Page 447 - Six Sigma Demystified

f i n a l e x a m 427

77. confidence intervals about the regression line are flared at the ends because
A. we can’t predict the y values at the extremes with as much certainty.
b. there are fewer data at the endpoints.
c. the error in predicting the slope causes greater uncertainty at the extremes.
d. All the above are true.
78. If the uncoded expression cycle time (in minutes) = 3 + 1.4 × (number of orders)
– 2.1 × (number of clerks) – 0.034 × (process distance),
A. the number of clerks is most important.
b. the process distance is least important.
c. the cycle time is about 5 minutes when the number of orders equals 5, the
number of clerks equals 2, and the process distance equals 14.
d. All the above are true.

2
79. A model of the form Y = b + b X + b X + b X + b X 12
1
11
1
1
0
2
12
2
A. will result in a surface that is nonlinear.
b. requires experimental trials where X is set to at least three distinct levels.
1
c. includes an interaction term.
d. All the above are true.
80. To estimate the regression model Y = b + b X + b X + b X ,
2
1
2
1
0
12
12
A. using four experimental trials allows for no estimation of error.
b. it is necessary to use experimental trials where X is set to at least three distinct
1
levels.
c. we can always use the F statistic to estimate significance of the parameters.
d. All the above are true.
81. If an ANOVA does not provide results of the F test, then
A. we probably did not run enough trials to estimate error.
b. the software must be flawed.
c. we can remove a factor with an effect close to zero to free up trials.
d. choices a and c are true.

2
82. A 3 experiment means that we are considering
A. two levels of three factors.
b. two dependent variables and three independent variables.
c. two go/no-go variables and three continuous variables.
d. three levels of two factors.
83. Which of the following purposes are served by replicating an experiment?
i. Provide a means for estimating the experimental error.
ii. Increase the number of treatments included in the experiment.
iii. Improve the precision of estimates of treatment effects.
A. Purposes i and ii only
b. Purposes i and iii only
c. Purposes ii and iii only
d. Purposes i, ii, and iii