BooksAnderson and Whitcomb · 2007
DOE Simplified: Practical Tools for Effective Experimentation
- Matches
- Very close
- Close
- Differs
Factorial gives the book's answers. Where they differ, this page says why: some are misprints in the book.
Examples
Each example, and how closely Factorial gives its answers.
Pages 1 to 99
- Total of a pair of dice in 50 rolls: mean, variance, standard deviation, standard error and a 95% intervalExample · pp. 7-15Matches
- 56 dice of four colours tossed once: does the colour change the average roll? A one-way F-testExample · pp. 24-28Close
- The same 56 dice after two cheaters turned blue and green up a dot and white and purple down: F-test and LSDExample · pp. 29-32Matches
- Seconds for four preschool children to stack three dice of 1 to 6 dots: blocking out the childrenExample · pp. 32-37Close
- Six games each of three bowlers: are their scores different? (Problem 2-1)Exercise · pp. 37-38Matches
- Microwave popcorn: brand, time and power in a 2^3 factorial, with taste and the weight of unpopped kernelsExample · pp. 43-74, 157-159Very close
- Tabletop hockey: puck weight, stick length, wind-up and puck place in a 2^4, the distance analysed in logsExample · pp. 75-86Misprint found
- The recommended two-level screening fraction for 4 factors, 12 runsDesign · p. 91Matches
- The recommended two-level screening fraction for 5 factors, 16 runsDesign · p. 91Matches
- The recommended two-level screening fraction for 6 factors, 32 runsDesign · p. 91Matches
- The recommended two-level screening fraction for 7 factors, 32 runsDesign · p. 91Differs
- Pulls to start a weed whacker: prime pumps, choke pulls, gas and choke settings in a 2^(5-1) half fractionExample · pp. 92-97Matches
- The popcorn 2^3 cut to the half fraction ABC = +: a resolution III design and its aliasesExample · pp. 97-99Matches
Pages 100 to 199
- The 2^(4-1) half fraction with ABCD = +: every two-factor interaction aliased with anotherDesign · pp. 100-101Matches
- Seven factors in 32 runs, the 2^(7-2): DE, DF and DG aliased with FG, EG and EFDesign · pp. 101, 126, 200-202Matches
- Resolutions of the standard two-level fractions for 3 to 7 factors in 4 to 64 runsDesign · p. 102Matches
- Seven factors in 16 runs, the 2^(7-3): its aliases among two-factor interactionsDesign · pp. 102-103Matches
- Readability of a projected screen: font size, font style, background and room lighting in a 12-run 3/4 fractionExample · pp. 104-110Matches
- Four factors in 12 runs, the irregular 3/4 fraction of the 2^4, and its aliasesDesign · pp. 104, 194-195Matches
- Part shrinkage of an injection molder: seven factors in a 2^(7-2) run on two production lines as blocks (Problem 5-1)Exercise · pp. 110-113Matches
- Dancing raisins: seven factors in a 2^(7-4) and its complete fold-over in a second blockExample · pp. 116-125Matches
- Yield density of a fermentation: six factors screened in a 12-run minimum-run resolution IV (MR4) designExample · pp. 129-130Matches
- A minimum-run resolution V (MR5) design for six factors in 22 runsDesign · pp. 130-132Matches
- Time for three spring toys to walk down a plank at two inclines: a 3 x 2 general factorial, replicatedExample · pp. 134-138Matches
- Four spring toys, two inclines and a child or adult operator: an unreplicated 4 x 2 x 2 general factorialExample · pp. 138-146Matches
- Flight time of paper confetti: a 2^2 with four centre points shows curvature, then augmented to a CCDExample · pp. 148-156Matches
- Melt point of gold and copper blends: a replicated two-component mixture designExample · pp. 162-165Very close
- Taste of jelly candy blends of apple, cinnamon and lemon: a simplex centroid with replicatesExample · pp. 166-169Matches
- Power to detect main effects of five factors in 8 runs (resolution III)Design · pp. 174-178Matches
- Power to detect main effects of five factors in 16 runs (resolution V)Design · pp. 174-178Very close
- Power to detect main effects of five factors in 32 runs (resolution Full)Design · pp. 174-178Very close
- Five factors in 16 runs, the 2^(5-1) with E = ABCDDesign · pp. 196-197Matches
- Six factors in 32 runs, the 2^(6-1) with F = ABCDEDesign · pp. 198-199Matches
Misprints
Misprints found in the book, by page. Each shows what the book prints and what its own data give.
Anova p: B·Cp. 61
Book 0.0006Data 0.0060
F = 28.44 on 1 and 4 degrees of freedom has p = 0.0060; the 0.0006 printed is the model's
Transformed observation: run 11p. 77
Book 1.658Data 1.676
The book prints 1.658 for log10(47.4) = 1.676
Effect: A·C·Dp. 78
Book -0.088Data -0.013
Table 4-3's log column swaps ACD and ABD: its -0.088 is the data's ABD
Effect: A·B·Dp. 78
Book -0.013Data -0.088
Table 4-3's log column swaps ACD and ABD: its -0.013 is the data's ACD
Not yet
Answers the book prints that Factorial cannot compute yet, by what it lacks.
- The library works the alias matrix against the two-factor interactions (and squares) left out of the model onlyp. 195
- A plotting position of the data, 100(i - 1/2)/n for the i-th smallest of n = 10; the library places a fit's effects at these positions and its residuals at scipy's (Filliben's), but has no probability plot of a sample's own valuesp. 19
- A plotting position of the residuals, 100(i - 1/2)/8 for the i-th smallest; the library's normal probability plot of residuals places them at scipy's (Filliben's) positions instead, and no check reads a plotpp. 65 and 66
- The library makes no minimum-run design, so it has none to work the power ofp. 174
- The library makes no minimum-run (MR4 or MR5) designpp. 126 and 132
Show the rest
- The library has no mode, which the tally of Figure 1-3 shows at a glancep. 7
- An irregular 3/4 fraction of the 2^4; the library makes regular fractions only, the 8-run half and the 16-run wholep. 91