Blatchley Chapter 15 Quiz

. Which of the following is true regarding non-parametric tests?

They require interval or ratio scales

They require homogeneous variances

They require a nominal or interval scale

They can test hypotheses

. Which of the following is the best example of ranked data?

Individuals'

Number of pets households own

The top 20 US colleges

Reaction time to a visual stimulus measured by milliseconds

.

Which score is in the 2^nd rank (higher is higher rank)?

Score
4
17
19
12
11
8

17

19

12

11

.

Using the data of test scores out of 100 below to answer the next two questions.

Score
89
81
79
77
76
77
78
79
81
82
91
64
81
89
82
83
85
85
89

How many ties are there in this data in terms of their rank?

4

3

5

6

. Rank the data. Which score is second lowest rank?

64

76

78

81

.

Using the data to answer the next four questions.

Does income correlate with happiness? A researcher collected a small sample by asking the respondent how much they earned a year and how satisfied they were with their lives (1 = extremely unsatisfied, 10 = extremely satisfied).

Income	Happiness
1,500,000	5
25,000	8
5,000,000	4
55,000	9
76,000	7
189,000	3
300,000	6

Why might a spearman correlation be appropriate for this dataset?

The data are on an interval scale

The sample is too large and on a ratio scale

The sample is too small and the distribution might not be normal

The distribution might not be normal and the data is on an ordinal scale

. What is the spearman correlation?

-.75

.75

.50

-.55

. What do the data suggest?

The results are statistically significant. Happiness is positively correlated with income

The results are statistically significant. Happiness is negatively correlated with income

The results are not statistically significant, but the direction of the correlation is positive

The results are not statistically significant, but the direction of the correlation is negative

. Given the size of the correlation, what might be influencing the outcome of this small survey?

The sample size is too small, so the results are not significant despite the size of the correlation

The sample is non-representative and therefore not statistically significant

The sample size is large enough, and therefore the results are significant

The sample is on an interval scale. If it were on a ratio scale, the results would not be significant

. The Mann-Whitney U-test is the nonparametric equivalent of which parametric test?

one-way ANOVA

independent samples t-test

dependent samples t-test

one sample t-test

. Which of the following is not an assumption of the Mann-Whitney U-test?

No strong requirement for distribution type

Interval or ratio data

Independence in observations

None of the above are assumptions that the Mann-Whitney U-test makes

. How do you determine if the Mann-Whitney U test results in a significant difference?

If your calculated U-value is smaller than that of the critical value, then the test is considered significant.

If your calculated U-value is larger than that of the critical value, than the test is considered significant.

If your W statistic is greater than the degrees of freedom, then the test is considered significant.

None of the above are correct.

.

Using this information and the data below to answer the next four questions.

Does drinking a cup of coffee in the morning enhance performance in a class? A researcher randomly assigned a class to drink either two cups of coffee or no coffee before lecture. They then ranked the students in terms of performance.

Coffee	No Coffee
1	10
5	3
7	2
6	11
9	5
8	4
14	12

Which of the following tests should be used?

Wilcoxon Signed Rank test

Dependent samples t-test

Spearman correlation

Mann-Whitney U-test

. What is the critical value for this test at α = .05, assuming two-tails?

6

7

8

9

. Calculate the Mann-Whitney U-test U-value and determine the smaller of the two values.

21.5

20

20.5

22.5

. Which of the following most accurate describes the outcome of the study?

The difference between the two groups were not statistically significant

The difference between the two groups were significant: Coffee helped with performance

The difference between the two groups were significant: Coffee did not help performance

The data is not sufficient to tell whether or not the two groups in this sample were different or not

. How does on determine if the results of a Wilcoxon signed-rank test is significant?

the observed X must be less than the critical X

the observed t must be less than the critical t

the observed t must be larger than the critical t

the observed t must be equal to the critical t

.

Using the data, below to answer the next four questions.

A marketing director is interested in examining whether individuals look longer at advertisements that “make sense” (e.g., no surprising elements) or those that don’t “make sense.” The direct recruited people to look at the advertisements that her team created and measured the number of seconds they spent viewing the ads. All participants viewed both types of advertisements.

Ads that "make sense"	Ads that don't "make sense"
12	22
16	14
9	14
8	11
5	7
3	9
12	18
10	13

In what sense are these data "matched?"

They were all about advertisements

Participants experienced every condition in the study

Participants were all from the same demographic

All the participants chose to participate

. For a Wilcoxon Signed-Rank test, what is the critical value for this test, assuming two-tailed α = .05?

1

2

3

4

. For a Wilcoxon Signed-Rank test, which of the following is the correct test statistic for the sample?

1

1.5

2

2.5

. Which of the following best describes the finding?

the two groups did not statistically differ in time spent looking at the ad

the two groups were statistically different in their time spent looking at the ad

the two groups were statistically equivalent in time spent looking at the ad

there is insufficient data to reach a conclusions about the sample

. Which of the following is true about the Wilcoxon Signed-Rank test?

It can only be used on ordinal data

It can be used on data that is skewed, even if the data is on an interval or ratio scale

It can be used on data that is skewed, but only if it is on an interval or ratio scale

It can only be used when the data is not normally distributed

. If you have very skewed data which of the following is least appropriate?

ranking the data and considering a nonparametric instead of parametric test

ignoring the skew and analyzing the data with a parametric test

transforming the data or considering an alternate method to analyze it

ranking the data and using a parametric test instead of a nonparametric test

. Which of the following are true regarding skewed distribution?

There is only one kind of skewed distribution, and it is always bad.

There are two types of skewed distribution: positive and negative.

There are two types of skewed distributions, in one case, parametric tests are still suitable and in the other, only nonparametric tests are suitable.

There is only one kind of skewed distribution, and both parametric and nonparametric tests are suitable.

. With a skewed distribution, it is generally the case that...

One tail of the distribution extends longer than the other tail.

The distribution is symmetric, but there is more than one mode.

Only parametric tests are suitable for data analysis.

The sample cannot be generalized to the population.

Blatchley Chapter 15 Quiz

Quiz Content