Question 1

Which of the following are assumptions underlying the use of most parametric tests (based on the normal distribution)?

Accepted Answer

A)   The data should be normally distributed.
B)   The samples being tested should have approximately equal variances.
C)   The data should be (at least)  interval level.
D)   All of the above. D

Question 2

What does independence of data mean?

Accepted Answer

A)   That we must never collect two sets of data from one person
B)   That independent researchers must collect the data
C)   That scores from one participant are free from influences from other participants
D)   That scores in one condition are free from influences from other conditions D

Question 3

Which of the following is not a transformation that can be used to correct skewed data?

Accepted Answer

A)   Log transformation
B)   Tangent transformation
C)   Square root transformation
D)   Reciprocal transformation B

Question 4

If the distribution of CLV for your two products (Q8) is multimodal, what does this mean?

Accepted Answer

A)   It doesn't have a normal distribution.
B)   The data have been entered incorrectly.
C)   It will be a normal distribution.
D)   It will have to be checked with a Levene's test.

Question 5

The Kolmogorov-Smirnov test can be used to test?

Accepted Answer

A)   Whether data are normally distributed.
B)   Whether group variance are equal.
C)   Whether scores are measured at the interval level.
D)   Whether groups means differ.

Question 6

Which of the following would be the best way to decide whether the skew in the example above is problematic for analysts at 'Wipe-It'?

Accepted Answer

A)   See if the z-score is bigger than 1.96 or smaller than -1.96.
B)   See if the skew is significant at p 
C)   Use the Kolmogorov-Smirnov test.
D)   None of the above, because of the large sample size.

Question 7

What is an outlier?

Accepted Answer

A)   A set of data outside the data file.
B)   A single score (e.g., participant response)  that is very different from others.
C)   A score derived from a participant who has lied.
D)   A variable that cannot be quantified.

Question 8

Which of the following tests whether variances are homogenous?

Accepted Answer

A)   Levene's test
B)   Bartlett's test
C)   Neither a nor b
D)   Both a and b

Question 9

You have income data for your entire customer database. Some of your clients are far wealthier than others. Which of the following do outliers least affect?

Accepted Answer

A)   The range
B)   The mean
C)   The median
D)   The standard deviation

Question 10

If a Kolmogorov-Smirnov test is conducted and the result is significant, what does this mean for the data sample?

Accepted Answer

A)   The data sample is normally distributed.
B)   The comparison used in the test is not valid.
C)   The data sample is not normally distributed.
D)   The test is wrong.

Question 11

You need to plot values of one continuous variable against two others, whilst also differentiating groups of cases with different-coloured dots. What type of scatterplot should you use?

Accepted Answer

A)   Simple 3-D scatterplot
B)   Grouped 3-D scatterplot
C)   Scatterplot matrix
D)   Grouped scatterplot

Question 12

You are comparing customer lifetime value (CLV) in £s for two of your firm's products - Mug-R and Jug-S. What does acceptance of the assumption of homogeneity mean for the two groups in this context?

Accepted Answer

A)   The variance is twice as big for Mug-R than Jug-S.
B)   Variances in the groups are approximately equal.
C)   The variance across the groups is proportional to the means of those groups
D)   The variance for both Mug-R and Jug-S is equal to the interquartile range

Question 13

Which of the following is not defined as a property for a variance ratio?

Accepted Answer

A)   It can be used to demonstrate homogeneity of variances.
B)   It is one variance divided by another.
C)   It is one variance multiplied by another.
D)   It can show the effect of a treatment on several groups.

Question 14

Some data were collected about consumer loyalty towards a (relatively) new brand of toilet roll - 'Wipe-It' (using a scale from 1 = I'd never go back to 10 = I will always purchase this brand). The sample was 15,467 people. When looking at the distribution, a skew of 1.23 (SE = 0.65) was identified. The mean rating was 4.78. What is the z-score for the skew in this data?

Accepted Answer

A)   1.89
B)   0.53
C)   -3.92
D)   3.36

Question 15

Which of the following does a box-whisker plot not display?

Accepted Answer

A)   The range
B)   The interquartile range
C)   The lower quartile
D)   The mean

Question 16

Why are z-scores used to check for outliers?

Accepted Answer

A)   They standardize scores for a known mean and standard deviation, allowing comparison.
B)   They allow you to allocate letters for missing values.
C)   A z-score is an outlier.
D)   They standardize scores in order to convert them to values closer to the mean.

Exam 5: Exploring Data With Graphs

Which of the following are assumptions underlying the use of most parametric tests (based on the normal distribution) ?