Errata for **Explaining Psychological Statistics, 3rd ed.** by Barry H. Cohen (rev. 8/6/12)

Please send any additional corrections, comments, suggestions, etc., to me at: barry.cohen@nyu.edu

Some really minor typos that are very unlikely to cause any confusion are not included in this list, but will

be corrected in future printings. If you are looking at the fifth (or a later) printing, all of the errors below have

been corrected, except for the ones whose page numbers are starred (e.g., p. 58*). To find out which printing

you have, look at the very bottom line of the copyright page (the page opposite the Dedication page, “For Leona”).

The bottom line has a string of numbers that begins like this: 10 9 8 … The last number in the series is the number

of your printing, so, for instance, the fifth printing will have the following string of numbers: 10 9 8 7 6 5

To avoid having to insert Greek letters, except for the upper-case sigma as summation sign, I have spelled out the

names of Greek letters; they should be assumed to be lower case, unless otherwise specified. To avoid creating

subscripts and superscripts, I use “_sub_” and “_super_”, respectively. Finally, D-bar means D has a bar over it.

Page No. |
Line/ Location |
Corrections (for first printing of the third edition) |

p. 15 |
Fifth line below Summation Rule 4 |
“1 · 4 + 2 · 5 = 3 · 6” should be “1 · 4 + 2 · 5 + 3 · 6” |

p. 58* |
Formula 3.7A |
“SS/N – 1” should be “SS/(N-1)” |

p. 66 |
Formula 3.12 |
“SS + ΣX2 …” should be “SS = ΣX2 …” |

p. 84* |
Formula 3.7A |
“SS/N – 1” should be “SS/(N-1)” |

p. 85 |
Formula 3.12 |
“SS + ΣX2 …” should be “SS = ΣX2 …” |

p. 121 |
Exercise #6 |
Change “(assume sigma = 15):” to “(assume sigma = 16):”. |

p. 193 |
Formula 7.4 |
There should be bars over the X’s in this formula. |

p. 195 |
Formula 7.7A |
“s_super_1_sub_2” should be “s_super_2_sub_1” |

p. 203 |
Line 3 |
“that” should be “than” |

p. 208 |
Second line below “HOV Tests and the … |
“the F text” should be “the F test” |

p. 221 |
Heading above Formula 7.12 |
End with “… trimmed t test method:”. [Delete “(where wi …)” at end.] |

p. 252 |
Exercise #6 |
The exercise should end with: “a. Exercise 7C5 b. Exercise 7C6”. [Not Exercise 7C1 and Exercise 7C2, as in the original first printing] |

p. 253 |
Formula 8.5 |
The caption above (i.e., that applies to) this formula should end like this: “ … and sample sizes are equal:” (instead of “and sample size is known:”). |

p. 268 |
Formula 9.3 |
“(ΣXY = NXY)” should be “(ΣXY - NXY)” [but note that there should be bars over both the X and the Y following N] |

p. 284 |
Formula 9.3 |
“(ΣXY = NXY)” should be “(ΣXY - NXY)” [but note that there should be bars over both the X and the Y following N] |

p. 295* |
Formula 10.7B |
“(1 – r2)” should be “(1 – r_super_2)” – i.e., the 2 should be an exponent |

p. 298 |
First eq’n under Finding the Regression Eqn |
The equation should end with “= .43”, not just = .4. |

p. 298* |
Second eq’n below “Finding the Regression Equation” |
“74 = .43(36)” should be “74 - .43(36)” – i.e., a minus sign rather than an equals sign |

p. 298* |
First eq’n below “Making Predictions” |
“.43(36) = 58.6” should be “.43(36) + 58.6” – i.e., a plus sign rather than an equals sign |

p. 301 |
First eq’n under An Example of a Conf Interval |
Under the first sqrt, the r in (1-r) should have an exponent of 2, not a sub-script of 2 |

p. 301 |
Second eq’n under An Exple of a Conf Int’l |
The first square root should be extended to cover the term that follows “1/40” (thus covering a total of three terms). |

p. 328 |
Line 7 of text |
Should be “D-bar = -1.8/10 = -.18” (should not end with -1.8) |

p. 336 |
Exercise #9 |
Change: “For the data in Exercise 10B4:” to “For the data in Exercise 10B6:” Also, part a should end with: “… slope in Exercise 10B6.” [Not Exercise 10B4.] |

p. 356 |
First equation on page (MSw = ) |
Where you see n1, n2, etc., the numbers should be subscripts. |

p. 356 |
Third equation on p. (MSbet = ) |
The first term (like the other terms in the numerator) should have a minus sign, instead of a plus sign. |

p. 356 |
Formula 12.7 (1/3 down the p.) |
The “2” at the end of the numerator should be an exponent. |

p. 359 |
Eq’n after first paragraph |
Two corrections: a) the first term of the eqn should look like the last term of Formula 12.8 at the top of the page – that is, Σ(T/n), with the super- and subscripts included; b) 12.544/4 should be “12,544/4”. |

p. 359 |
First line of text below the first eqn (Finally …) |
Two more corrections: a) the capital S should be a summation sign, and T/n should be enclosed in parens – that is, Σ(T/n), with the super- and subscripts included; b) 15,808 should be preceded by an equals sign: Σ(T/n) = 15,808. |

p. 360 |
Formula 12.11 (top of page) |
The second “equals” sign should be a “minus” sign. |

p. 363 |
Formula above “Effect Size” (halfway down p.) |
Change “sigma =” to “phi =” |

p. 367* |
Next to last line of first full paragraph |
“N – 1, as compared to N – k” should be “N – 2, as compared to N – k” |

p. 370* |
Summary point # 6 |
Where it says f = g/2, it should say: f = g / sqrt (2). [Note: it is bold f that equals bold d divided by 2.] |

p. 373* |
Exercise #13, pt. a |
Where it says (in the Hint) that f = d/2, the f should be printed in boldface, to represent a population parameter. |

p. 373 |
Exercise #13, pt. b |
Should read: “b. Recalculate 8B5a for a two-tailed test; begin by assuming …” |

p. 377 |
First line |
“ the numbers 10, 20, and 80” should be “the numbers 10, 20, 40, and 80” |

p. 383 |
Formula 12.10 |
“SSW” should be “SS_sub_W” |

p. 384 |
Heading above Formula 12.14 |
Should read: “Phi as a function of sample size and population effect size …” |

p. 384 |
Formula 12.15 |
The first letter, “f”, should NOT be in boldface. [It is a sample statistic, not a population parameter.] |

p. 405 |
The line above the calculation of SScontrast, near bottom of page |
Change: “The next step is to use Formula 13.13 …” to “The next step is to use Formula 13.14 …” |

p. 423 |
Exercise #4, pt. c |
Delete second question, which begins “How could you justify …” |

p. 451* |
8 lines from the bottom |
“ordianry g” should be “ordinary g” |

p. 477 |
Third line of first full paragraph |
Change “… their biased variance equals 257.75 …” to “… their biased variance equals 257.25 …” [The rest of the numbers on that page are written correctly.] |

p. 511 |
Fifth line under Publishing the Results of an … |
The terms “auditory” and visual-successive” should be reversed in that sentence, so that it reads as follows: “… by visual-successive (M = 12.17), and then auditory (M = 10).” |

p. 517* |
Exercise #10, pt. b |
Should end with: “… have power equal to .99 for a .05, two-tailed test?” |

p. 522 |
Beginning of fourth line |
Should read: “… MS_sub_inter = 125.67 / 10 = 12.57 …” [Not 125.7] |

p. 527 |
Formula 15.6 |
Should begin with: “Est. omega squared = …” [The “2” should be an exponent, not a subscript.] |

p. 568 |
Exercise #3, pt. a |
The first sentence should read: “Ignoring the therapy conditions, test the significance of the linear and quadratic trends over time.” In the second sentence, “preformed” should be changed to “performed”. |

p. 641 |
Caption for Figure 18.5 |
“Exercise 18A6” should be “Exercise 18A8”. |

p. 651* |
Formula 18.11 |
“(1 + F_x) /df_error” should be “1 + (F_x) /df_error)” [This formula is printed correctly on p. 676.] |

p. 748 |
Exercise #9, pt. b |
Should read: “Repeat the test in part a, using Formula 21.5.” [Not Formula 21.4] |

p. 748 |
Exercise #10, pt. b |
Should read: “Repeat the test in part a, using Formula 21.5.” [Not Formula 21.4] |

p. 781 |
Answers to Exercise #6 in Chap. 4, sec. C |
The answer to part d should read: “2 X (.2643) (.1736) = .092. |

p. 782 |
Answers to Exercise #1 in Chap. 7, sec. A |
In the answer to part a, instead of 42/100 under the square root sign, it should read 4-squared over 100 -- i.e., (4_super_2) / 100 |

p. 785 |
Answers to Exercise #3 in Chap.11, sec. A |
The answer to part a should read: “… = -1.55”, instead of “… = -.155” |

p. 788* |
Answer to Ch. 14, sec. B, #2d |
“eta-sq_p” should be “eta-sq_ord” (i.e., ordinary eta squared instead of partial eta squared). |

p. 788 |
Answer to Exercise #10a in Ch.14, sec. B |
The third (i.e., bottom) entry in the p column of the ANOVA Summary table should be “> .05” (i.e., not sig.), rather than < .01. |