Errata

Durbin-Watson is not in the index. I looked up autocorrelation but the D-W is not associated with it in the text. It is mentioned on pg. 203

Note from the Author or Editor:
Add "Durbin-Watson statistic" to the index, with a reference to page 203.

Text states: "The idea behind the Six Sigma program is to reduce variability sufficiently that output in the range of ±3 sigma will still be acceptable to the customer." That is incorrect.

The idea behind Six Sigma is to reduce variabiliby sufficiently such that output in the range of ±6 (not ±3) sigma will still be acceptable to
the customer.

Output in the range of ±3 sigma would still produce an out of specification condition on average about once of every 370 times/parts. This is considered unacceptable in high-volume production. At ±6 sigma, the out of specificaiton conditions are measured in defects per million opprotunities (DPMO).

See: https://web.archive.org/web/20060203042658/http://www.motorola.com/content/0,,3088,00.html

...and...

http://en.wikipedia.org/wiki/Six_Sigma#Etymology_of_.22six_sigma_process.22

...as well as numerous printed references (e.g. Tennant, "SIX SIGMA: SPC and TQM in Manufacturing and Services")

Note from the Author or Editor:
p. 344 1st pargraph 2nd sentence replace the "3 in "+- 3 sigma" with a "6" so it reads "output in the range of +- 6 sigma…"

It should be noted that although many physical measurements are interval-level,

should be read as

It should be noted that although many physical measurements are ratio-level,

Note from the Author or Editor:
p. 5, 1st sentence of first complete paragraph should begin "It should be noted that although many physical measurements are ratio-level..."

• xis the number whose mean we are calculating.

should be read as

• xis the set of numbers whose mean we are calculating.

Note from the Author or Editor:
p. 22 fist bullet point following Figure 2-1 should read: "x is the set of numbers whose mean we are calculating."

(with "x" in italics)

[(0.95)(0.010]+(0.01)(0.99)]
should be read as
[(0.95)(0.010]+[(0.01)(0.99)]

0.0095+.00099
should be read as
0.0095+0.0099

Note from the Author or Editor:
p. 35 Figure 2-13 first denominator is missing a square bracket and should read "[(0.95)(0.010]+[(0.01)(0.99)]"

second denominator is missing a leading zero and should read:
"0.0095+0.0099"

[(0.95)(0.005]+(0.01)(0.995)]
should be read as
[(0.95)(0.005]+[(0.01)(0.995)]


0.00475+.00995
should be read as
0.00475+0.00995

Note from the Author or Editor:
p. 36 first denominators is missing a square bracket and should read: "[(0.95)(0.005]+[(0.01)(0.995)]"

second denominator is missing a leading zero and should read:
"0.00475+0.00995"

E = (E = 1) ∪(E = 2∪(E = 3) ∪(E = 4)
should be read as
E = (E = 1) ∪(E = 2) ∪(E = 3) ∪(E = 4)

Note from the Author or Editor:
p. 38 3rd solution to the problem beginning "If I roll a die once..." there is a parenthesis missing following the 2; the formula following "that is," should read: "E = (E = 1) U(E = 2) U(E = 3) U(E = 4)"

U is the symbol for union (not a capital U)

This figure is drawn incorrectly. The vertical divisions between the regions do not align with the markings on the horizontal axis. This is most evident at the "1 sigma" mark (the division in the area above is noticably to the left), and the ""2 sigma" mark (the division above is even further to the left). The figures given (cumulative propabilities between integer mutiples of the standard deviation) are correct only for the markings along the horizontal axis; not for the incorrectly aligned shaded areas.

Note from the Author or Editor:
p. 48 figure 3-2 horizontal axis does not align with breaks in shaded figure, particularly on the positive side of 0 , particularly for 1-sigma and 2-sigma labels

notation for Normal distribution, x ~ N (100, 5) is used throughout the page,
however, in strictly speaking, this should be read as
x ~ N (100, 5^2) 2 is superscript

Note from the Author or Editor:
p. 49 Following 8th line from the top (ending "as shown in Figure 3-4.") add "Note that some texts use the mean and variance (rather than standard deviation) to describe normal distributions, in which case we would write "x ~ N (100, 25)" in this case."

about 2.5%
should be read as
about 2.3%

Note from the Author or Editor:
p. 49 9 lines from the bottom change "about 2.5% of the population" to "about 2.3% of the population"

less than half of 1%

should be read as
less than 0.14%

because the precise number is 0.00135

Note from the Author or Editor:
p. 49 next to last paragraph, last sentence (beginning "A score of 100 is more unusual"): change "less than half of 1% of the population" to "less than 0.14% of the population"

The line "np = 40(0.5) = 20 n (1 - p) = 40(1 - 0.5) = 20" appears as one long equation/calculation. However, it should be two clearly seperated equations/calculations, such as: "np = 40(0.5) = 20, and n(1 - p) = 40(1 - 0.5) = 20.

Note from the Author or Editor:
p. 52 formula after the first paragraph: insert "and" after the first 20 so it reads:

np = 40(0.5) = 20 and n(1 - p) = 40(1 - 0.5) = 20

also remove space between n and (1 - p) following the "and"

P(k ≥= 3) = 1 − P(k ≤= 2) = 1 − (0.11 + 0.27 + 0.30) = 0.32

should be read as

P(k ≥ 3) = 1 − P(k ≤ 2) = 1 − (0.11 + 0.27 + 0.30) = 0.32

Note from the Author or Editor:
p. 78 4th formula from top should read:
P(k ≥ 3) = 1 − P(k ≤ 2) = 1 − (0.11 + 0.27 + 0.30) = 0.32

(i.e., take extra equals sign out of "P(k ≥= 3)")

Text reads: "...the table probability of b(2; 10; 0.5) is 0.67780;..."

For this example, p=0.2, not 0.5.

The text should read: "...the table probability of b(2; 10; 0.2) is 0.67780;..."

Note from the Author or Editor:
p. 78 third paragraph (beginning "Using Figure D-9), following comma should read "...the table probability of b(2; 10; 0.2) is 0.67780…" (third value in the parentheses, it should be 0.2 not 0.5)

The x-axis (vertical axis) in a histogram

should be read as

The x-axis (horizontal axis) in a histogram

Note from the Author or Editor:
p. 109 paragraph 1 last sentence should begin "The x-axis (horizontal axis) in a histogram…"

the bottom number of Percent Obese 10 should be read as 0
as described in the sentence below the figure (05-100%)

Note from the Author or Editor:
p. 116 Figure 4-36 on the Y-axis, the lowest number should be 0, not 10

see the references on Excel in Appendix C for more on this.

should be read as

see the references on Excel in Appendix B for more on this.

There are no specific references on Excel in Appendix C.

Note from the Author or Editor:
p. 117 2nd paragraph after "Exercises" next to last sentence following semicolon should read "see the references on Excel in Appendix B for more on this."

The example subscript in the lower right cell is incorrect. It should be Cell22 instead of Cell21

Note from the Author or Editor:
For table 5-4, change subscript of fourth cell (row 2, column 2) to 22 (not 21)

Text states: "I also determined from a computer analysis that the exact probability of getting a chi-squared stastic with one degree of freedom at least as extreme as 6.43 is 0.017..."

I have attempted to duplicate that result without success. Using R I get a p-value of 0.01123. R session follows:

> A<-array(c(15,10,25,20),c(2,2))
> mcnemar.test(A, correct=F)

McNemar's Chi-squared test

data: A
McNemar's chi-squared = 6.4286, df = 1, p-value = 0.01123

Similarly, using MS Excel: "=CHISQ.DIST.RT(6.43,1)" returns approximately 0.01123.

Perhaps the text should read: "...as extreme as 6.43 is 0.011..."

Note from the Author or Editor:
p. 136 1st paragraph third line from the bottom change 0.017 to 0.011 ("at least as extreme as 6.43 is 0.011…"

The definition of r_{pb} in Figure 5-19 (point-biserial correlation coefficient) shows the numerator of the fraction as
\bar{X}_{1}-\bar{X}_{0}\sqrt{p(1-p)}
i.e. it indicates that only the X0 term is multiplied by sqrt{p(1-p)}.

The example in Figure 5_20 indicates that the whole (X1 - X0) term which should be multiplied by sqrt{p(1-p)}. The definition in 5_19 could be amended to put brackets around the whole (X_1 - X_0) term.

Note from the Author or Editor:
For Figure 5-19 and F-20, the X-bar-sub1 minute X-bar-sub0 term should be in parentheses

The caption of Table 5-16 is:
Example data to calculate gamma

This is not appropriate, because at this step, the data is introduced only as
Is there a strong relationship between weight and blood pressure in
the data set shown in Table 5-16?

Gamma calculation starts from the following paragraph.

Example data on BMI amd B;ppd pressure (which will be used to calculate gamma)
should be more appropriate for the reader.

Note from the Author or Editor:
p. 143 caption for Table 5-16 should read:

"Example data for BMI and blood pressure"

Prior errata states:
"In Figure 6.12, in the denominator following the first = sign, replace 37.18 with 31.78 (the rest of the equation is correct)"

No, the rest of the equation is not all correct. The corrected value (31.78 vice 37.18) results in a value 2.52 (not 2.73) in the intermediate denominator result. Finally, 2.76/2.52 results in a t-stastic of 1.10 (following the rounding conventions utilized in the text).

Note from the Author or Editor:
p. 163 Figure 6-12 intermediate result (following second equals sign) should be 2.76/2.52

final result should be 1.10

This is a sequela of the error in Figure 6-12, already discussed in the errata. The same incorrect "2.73" denominator value from Figure 6-12 (which resulted from the use of 37.18 vice 31.78 for the pooled variance) was used again here, in Figure 6-14.

"...(2.10)(2.73) = (-2.97,8.49)" should instead read: "...(2.10)(2.52) = (-2.53, 8.05)".

Note from the Author or Editor:
p. 164 Figure 6-14 following plus/minus symbol should read:
(2.10)(2.52) = (-2.53, 8.05)

The listed figures are in a section entitled "Relationships Between Continuous Variables". In addition, almost all bear titles starting with "Graph of the model y=...". As such, they would certainly be clearer if depicted as (continous) line graphs rather than graphs of discreet points (e.g. "data" points). To graph them as points, as they are, gives the incorrect implication that only integer values of x are valid.

If, for some other reason(s), a graph of discreet points is insisted upon, I see no reason that the x=0 points should be omitted, as they are in these six Figures (including the y=0 point in Figure 7-5). Its omission gives the impression that there's something special about the x=0 (or y=0) value, when there isn't.

Note from the Author or Editor:
p. 177 1st paragraph add sentence "Note that Figure 7-2 through 7-5 are depicted as a series of points on a line, but could also be depicted by a solid line."

the sum of squares of x and y

should be read as

the sum of cross-products of x and y

as with page 184 errata correction

Note from the Author or Editor:
p. 185 caption for figure 7-18 should be "Calculating the sum of cross-products of x and y"

and for figure 7-19 should read "Computational formula for calculation the sum of cross-products of x and y"

sum of squares

should be read as

sum of cross-products

Note from the Author or Editor:
p. 185 first paragraph beginning "There is also a computation formula..." (after Figure 7-18): change "sum of squares" to "sum of cross-products"

-0.663 is placed under t column, it should be placed under Beta column.

Note from the Author or Editor:
p. 216 table 8-10 headers are left-aligned and values are right-aligned. This is misleading because the column "standardized coefficients" is so wide that the value for "Bet"a (-0.663) appears to be in the column for "t". Could the headers be left-aligned or that column made narrower, perhaps by using two lines for "Standardized coefficients" so it is clear that the value is in the "Beta" column?

(p − .001)
should be read as
(p = 0.001)

Note from the Author or Editor:
p. 218 2nd line from the bottom, value in parentheses should be:
" (p = 0.001)"

H0: there is no difference in grip strength between men and women.
H1: there is a difference in grip strength between men and women.

should be read as

H0: there is no difference in grip strength between different alcohol consumption.
H1: there is a difference in grip strength between different alcohol consumption.

You can say
Alcohol consumption affects (does not affect) grip strength
as well.

Note from the Author or Editor:
p. 227 top of page, after heading "Main effect for alcohol" the two hypotheses should read:

"H0: there is no difference in grip strength between people with different levels of alcohol consumption.
H1: there is a difference in grip strength between people with different levels of alcohol consumption."

Figure 10-7. Means for the natural log of the adolescent fertility rate for low, medium, and high levels of the log of per capita national income and expected years of schooling

should be read as
Figure 10-7. Means for the natural log of the adolescent fertility rate to the log of per capita national income with the low, medium, and high levels of expected years of schooling

Note from the Author or Editor:
p. 254 caption for Figure 10-7 should read "Means for the natural log of the adolescent fertility rate for low, medium, and high levels of the log of per capita national income and low, medium and high levels of expected years of schooling"

Figure 10-10. Histogram of standardized residuals for model 3

should be read as
Figure 10-10. Scatterplot of standardized residuals for model 3

Note from the Author or Editor:
p. 257 Figure 10-10 caption should be "Scatterplot of standardized residuals for model 3"

The medium-sized scanner (4 cm) showed no improve-

should be read as

The medium-sized scanner (4 sq cm) showed no improve

Note from the Author or Editor:
p. 271 4th line from the bottom, following period sentence should begin "The medium-sized scanner (4 sq cm) showed…"

In Table 11-4, the rightmost two columns' labels are:
Lower Upper.

This is a bit misleading, and they should be read as
Lower Bound Upper Bound

as was in Table 11-8, page 281 for clarification.

Note from the Author or Editor:
p. 277 Table 11-4 two columns furthest right should be labeled "Lower Bound" and "Upper Bound" [add "Bound"]

(1.018, 1.014)

should be read as
(1.018, 1.054)
following the data in Table 11-4

Note from the Author or Editor:
p. 278 3rd line of 3rd paragraph, beginning "and the 95%…": the second number in the parenthesis should be 1.054 (not 1.014)

Male=1 should be read as Male (3 occurrences)

Note from the Author or Editor:
p. 281 Table 11-8 2nd column change "Male = 1" to "Male" (occurs 3 times)

For the linear model Y= β0+ β1X1 + e, where Y is performance and x is caffeine,

should be read as

For the linear model Y= β0+ β1X1 + e, where Y is performance and X1(subscript) is caffeine,

Note from the Author or Editor:
p. 284 first sentence following Table 11-9, following comma, should read "where Y is performance and X1 is caffeine..."

The 1 should be subscript to the capital X

The label, "Parameter estimates" should locate at the rightmost four columns,
not the column with "Sig." which belongs to "Model Summary."

Note from the Author or Editor:
p. 286 Table 11-12 top line: "Parameter Estimates" should be located over the "Constant" column (not the "Sig" column)

In Table 11-13, the rightmost two columns' labels are:
Lower Upper.

This is a bit misleading, and they should be read as
Lower Bound Upper Bound

as was in Table 11-8, page 281 for clarification.

Note from the Author or Editor:
p. 288 Table 11-13 final two columns on right should be labeled "Lower Bound" and "Upper Bound" [add "Bound"]

Some information are missing in this table:

1) function numbers are not put on two functions. the row of 1 2 should be added under the row Cluster number of case and function.

2) No information are given to the cluster numbers.
1 (Reading, Verbal, Spelling) 2 (Arithmetic, Geometry) 3 (Music, Sports)

3) Explanation for the value is missing. It would be better to add such as:
The value is the distance of the cluster centroid from discrimination functions as represented as a line

Note from the Author or Editor:
p. 304 Table 12-20 several edits

1. add a row following the gray heading over the second and 3rd columns, with numbers 1 (over 4.804) and 2 (over -0.169)

2. In column 1, identify the clusters as follows:

1 Reading, Verbal Spelling
2. Arithmetic, Geometry
3. Music, Sports


3. add a line below the table including the explanation: "The value is the distance of the cluster centroid from discrimination functions as represented as a line."

In this formula, W is the smaller of the two rank sums, µW is the expected sum of
ranks as calculated previously, and σW is the standard error, calculated as shown in Figure 13-5.

should be read as

In this formula, W is the smaller of the two rank sums, µW is the expected sum of
ranks as calculated previously, and σW is the standard deviation, calculated as shown in Figure 13-5.

Note from the Author or Editor:
P. 312 sentence following Figure 13-4, should read as follows:

"In this formula, W is the smaller of the two rank sums, µW is the expected sum of
ranks as calculated previously, and σW is the standard deviation, calculated as shown in Figure 13-5."

The measurement of interest in this particular study is a 100-point measure of mood state in which 0 is very low affect, and 10 is very high affect.

should be read as
The measurement of interest in this particular study is a 100-point measure of mood state in which 0 is very low affect, and 100 is very high affect.

The scale is 100 instead of 10.

Note from the Author or Editor:
p. 317 3rd sentence of 3rd paragraph, beginning "The measurement of interest" the end should read "and 100 is very high affect." (not "and 10 is very high affect").

No particular pattern in the data is presented in Figure 14-14

should be read as

No particular pattern in the data is presented in Figure 14-13

Note from the Author or Editor:
p. 341 last paragraph, change reference to Figure as follows:

"No particular pattern in the data is presented in Figure 14-13..."

The caption of Figure 14-13
Figure 14-13. Control chart of weight in ounces for 40 screws (individual values) with a process mean of 3.0

should be read as
Figure 14-13. Run chart of weight in ounces for 40 screws (individual values) with a process mean of 3.0

as the preceding sentence describes "Both features are displayed in the hypothetical run chart in Figure 14-13." and the chart is not a control chart which needs control lines as in Figures 14-24 and 14-25.

Note from the Author or Editor:
p. 341 Figure 14-13 caption replace "Control chart" with "Run chart" so it reads "Run chart of weight in ounces…"

In the solution to this example, list item #3 indicates a violation: "Six points in a row in the same direction (rule 3)".

First, this is an subtly incorrect simplification of the rule, which is more correctly stated on p345 as "If 6 consecutive points fall in the same direction, that is, all increasing or all decreasing".

In the charts (Figures 14-24 & 14-25), point 28 decreases from point 27. The subsequent 5 points (numbers 29-33, inclusive) each "fall in the same direction" (i.e. all increase). Finally, the next point, #34, decreases. Thus, only 5 points (29-33) "fall in the same direction", not 6, and this series of points is NOT a rule violation.

An example of an actual violation of Nelson's control chart rule #3 is shown at:
http://en.wikipedia.org/wiki/Nelson_rules

Specifically, this graphic image:
http://en.wikipedia.org/wiki/Nelson_rules#mediaviewer/File:Rule_3_-_Control_Charts_for_Nelson_Rules.svg

Note from the Author or Editor:
p. 349 next to last sentence (numbered 3): reword as:

"Not a rule violation, but close: five consecutive point in the same direction"

Even though hospital 1 had more postsurgical infections in the period studied, these
should be read as

Even though hospital 1 had less postsurgical infections in the period studied, these

Note from the Author or Editor:
p. 357 first line should begin "Even though hospital 2 had more..." (not "hospital 1 had more...")

The parenthesized part of
"Looking at the age distribution and age-specific rates for the employed versus unemployed　populations, we see that for each age group, the rates of arthritis are　somewhat higher in the unemployed group than in the employed group (the opposite
pattern from that seen when data from all the age categories is combined)." that is,
" (the opposite pattern from that seen when data from all the age categories is combined)" seems not true at the last row of Table 15-5.
Either this should be deleted or revised to be precise.

Note from the Author or Editor:
p. 359 paragraph following Table 15-5, first sentence, should read "Looking at the age distribution and age-specific rates for the employed versus unemployed populations, we see that for each age group, the rates of arthritis are only slightly higher in the unemployed as compared to the employed group, with the exception of the 65+ group.

The total persons with arthritis for the Employed category does not sum correctly. 127 + 260 + 105 = 492, not 387 that is printed.

Note from the Author or Editor:
p. 359 table 15.5 3rd column "Persons with arthritis" under "Employed": total (bottom line) should be 492 no 387. For that same line, rate/1000 should be 49.2 not 38.7.

Same page, table 15-4, row "Employed" column "Persons with arthritis should be 492 and "rate per 1,000" should be 49.2.

Caption for Figure 15-11 reads "Expected cases for the employed, age 25-44, category". This category does not exist. The caption should read "Expected cases for the employed, age 18-44, category".

Note from the Author or Editor:
p. 360 figure 15-11 change caption to read "Expected cases for the employed, age 18-44, category"

Probability = odds (1 + odds)

should be read as

Probability = odds / (1 + odds)

Note from the Author or Editor:
p. 369 in box "Odds" , 9 lines from the bottom, second formula (following comma) is missing slash, should read:

"Probability = odds/(1 + odds)

the rightmost fraction
p1(1-p1) / p2(1-p2)

should be read as
p1(1-p2) / p2(1-p1)

Note from the Author or Editor:
numerator should be: p1(1-p2)

denominator should be: p2(1-p1)

(the 1 and 2 in p1 and p2 are subscripts)

about 66% of the scores will be within one standard deviation of the mean

should be read as

about 68% of the scores will be within one standard deviation of the mean

Note from the Author or Editor:
p. 388 10 lines from bottom, change "about 66% of the scores will be...to "about 68% of the scores will be..."

Caption, Figure 16-12. Calculating the coefficient of equivalence
should be read as
Figure 16-12. Calculating the estimated reliability from the coefficient of equivalence

Note from the Author or Editor:
p. 397 change caption for Figure 16-12 to "Calculating the estimated reliability from the coefficient of equivalence"

Caption,
Figure 16-13. An alternative formula for the coefficient of equivalence

should be read as

Figure 16-13. An alternative formula for the estimate of reliability from the coefficient of equivalence

Note from the Author or Editor:
p. 398 caption 16-13 change to: " An alternative formula for the estimate of reliability from the coefficient of equivalence"

With the line,
is the total variance for the test (usually estimated by s2
x.

we need missing closing parenthesis,
and the right form for
s (subscript x) (superscript 2 meaning square of (sx))

Note from the Author or Editor:
p. 400 3rd line from the top needs closing parenthesis just before the period, and the subscript x should be directly below the square sign (as in the sigma-hat-sub-x-squared) at the beginning of this line)

Something wrong with the sentence: "Value labels are assigned to variable labels but are assigned to the values of individual variables." Perhaps, it can read as "Value labels are assigned to the values of individual variables."

Note from the Author or Editor:
p. 421 2nd paragraph 7th sentence from the top should read: "Value labels are assigned to the values of individual variables."

"the coding systems most commonly used are EBCDIC (Extended Binary Coded Decimal Interchange Code) and ASCII (American Standard Code for Information Interchange). " is misleading and/or obsolete.
We should include Unicode rather than EBCDIC, especially for PC and Web environments and places other than United States.

Note from the Author or Editor:
p. 422 3nd paragraph edit the phrase in question as "the coding systems most commonly used are Unicode and ASCII (American Standard Code for Information Interchange)."

−8^0 = 1

should be read as

(−8)^0 = 1

Note from the Author or Editor:
p. 477 middle of page, following bullet point for Zero exponent": put parentheses around "-8" so it reads (-8)**0 = 1"

(** means the 0 should be an exponent, i.e., superscript)

when both a and b >= 0

should be read as

when a>=0 and b > 0
because b is a denominator and should not be 0

Note from the Author or Editor:
p. 478 Figure A-4 end should read "when a >= 0 and b > 0"

Substitute this value into the first equation to solve for y:
should be read as
Substitute this value into the solution for y from the first equation:

the equation given is the value for y, not the first equation itself, so I changed the explanation. We can change the equation as well, though.

Note from the Author or Editor:
p 481 6th paragraph, beginning "Substitute this value into the first..." should read "Substitute this value for x into the solution for y:"

The two equations are typeset as a fraction, with one equation as the numerator and the other as the denominator, with a plus sign aligned with the viniculum.

I presume what was meant was to show columnar addition, with a horizontal line below the lower equation, and the plus sign to the left of the lower equation.

Note from the Author or Editor:
p. 482 Figure A-6 remove the horizontal line between 4x + 2y = 12 and 3x - 2y = 16

Exponents, Roots, and Logarithms
should be read as
Exponents and Roots

Note from the Author or Editor:
p. 495 second header should read "Exponents and Roots"

Natural Logarithms
should be read as
Logarithms

Note from the Author or Editor:
p. 496 1st heading should be "Logarithms" not "Natural Logarithms"

ln3 = 3

should be read as
ln e(superscritpt 3) = 3

Note from the Author or Editor:
p. 496 #5 under "Natural Logarithms" should read:

ln e**3 = 3

[the 3 following the e is a superscript)

The link (http://www.daheiser.info/excel/frontpage.ht) does not have proper information. the link was changed or lost the information.

Note from the Author or Editor:


substitute the URL http://www.statisticalengineering.com/Weibull/excel.html

for http://www.daheiser.info/excel/frontpage.ht
on the 8th line of the first Excel paragraph, p. 510

In Figure B-13, you can see a worksheet (Sheet3, as you can tell from the lower tab

should be read as

In Figure B-13, you can see a worksheet (Sheet1, as you can tell from the lower tab

Note from the Author or Editor:
p. 510 last paragraph first sentence should refer to Sheet1, not Sheet3 (…"Sheet1, as you can tell from the lower tab)."

Two websites run by the experienced SPSS programmer Raynald Levesque; both are loaded with tips, tricks, and sample code.

should be read as

The website run by the experienced SPSS programmer Raynald Levesque; it is loaded with tips, tricks, and sample code.

There is only one website by Raynald.

Note from the Author or Editor:
p. 523 first bullet point following url should begin "The website run...

http://www.math.unb.ca/~knight/utility/
Not Found

the link maybe obsolete.

Note from the Author or Editor:
at end of 3rd paragraph,in the final sentence, remove the text "and are available..." to the end of the sentence. (refers to Knight's website, which no longer exists)

http://www.math.unb.ca/~knight/utility/
Not Found

the link maybe obsolete.

Note from the Author or Editor:
Remove the text "; the complete tables are available" to the end of the sentence.

Delete the first bullet that is now the following:

StatLib: Data, Software and News from the Statistics Community (http://lib.stat
.cmu.edu/index.php)
A website dedicated to distributing statistical software, data sets, and information, maintained on the website of Carnegie Mellon University.

Because the site is not accessible, and the whole CMU Statistics site is not available.

Note from the Author or Editor:
p. 540 replace first bullet point with:

Consortium for the Advancement of Undergraduate Statistics Education (https://www.causeweb.org)

The website of CAUSEweb, a national organization supporting advanced undergraduate statistics education, with links to research, resources, professional development opportunities, etc., mainly of interest to teachers of statistics.

http://www.math.unb.ca/~knight/utility/
Not Found

the link maybe obsolete.

Note from the Author or Editor:
remove this bullet point: the web page no longer exists

when you look at the followings:

Qit Quantity of product i for time t (business statistics)
Pit Price of product i for time t (business statistics)
Tt Secular trend (time series)
Ct Cyclical effect (time series)
St Seasonal effect (time series)
Rt Residual or error effect trend (time series)

you would notice the subscirpt (of subscript) are not properly typeset as in the page 329, 331 and 332.

The typeset should follow in the Chapter 14.

Note from the Author or Editor:
p. 545 lines 10-14:
Qit: the "i" should be subscript and the "t" double subscript (as in Figure 14-6 p. 239)
Pit: the "i" should be subscript and the "t" double subscript (as in Figure 14-6 p. 239)
Tt: the "t" should be subscript
Ct: the "t should be subscript
St: the "t" should be subscript
Rt: the "t" should be subscript

The definition of "Interaction variable",
A variable for which the relationships between two other variables are different,
depending on the level of the interaction variable.

looks cyclic. It should be read as
A variable for which the relationships between two other variables are different,
depending on the level of this variable.

Note from the Author or Editor:
p. 547 definition for "Interaction variable" should read: "A variable for which the relationship between two other variables differs depending on the level of this variable. For instance, if the relationship of variables A and B differs depending on the level of variable C, then variable C is an interaction variable for variables A and B."

Index refers to EBCDIC, pointing to Page 422, but the referenced page does not refer to that encoding, just to Unicode and ASCII (probably updated from 1st edition).

Note from the Author or Editor:
Remove references to EBCDIC in index. There are two: p. 558 "EBCDIC" and p 559 "Extended Binary Coded Decimal Interchange Code".

In a series of examples of dividing out common factors to simplify fractions, (4x^3y^2)/(2xy^3) is said to reduce to 2xy^-1 instead of 2x^2y^-1.

Note from the Author or Editor:
the second equation in Figure A-15 (p. 487 in the pdf) should solve to 2x^2y^-1

in other words, it should be x-squared rather than x in the solution