Deck 13: Matching Statistical Tests to Business Problems

Question 1

To conduct any of the statistical tests covered in this text, each case must be ______.

A) independent
B) dependent
C) robust
D) none of these

Answer 1

A

Question 2

If you wanted to conduct a paired-samples t-test, but your outcome data were ordinal, a ______ might be more appropriate.

A) Kruskal-Wallis test
B) Wilcoxon signed-rank test
C) Spearman's rho
D) McNemar's test

Answer 2

B

Question 3

If you want to conduct regression but your data contains dependences, ______ might be more appropriate.

A) hierarchical linear modelling
B) nonparametric regression
C) repeated-measured ANOVA
D) McNemar's test

Answer 3

A

Question 4

The tests you've learned in this text book require ______ data.

A) normally distributed, dependent
B) skewed, dependent
C) skewed, independent
D) normally distributed, independent

Question 5

Outcome variables are also called ______ variables.

A) ordinal
B) interval
C) dependent
D) independent

Question 6

If a dependent variable is measured twice, it is considered a ______ dependent variable.

A) repeated measures
B) paired
C) repeated
D) all of these

Question 7

If we expect increased product costs to reduce sales, product costs would be the ______.

A) paired variable
B) outcome variable
C) independent variable
D) dependent variable

Question 8

With one interval-level DV and no IVs, you would most likely need to conduct a ______.

A) one-sample t-test or z-test
B) one-way ANOVA
C) χ² goodness-of-fit test
D) correlational analysis

Question 9

With one ordinal-level IV, you would most likely need to conduct a ______.

A) one-sample t-test or z-test
B) one-way ANOVA
C) χ² goodness-of-fit test
D) none of these

Question 10

If you're interested in investigating the interaction between two nominal variables, you would most likely need to conduct a ______.

A) χ² test of independence
B) one-way ANOVA
C) χ² goodness-of-fit test
D) none of these

Question 11

With one nominal-level outcome variable, you would most likely need to conduct a ______.

A) one-way ANOVA
B) χ² goodness-of-fit test
C) regression analysis
D) one-sample t-test or z-test

Question 12

With one three-level nominal variable and one interval-level outcome variable, you would most likely need to conduct a ______.

A) paired-samples t-test
B) independent-samples t-test
C) z-test or one-sample t-test
D) one-way ANOVA

Question 13

To determine the difference between men and women on a ratio-level DV, you would most likely need to conduct a ______.

A) correlation analyses
B) independent-samples t-test
C) χ² test of independence
D) z-test or one-sample t-test

Question 14

If you're interested in determining the strength of the relationship between sales (in pounds) and Likert-type survey scores, you would most likely need to conduct ______.

A) one-way ANOVA
B) χ² test of independence
C) correlation analyses
D) z-test or one-sample t-test

Question 15

If you wanted to see if customer satisfaction survey scores were different between morning shoppers, mid-day shoppers, and night shoppers, you would most likely need to conduct a ______.

A) paired-samples t-test
B) one-way ANOVA
C) χ² goodness of fit test
D) regression analyses

Question 16

Over the past year, you've given employees the choice each month of whether to have a pizza party or a game night to celebrate high sales figures. If you wanted to see if employees picked one more often than the other, you would most likely need to conduct a ______.

A) independent-samples t-test
B) correlation analyses
C) χ² test of independence
D) χ² goodness-of-fit test

Question 17

Correlation can be used to explore the relationship between ______.

A) two DVs
B) one IV and one DV
C) both a and b
D) neither of these

Question 18

Based upon the decision process given in this text, there are special requirements for ______.

A) z-test
B) one-sample t-test
C) independent-samples t-test
D) all of these

Question 19

Which of the following is an example of exploratory data analysis?

A) data mining
B) confirmatory analysis
C) hypothesis testing
D) none of these

Question 20

Clean data should ______.

A) contain independent cases
B) lack missing data
C) contain all needed variables
D) all of these

Question 21

A code sheet is used to ______.

A) clean data
B) provide context to data
C) enable confirmatory analyses
D) all of these

Question 22

In a dataset containing 1 IV and 1 DV, if the middle 50% of values on the DV are missing from the sample, ______ will occur, which artificially ______ the size of correlations

A) range enhancement, increases
B) range enhancement, decreases
C) range restriction, increases
D) range restriction, decreases

Question 23

In a dataset containing 2 DVs, if the upper 30% of values on the DV are missing from the sample, ______ will occur, which artificially ______ the size of correlations

A) range enhancement, increases
B) range enhancement, decreases
C) range restriction, increases
D) range restriction, decreases

Question 24

Under what conditions can a correlation coefficient indicate causation?

A) if a t-test is run to test the significance of the correlation
B) if a coefficient of determination is calculated
C) if an experiment is used to manipulate the IV
D) A correlation can never indicate causation.

Question 25

The coefficient of determination alone can be used to assess ______.

A) statistical significance
B) practical significance
C) both a and b
D) neither of these

Question 26

With an independent-sample t-test, we identify a significant difference between men and women on job satisfaction. This means that the difference between men and women is ______.

A) real
B) important
C) practically significant
D) none of these

Question 27

With an independent-sample t-test, we identify that women have significantly higher job satisfaction than men. If we were to apply this to conclude that Employee A (who is female) has higher job satisfaction than Employee B (who is male), we have ______.

A) made a correct decision
B) committed an ecological fallacy
C) committed a Type I error
D) committed a Type II error

Question 28

Wicked problems ______.

A) are relatively infrequent
B) can be addressed with careful statistical analyses
C) are difficult to define and answer
D) all of these

Question 29

Statistics alone provide the final answer to ______ meaningful questions.

A) no
B) some
C) most
D) all

Question 30

Effect sizes can be ______.

A) over-interpreted by ignoring practical significance
B) under-interpreted by committing the ecological fallacy
C) inappropriately ignored by relying on statistical significance
D) all of these

Question 31

If your data are not independent, you can always fix this by adding dependent values together.

Question 32

By specifying an independent and dependent variable, we can make conclusions about causality.

Question 33

If you wanted to use an independent-samples t-test and your data were not normally distributed, you might use a Mann-Whitney U.

Question 34

If you wanted to use a one-way ANOVA and your data were not normally distributed, you might use Spearman's rho.

Question 35

If you wanted to examine the relationship between two ratio variables, you could use correlation.

Question 36

If you wanted to examine the difference between morning shift and evening shift on a job satisfaction survey, you could use a χ² goodness of fit.

Question 37

If you wanted to predict sales numbers from the number of sick days employees have taken, you could use regression.

Question 38

If you want to compare your sample mean to a population mean, you should always use a z-test.

Question 39

If you want to know "How much does training affect customer satisfaction?", you are interested in statistical significance.

Question 40

Data mining is your primary tool for analysing data.

Question 41

A code sheet is used to explain the meaning of variables in a dataset.

Question 42

Conducting analyses when statistical assumptions are violated can invalidate the results of the analysis.

Question 43

A statistician's personal biases can lead to misinterpretation of results.

Question 44

An independent-samples t-test could also be conducted by converting the IV into 1's and 0's and conducting a regression analysis.

Question 45

The eight-step hypothesis testing process used in this book is an example of an exploratory approach to data analysis.

Question 46

There is one "right" way to approach analysis to any particular research question.

Question 47

Immediately after collecting data is the first point at which you should consider what data analytic approach you will take.

Question 48

Range enhancement results in artificially high correlations.

Question 49

The coefficient of determination is used as a measure of practical significance.

Question 50

What is the difference between exploratory and confirmatory data analytic approaches?

Question 51

What questions should you ask about the variables available in your dataset to determine what statistical tests are appropriate?

Question 52

Data should always be clean before analysis. How do you clean data?

Question 53

What is the difference between range restriction and range enhancement? What is the effect on a correlation when each occurs?

Question 54

Is there a relationship between Sales at Time 1 and Time 2? Complete the full hypothesis testing process as described in Chapter 7 to examine this question and draw all appropriate conclusions.

Question 55

You would like to compare employees in each of the four departments of your organization: the sales department, the service department, the warehouse, and management. Specifically, you want to know if training costs are different due to Department. Complete the full hypothesis testing process as described in Chapter 7 to examine this question and draw all appropriate conclusions.

Question 56

In the population, we'd expect Days Without a Complaint to have a mean of 24 and standard deviation of 1, but you want to know if your sample comes from a different population with a higher mean. Complete the full hypothesis testing process as described in Chapter 7 and draw all appropriate conclusions.

Question 57

Do job satisfaction survey scores change over time? Complete the full hypothesis testing process as described in Chapter 7 to examine this question and draw all appropriate conclusions.