Corrections to Sullivan’s Fundamentals of Statistics

► Should fix: students likely won’t recognize it as an error, but it doesn’t impact the core of the subject

◊ Wrong, but students will likely be able to supply the correction without help

Chapter 1

► Page 19: The random number display is not accurate. With a seed of 34 as shown, you should get 11, 4, 20, 29, and 11 as your first five random numbers with TI-83 or TI-84.

► Page 27: The book defines a convenience sample as one that is easily obtained. True, most convenience samples take less work than most random or systematic samples, but the real issue is that a convenience sample is not obtained in a planned or structured way, just by taking whatever subjects are handy. Taking a good sample (random, systematic, or other types) requires advance planning and thought.

► ► Page 50: (Problem 13) The correct answer is qualitative (or attribute or categorical). Some printings of your book give the incorrect answer quantitative.

► ► Page 51: (Problem 27) This problem has problems! I guess you’re meant to identify the problems with this sampling procedure.

Chapter 3

► Page 107: In the shaded definitions at the bottom, the symbol for population arithmetic mean is given as m. That should be μ, pronounced “mew” as the text says. (It’s spelled “mu” if you have to write it out in English.)

Chapter 4

► ► Page 185: The last sentence of the second paragraph is wrong. In fact, if |r| is less than the value from Table VIII then you can’t tell whether there is a linear relation or not. See Decision Points for Correlation Coefficient.

► Page 217: There are two answers to problem 5e because there are two data points at x = 3.8. The residuals are 1.2 and 0.2.

Chapter 5

► ► Page 281: The book’s answers to problem 15(c) through (e) are accurate. However, you could also use the rule of thumb on page 260, that when the sample is <5% of the population you can assume independence even when sampling is without replacement. These answers would also be acceptable: (c) (301/505)×(301/575) =0.2740; (d) (274/575)×(274/575) =0.2271; (e) 1−(d) =0.7729.

Chapter 6

► Page 298: In the last bullet point at the bottom of the page, X (upper case) should be x (lower case).

► ► Page 309: “Should We Convict?” tells you to assume that the characteristics are independent, but they aren’t. For example, someone with a beard is more likely to have a mustache than someone without a beard, and “woman with blonde hair” plus “black man with beard” is much more likely to be “interracial couple in car” than just any randomly selected man and woman. Therefore, the probability computed in part (a) is too low.

► ► Page 316: (Problem 11f) , the text should read “... resulted in not more than 12 of them having ...”. On that basis, the probability is about 0.0334, which would be considered unusual. For the problem as written, the probability is 0.0157, which would still be considered unusual.

Chapter 7

General note: Your book rounds z and then looks up the rounded result in a table, where your calculator is accurate to many decimal places. This means that your answer will often differ from the book’s by a rounding difference. Usually it will be obvious whether your answer is different from the book’s by more than just a rounding difference, but by all means feel free to ask.

► ► Page 322: Figures 5(a) and (b) are not in the same vertical or horizontal scale. The red curve in Figure 5(b) should have exactly the same height and width as either curve in Figure 5(a). The blue curve in Figure 5(b) should be half as tall and twice as spread out as the red curve.

► Page 369: (Problem 25) More accurate answers are a. 0.4105; b. 0.4680; c. 0.2180; d. 120; e. 144.155308≈145

Chapter 8

General note: Your book does table lookups, which are not quite as accurate as your calculator results. Usually it will be obvious whether your answer is different from the book’s by more than just a rounding difference, but if in doubt please feel free to ask.

► Page 376: “In Other Words” is poorly worded. It should say that if the population size is a definite positive integer, the population is finite. (All population sizes are positive integers. We say a population is infinite if it’s not possible to list all the members, even in principle. “Smokers” are an infinite population because people stop or start smoking all the time.)

► Page 396: The problem statement for Example 4(b) should end “results in 10 or less having hearing trouble?”

► Page 396: In Example 4, the book doesn’t mention that the probabilities computed for p̂ are only approximate, because they don’t use the continuity correction from pages 363–364. The smaller the sample, the more difference there will be if the correction is not applied. In most real-life situations in inferential statistics, a sample of binomial data will be considerably larger than 120, and the continuity correction is usually small enough to ignore.

► Page 401: (Problem 5b) Your book reduces the standard error fraction part of the way but not the whole way. Convert your answer and the book’s answer to decimals and they should match.

► Page 401–402: More accurate answers: 5c. 0.0127; 9b. 0.1376; 9c. 0.0508, not unusual; 11b. 0.0691; 11c. probability is 0.0419

► Page 402: (Problem 13) In some printings, the answer in the back says P(x̄ < −65000); the minus sign does not belong. Also, a more accurate answer is 0.0072.

Chapter 9

◊ Page 426: In the first line of the paragraph in the middle of the page, “freedom” is misspelled.

Chapter 10

Hypothesis tests: Your book presents the classical approach and the p-value approach. Our class uses the p-value approach, and you should just skip over the classical approach.

◊ Page 457: In the box near the bottom of the page, your book labels the two columns as “H₀ is true” / “H₁ is true”. While that’s quite correct, most books would present the two possibilities as “H₀ is true” / “H₀ is false”. It’s just a different way of expressing the same idea.

► ► Page 459: It’s true that we never accept the null hypothesis. However, when p>α you don’t just say there is insufficient evidence to reject H₀, which still leaves your audience with the impression that H₀ is true. Say that there is insufficient evidence to draw a conclusion, or that there’s insufficient evidence to decide between (H₀ in words) and (H₁ in words). For more, see When p>α, you fail to reject H₀

► Page 487–488: More accurate p-values: 11d. 0.0038; 13. 0.0004; 15. 0.0925

► Page 495–496: In Example 1, the book commits The Big No-no. p̂ is about 0.7005, but the book rounds to two decimal places, and using p̂ = 0.70 the book gets z = 3.32 instead of the correct 3.36. The book’s p-value of 0.0005 is larger than the correct 0.0004, though that doesn’t affect the conclusion of the hypothesis test.

► Page 502: More accurate p-values: 13. 0.1813; 15. 0.000033465622, which we write as p < 0.0001

► Page 504–506: More accurate p-values: 12. 0.0022; 15. 0.0636; 17b. 0.0065; 19. 0.4080; 21. 0.0218

Chapter 11

► Page 529–531: 9(b) More accurate endpoints are (2.0412, 11.359); 15(a) p = 0.1780; 15(b) More accurate endpoints are (−1.767, 4.5918)

► Page 543–544: 11(b) More accurate endpoints are (.0049, .07719); 21 p = 0.1586

► Page 548–549: 13(b) p = 0.6896; 15(b) p = 0.0003; 17(a) p = 0.0033; 17(b) (−.0577, −.0093); 23. (−128.5, −42.74)

Chapter 12

Page 601–604: 1. p = 0.7489; 3. p = 0.0836; 5d. p-value = 0.5314, fail to reject H₀; 7. p < 0.0001 (3.71254E-43); 9. p = 0.2751; 11. p < 0.0001 (3.619764E-27)

What’s New?

home page | problems with viewing?

This page is used in instruction at Tompkins Cortland Community College in Dryden, New York; it’s not an official statement of the College. Please visit www.tc3.edu/instruct/sbrown/ to report errors or ask to copy it.

For updates and new info, go to http://www.tc3.edu/instruct/sbrown/stat/