Question 1

Levene's test tests whether?

Accepted Answer

A)   Data are normally distributed
B)   The variances in different groups are equal
C)   The assumption of sphericity has been met
D)   Group means differ B

Question 2

The total variation in response, assuming no bias, is due to error (unexplained variation) plus any differences due to treatments (known variation). If known variation is large compared to unknown variation, which of the following conclusions can be drawn?

Accepted Answer

A)   There is no difference in response due to treatments.
B)   There is a difference in response due to treatments.
C)   The treatments are not comparable.
D)   The cause fo the response is due to something other than treatements. B

Question 3

What kind of trend does the following graph show?

Accepted Answer

A)   Linear
B)   Quadratic
C)   Cubic
D)   Quartic B

Question 4

A Bonferroni correction is when?

Accepted Answer

A)   You apply a criterion for significance based on the usual criterion for significance (.05)  divided by the number of tests performed.
B)   You divide the F-ratio by the number of tests performed.
C)   The degrees of freedom are corrected to make the F-ratio less significant.
D)   The error in the model is adjusted for the number of tests performed

Question 5

In the above table what is the likely interpretation for whether any difference in net promoter score exists between the defined groups?

Accepted Answer

A)   The alternative hypothesis is rejected.
B)   The null hypothesis is rejected.
C)   Not enough information is provided.
D)   The null hypothesis is accepted

Question 6

What is the definition of 'mean square'?

Accepted Answer

A)   A sum of squares divided by its degrees of freedom
B)   The square root of the mean.
C)   The square of the mean.
D)   A table of means with four cells

Question 7

Do coffee products from China, India or South Africa take more or less time to ship to the UK? An ANOVA was employed to test this question. What is the missing value (??) for F in the table below?

Accepted Answer

A)   0.74
B)   16.62
C)   4.71
D)   2.71

Question 8

If the critical region for a hypothesis test is F > 9.48773 and the computed value of F from the data is 0.86, which of the following is correct?

Accepted Answer

A)   The null hypothesis should be rejected.
B)   The alternative hypothesis is two-tailed.
C)   The significance level is given by the area to the right of 9.48773 under the appropriate F-distribution.
D)   The significance level is given by the area to the left of 9.48773 under the appropriate F-distribution

Question 9

Consider this table of customer net promoter scores for the iPad 4:
Analysis of variance
 
-If all samples sizes are equal then the study used.

Accepted Answer

A)   4 samples of size 10
B)   5 samples of size 10
C)   4 samples of size 9
D)   5 sample of size 9

Question 10

The F-ratio of 15.3 in the above table is calculated by?

Accepted Answer

A)   Dividing the model mean squares by the residual mean squares.
B)   Taking the sum of residual mean squares and model mean squares
C)   Dividing the residual mean squares by the model mean squares
D)   Multiplying the model mean squares and the residual mean squares

Question 11

A training consultant wanted to see the effects of different learning strategies for teaching management accounting skills to employees. A control group read the book Accounting for Managers (book), a second group read the book and completed the 'end of chapter exercises' (book and exercises), and a third group read the book, did the end of chapter examples and also completed the web materials (all activities). The trainer predicted that the 'all activities' and 'book and exercises' groups would perform better than the 'book' group on a subsequent test, but that the 'book and exercises' group would be worse than the 'all activities' group. Which coding scheme would test these hypotheses in a set of planned comparisons?

Accepted Answer

A)   
B)   
C)   
D)

Question 12

Using the same data, post hoc tests were run. Based on the output for these tests, what should the HR manager conclude?

Accepted Answer

A)   The scheme group was significantly different from the post-scheme and control groups, and the post-scheme group was significantly different from the control group.
B)   The scheme group was significantly different from the control group but not the post-scheme group, and the post-scheme group was not significantly different from the control group.
C)   The scheme group was significantly different from the control group but not the post-scheme group, and the post-scheme group was significantly different from the control group.
D)   The post hoc tests are inconclusive.

Question 13

A follow-up experiment to better understand net promoter scores was conducted using a one-way factorial design with K sample means, each based on n scores. What is the number of degrees of freedom for the between mean squares?

Accepted Answer

A)   n -1
B)   K- 1
C)   n -K
D)   (n - 1) (K - 1)

Question 14

An HR manager was looking at the effects of an employee involvement scheme on commitment levels of the organization's staff. Three groups were used: no scheme applied, experiencing the scheme and post scheme (a group who had been subjected to the scheme 6 months before). The change in commitment over the time that the scheme was recorded is the dependent variable (although bear in mind that only the treatment group actually got any treatment during this time). The SPSS output is below.

Based on this output what should the researcher conclude?

Accepted Answer

A)   The employee involvement scheme groups did not have a significant effect on any change in commitment levels, F(2, 26.44)  = 4.35
B)   The employee involvement scheme groups had a significant effect on the change in commitment levels, F(2, 35.10)  = 5.1.
C)   The employee involvement scheme groups had a significant effect on the change in commitment levels, F(2, 45)  = 5.11.
D)   The employee involvement scheme groups did not have a significant effect on the change in commitment levels, F(2, 45)  = 5.11.

Question 15

When variances across groups are unequal, which of the following is not an appropriate course of action?

Accepted Answer

A)   Welch's F-ratio
B)   Games-Howell post hoc tests
C)   Transforming the data
D)   Friedman's ANOVA

Question 16

In a simple analysis of variance problem, which of the following is an estimate of the variance of individual measurements (after the various effects have been accounted for)?

Accepted Answer

A)  MS (between) 
B)   MS (within) 
C)   MS (total) 
D)   None of the above

Exam 12: Comparing Several Means