Interpret results.  Formulate an analysis plan. we cannot reject the null hypothesis.p ) * [ (1/n3) + (1/n4) ] } SE = sqrt [ 0. we have little support for the notion that Woods plays better on par 3 holes.214 * 0. State the hypotheses. We use the Normal Distribution Calculator to find P(z > 0.18.0175 }= sqrt [0. Using those measures. which is shown below. Thus. you may also want to mention why this approach is appropriate.20)/0. n3 is the number of par 3 holes.p4) / SE = (0. the approach is appropriate because the sampling method was simple random sampling. Note: If you use this approach on an exam. Since we have a one-tailed test.25 . we calculate the pooled sample proportion (p) and the standard error (SE).214 SE = sqrt{ p * ( 1 .786 * 0.0029548] = 0. Specifically. Analyze sample data.18.214 * 0. and each sample included at least 10 successes and 10 failures. the test method is a two-proportion z-test.92) = 0. Since the P-value (0.0544 z = (p3 .786 * ( 1/80 + 1/200 ) ] = sqrt[ 0. Using sample data.0544 = 0. For this analysis. where p 4 is the sample proportion of birdies on par 4. . each population was at least 10 times larger than its sample.0.18) is greater than 0.10.

II. A nonlinear transformation is used to increase the relationship between variables.100 (B) 0.See also: Hypothesis Test for Difference Between Proportions Problem 37 In the context of regression analysis. Non-random patterns in a residual plot suggest a departure from linearity in the data being plotted. If the CEO's claim were true. The CEO claims that an average Acme light bulb lasts 300 days. which of the following statements are true? I. III. A residual plot reveals departures from linearity. A linear transformation neither increases nor decreases the linear relationship between variables. The sampled bulbs last an average of 290 days. (A) I only (B) II only (C) III only (D) I and II (E) I. A logarithmic model is the most effective transformation method. See also: Transformations to Achieve Linearity Problem 38 Acme Corporation manufactures light bulbs.226 (C) 0. what is the probability that 15 randomly selected bulbs would have an average life of no more than 290 days? (A) 0. A linear transformation increases the linear relationship between variables. The most effective transformation method depends on the data being transformed. a logarithmic model may be more effective than other methods.334 . II. but it other cases it may be less effective. In some cases. and III Solution The correct answer is (C). with a standard deviation of 50 days. A researcher randomly selects 15 bulbs for testing. it preserves the relationship.

and n is the sample size.226. Then. Then. II. For this practice test. The first thing we need to do is compute the t score. Since we know the t score. The calculator displays the cumulative probability: 0.6% chance that the average bulb life for 15 randomly selected bulbs would be less than or equal to 290 days.909945 = . we can use the T Distribution Calculator. Stat Trek's free T Distribution Calculator). a handheld graphing calculator.g.0. The null hypothesis involves a continuous variable.. based on the following equation: t = [ x .7745966 where x is the sample mean. you may need to use a graphing calculator or a t distribution table.(D) 0.1 = 14. The t score is equal to .300 ) / [ 50 / sqrt( 15) ] = -10 / 12.μ ] / [ s / sqrt( n ) ] t = ( 290 .0. III. we find the cumulative probability associated with the t score.7745966.775 Solution The answer is (B). Hence. The standard deviation of the population is unknown. using an online calculator (e. we select "T score" from the Random Variable dropdown box of the T Distribution Calculator. but on the actual AP Statistics Exam. s is the standard deviation of the sample. if the true bulb life were 300 days. . The sample size is large (greater than 40). or the t distribution table. there is a 22. we enter the following data:   The degrees of freedom are equal to 15 .443 (E) . μ is the population mean. See also: Student's t Distribution Problem 39 Which of the following would be a reason to use a one-sample t-test instead of a one-sample z-test? I.

what is the expected frequency count of male Independents? (A) 40 (B) 50 (C) 60 (D) 180 (E) 270 Solution . the t-test is preferred. Either test can be used when the null hypothesis involves a continuous variable.(A) I only (B) II only (C) III only (D) I and II (E) I and III Solution The answer is (A). When the standard deviation of the population is unknown. Democrat. or when the sample size is large. Voting Preferences RepublicanDemocratIndependentRow total Male Female Column total 200 250 450 150 300 450 50 50 100 400 600 1000 If you conduct a chi-square test of independence. Respondents were classified by gender (male or female) and by voting preference (Republican. See also: Student's t Distribution Problem 40 A public opinion poll surveyed a simple random sample of voters. or Independent). Results are shown below.

c is the expected frequency count when gender is level r and voting preference is level c. n is the number of observations in the sample. To apply the chi-square test for independence. using the following equation. . Er.3 = (400 * 100) / 1000 = 40000/1000 = 40 where r is the number of levels of gender. and Or.The answer is (A). c is the number of levels of the voting preference.c = (nr * nc) / n E1. Er. The computation for males who are classified as Independents is shown below. n r is the number of observations from level r of gender. we compute the expected frequency counts for each cell of the table.c is the observed frequency count when gender is level r voting preference is level c. nc is the number of observations from level c of voting preference.