This is an ongoing listing of errors in the 6th edition. I give credit below to those who point out errors. When I don't know the name of the person who pointed out the error, I am still grateful even if I can't give credit.
Because of the way I expect to distribute this list, I cannot be sure that subscripts and Greek symbols will come across. Subscripts are usually shown in parentheses [as in s(1)], superscripts as s^2, and Greek letters are spelled out.
| Page | Line | Correction |
| 59 | Before exercises | I gave the wrong URL. Change DataSets to DataFiles and ADD.dat to Add.dat. I have fixed it so that the old link will work, but it upsets my tidy little mind. |
| 61 | Ex2.21 | This is a terrible question because the data are not paired as you would expect when you compute sigma(XY). The only thing I can do now is apologize and ask you to pretend that they are paired in some fantastic way. Also ignore the "+" at the end of part c. |
| 61 | Ex2.23 | The reference should be to page 17 instead of page 19. |
| 101 | line 20 line 38 |
Change significant to significance. Change mu(eta) to mu(h). |
| 135 | Ex5.38 and 5.39 | Should be numbered Ex5.37 and 5.38. |
| 205 | lines1 & 4 | Reverse < and > in two places. |
| 207 | Exercise 7.14 and 7.15 | I was asking for a two sided test and two sided confidence limits in these two questions. |
| 215 | Figure 8.1 | The arrow should point at the vertical line separating the dark and light shaded areas. |
| 224 | line 2 | Replace XXX with 193. |
| 266 | Dale Berger has correctly pointed out that for power calculations for Pearson's r should be using Fisher's transformed value (r') and not the standard Pearson r. The changes need on that page should be evident. Dale calculated that the new sample size should be 83 instead of 88. | |
| 351 | bottom | I modified my formulae but forgot to modify the calculations accordingly. Psi is -1.25, and that correction carries over to next line, where the denominator is 1.5. The asnwer remains the same. Thanks to Cara Swain at Univ. of Oregon. David Bowlby caught the same error, as did Sam Ballard. |
| 364-65 | Mark Faust pointed out that I made an error in my discussion of confidence limits and it flows into the discussion of effect sizes. Please go to New364-365.pdf . | |
| 388 | Ex12.14 | Ignore the reference to Table 12.2. Go to the discussion at contrasts for unequal n's |
| 430 | Fig 13.4 | The figure doesn't quite match the data in the table, but the overall appearance is close enough that it won't mislead anyone. |
| 436 | Ex13.11 | You should refer to Ex11.3a, not Ex11.3b. |
| 437 | Ex13.26 and 13.27 | These should both refer back to Exercise 13.25 |
| 506 | Dale Berger caught me again. In the section on sample sizes it is R-squared, not R, that has an expected value of p/N - 1). Dale also recommended an excellent paper by Scott Maxwell [Maxwell, S. E. (2000). Sample size and multiple regression analysis. Psychological Methods, 5 , 434-458.] | |
| 530 | Table 15.6 | The entries in the last column should be .323, .106, and 3.04. The calculations immediate above were correct. |
| 581 | Top table | The error term should be given as SSresidual instead of SSregression. |
| 645 | Figure 18.6 | "less than or equal to 1 -04" should be "less than -4" |
| 654 | top | I forgot to mention that n(1) is the smaller of n(1) and n(2). Thanks Dale |
| 663 | Ex18.9 | It should ask you to rerun the analysis in Exercise 18.8, not 18.7. |
| 710 | Ex3.19 | It should read "The data are based on a sample, not a populations, so therew will be slight differences due to sampling error." |
| 7.13 | Ex7.13b | It would be better to say that the result "does not speak to the question of whether the SAT is a valid measure." |
| 713 | Ex7.25 Ex7.27 |
The correct value of t is -4.185 on 16 df,
which is significant at p = .001. The confidence interval should be -10.94 < mu < -3.58. |
| 716 | Ex11.27 | This is actually the answer to Exercise 11.28. The answers are correct in the Student Manual at http://www.uvm.edu/~dhowell/methods/StudentManual/StudentManual.html |
There is a self-extracting archive of all of the data found in DataAll.zip. Simply click on this link and tell the resulting dialog box where you want to store the download--the desktop is fine. After the file is downloaded, click on the downloaded file and indicate where you want the data files to be stored. Once the data are extracted, you can safely delete the archive file itself.
Last revised: 12/31/2006