Question 1

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'?&#10;A) See if the z-score is bigger than 1.96 or smaller than -1.96.&#10;B) See if the skew is significant at p < .05.&#10;C) Use the Kolmogorov-Smirnov test.&#10;D) None of the above, because of the large sample size.

Accepted Answer

The large sample size at 'Wipe-It' would likely render the skew less problematic for analysis, as many statistical methods become more robust with larger samples. The specific methods listed (A, B, C) are not universally the best ways to decide on the problematic nature of skew without considering the context of the sample size and the analysis being performed.

Question 2

The Kolmogorov-Smirnov test can be used to test?&#10;A) Whether data are normally distributed.&#10;B) Whether group variance are equal.&#10;C) Whether scores are measured at the interval level.&#10;D) Whether groups means differ.

Accepted Answer

The Kolmogorov-Smirnov test is a non-parametric test used to determine if a sample comes from a population with a specific distribution. It is commonly used to test whether data are normally distributed.

Question 3

Which of the following are assumptions underlying the use of most parametric tests (based on the normal distribution)?&#10;A) The data should be normally distributed.&#10;B) The samples being tested should have approximately equal variances.&#10;C) The data should be (at least) interval level.&#10;D) All of the above.

Accepted Answer

All three options are assumptions that underlie the use of most parametric tests based on the normal distribution.

A) Normality assumption: If the data is not normally distributed, then the results of the statistical tests may be misleading. 
B) Equal variance assumption: Parametric tests assume that the variance of the population being sampled is equal across the groups being compared. If this assumption is violated, then the results may not be reliable. 
C) Interval level assumption: Most parametric tests require interval level data, which means that the intervals between values are equal. If the data is not at least interval level, then parametric tests may not be appropriate.

Question 4

Which of the following does a box-whisker plot not display?&#10;A) The range&#10;B) The interquartile range&#10;C) The lower quartile&#10;D) The mean

Accepted Answer

A box-whisker plot does not display the mean. It only shows the minimum value, maximum value, lower quartile, upper quartile, and median of a set of data.

Question 5

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?&#10;A) Simple 3-D scatterplot&#10;B) Grouped 3-D scatterplot&#10;C) Scatterplot matrix&#10;D) Grouped scatterplot

Accepted Answer

A grouped 3-D scatterplot allows plotting values of one continuous variable against two others and differentiates groups of cases with different-coloured dots, effectively meeting the requirements described.

Question 6

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?

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

Accepted Answer

The z-score for skewness is calculated by dividing the skewness by its standard error: 1.23 / 0.65 = 1.89.

Question 7

You are comparing customer lifetime value (CLV) in &#163;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?&#10;A) The variance is twice as big for Mug-R than Jug-S.&#10;B) Variances in the groups are approximately equal.&#10;C) The variance across the groups is proportional to the means of those groups&#10;D) The variance for both Mug-R and Jug-S is equal to the interquartile range

Accepted Answer

Acceptance of the assumption of homogeneity means that the variances in the groups are approximately equal. In this context, it means that the CLV variance for Mug-R is similar to the CLV variance for Jug-S.

Question 8

If a Kolmogorov-Smirnov test is conducted and the result is significant, what does this mean for the data sample?&#10;A) The data sample is normally distributed.&#10;B) The comparison used in the test is not valid.&#10;C) The data sample is not normally distributed.&#10;D) The test is wrong.

Accepted Answer

If the Kolmogorov-Smirnov test results are significant, it indicates that the data sample is not normally distributed. Normality assumption is one of the underlying assumptions of many statistical tests such as t-test, ANOVA etc. Violation of the normality assumption can affect the reliability of the test results. Therefore, it is important to check the normality assumption before applying some statistical tests.

Question 9

What does independence of data mean?&#10;A) That we must never collect two sets of data from one person&#10;B) That independent researchers must collect the data&#10;C) That scores from one participant are free from influences from other participants&#10;D) That scores in one condition are free from influences from other conditions

Accepted Answer

The answer of What does independence of data mean?&#10;A) That...

Question 10

What is an outlier?&#10;A) A set of data outside the data file.&#10;B) A single score (e.g., participant response) that is very different from others.&#10;C) A score derived from a participant who has lied.&#10;D) A variable that cannot be quantified.

Accepted Answer

An outlier is a single score that is very different from the other scores in a data set. It can be either unusually high or unusually low and is often a result of measurement error or some other kind of anomalous event.

Question 11

Which of the following tests whether variances are homogenous?&#10;A) Levene's test&#10;B) Bartlett's test&#10;C) Neither a nor b&#10;D) Both a and b

Accepted Answer

The answer of Which of the following tests whether variances...

Question 12

Which of the following is not defined as a property for a variance ratio?&#10;A) It can be used to demonstrate homogeneity of variances.&#10;B) It is one variance divided by another.&#10;C) It is one variance multiplied by another.&#10;D) It can show the effect of a treatment on several groups.

Accepted Answer

The answer of Which of the following is not defined...

Question 13

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?&#10;A) The range&#10;B) The mean&#10;C) The median&#10;D) The standard deviation

Accepted Answer

The answer of You have income data for your entire...

Question 14

Why are z-scores used to check for outliers?&#10;A) They standardize scores for a known mean and standard deviation, allowing comparison.&#10;B) They allow you to allocate letters for missing values.&#10;C) A z-score is an outlier.&#10;D) They standardize scores in order to convert them to values closer to the mean.

Accepted Answer

The answer of Why are z-scores used to check for...

Question 15

Which of the following is not a transformation that can be used to correct skewed data?&#10;A) Log transformation&#10;B) Tangent transformation&#10;C) Square root transformation&#10;D) Reciprocal transformation

Accepted Answer

The answer of Which of the following is not a...

Question 16

If the distribution of CLV for your two products (Q8) is multimodal, what does this mean?&#10;A) It doesn't have a normal distribution.&#10;B) The data have been entered incorrectly.&#10;C) It will be a normal distribution.&#10;D) It will have to be checked with a Levene's test.

Accepted Answer

The answer of If the distribution of CLV for your...

Deck 5: Exploring Data With Graphs