Deck 18: Determining and Interpreting Associations Among Variables

Question 1

You do not have a "relationship" that links the labels (or levels)for two variables unless the relationship is:

A) causal and consistent
B) consistent and systematic
C) systematic and causal
D) systematic and important
E) consistent and important

Answer 1

B

Question 2

The product life-cycle curve that describes the sales pattern of a new product over time (growing slowly during its introduction and then spurting upward rapidly during its growth stage and finally plateauing or slowing down considerably as the market becomes saturated)is an example of which type of relationship?

A) causal relationship
B) linear relationship
C) nonmonotonic relationship
D) monotonic relationship
E) curvilinear relationship

Answer 2

E

Question 3

We know that McDonald's customers drink coffee for breakfast and soft drinks at lunch. This is an example of what type of relationship?

A) causal relationship
B) linear relationship
C) monotonic relationship
D) nonmonotonic relationship
E) curvilinear relationship

Answer 3

D

Question 4

What type of analysis would be used to answer a question such as "Which customer demographic is most strongly related to product purchase/nonpurchase"?

A) association or relationship analysis
B) predictive analysis
C) predictive or relationship analysis
D) analysis of variance
E) canine/feline regression analysis

Question 5

Which type of relationship is described by relationships that may be S-shaped or J-shaped?

A) causal relationship
B) linear relationship
C) curvilinear relationship
D) nonmonotonic relationship
E) irregular relationship

Question 6

Which type of relationship would you have when you have a general direction assigned to the relationship; that is, as one variable increases, the other variable may increase (or decrease)?

A) causal relationship
B) duotonic relationship
C) monotonic relationship
D) nonmonotonic relationship
E) curvilinear relationship

Question 7

Strength of association might be described as:

A) strong
B) weak
C) moderate
D) all of the above
E) none of the above

Question 8

Which type of relationship is a "straight-line" relationship between two variables and for which allows us to know the knowledge of one variable if we have knowledge of the other?

A) causal relationship
B) linear relationship
C) monotonic relationship
D) nonmonotonic relationship
E) curvilinear relationship

Question 9

A relationship in which the presence (or absence)of one variable is systematically associated with the presence (or absence)of another is:

A) causal relationship
B) linear relationship
C) monotonic relationship
D) nonmonotonic relationship
E) alinear relationship

Question 10

Associative analyses determine whether stable relationships exist between:

A) costs and expenses
B) two variables
C) 12 or more variables
D) marketing and sales
E) statistics and results

Question 11

The owner of a shoe store knows that as children increase in age, their shoe size tends to get larger. This is an example of what type of relationship?

A) causal relationship
B) duotonic relationship
C) monotonic relationship
D) nonmonotonic relationship
E) linear relationship

Question 12

The four basic types of relationships between two variables are:

A) anonmonotonic, duotonic, linear, and curvilinear
B) nonmonotonic, duotonic, sublinear, and curvilinear
C) nonmonotonic, monotonic, linear, and curvilinear
D) causal, consistent, systematic, and linear
E) duotonic, linear, sublinear, and alinear

Question 13

Which of the following refers to the finding that a systematic relationship exists between the two variables of interest in the population?

A) presence
B) direction
C) strength
D) pattern
E) concreteness

Question 14

Associative analysis procedures are useful because they determine if there is a consistent and systematic relationship between the presence (label)or amount of one variable and the:

A) nonpresence of a regressive relationship
B) presence (label) or amount of another variable
C) presence (label) or amount of the same variable
D) covariance of the other variable
E) presence of unobservable variables

Question 15

Which of the following were discussed by the authors as being ways that we may characterize relationships?

A) presence, direction, concreteness
B) direction, strength, concreteness
C) presence, direction, strength
D) presence, strength, concreteness
E) presence, pattern, concreteness

Question 16

Pontiac wants to know what types of persons respond favorably to proposed style changes in the Firebird. Frito-Lay wants to know what kinds of people buy from the Frito-Lay line. These are questions that may be answered through:

A) relationship analysis
B) chi-square analysis
C) associative analysis
D) analysis of variance
E) regression analysis

Question 17

When a scale has "labels" as opposed to "levels," we can normally assume the level of measurement is:

A) nominal or categorical
B) ratio
C) interval or ordinal
D) interval or ratio
E) metric

Question 18

Which type of relationship is described by the formula: y = a + bx?

A) causal relationship
B) linear relationship
C) monotonic relationship
D) nonmonotonic relationship
E) curvilinear relationship

Question 19

Which of the following is used when describing the general pattern of nonmonotonic relationships?

A) presence
B) direction
C) strength
D) pattern
E) concreteness

Question 20

When a consistent and systematic relationship is found between two variables, the linkage is:

A) causal
B) statistical
C) managerially significant
D) important
E) relevant

Question 21

In the textbook you were given an example of running a chi-square test using SPSS. The output shows a "Pearson Chi-Square" value of 23.272 and df =7. This information alone means:

A) there is a significant difference
B) there is no significant difference
C) the difference is associative
D) the test was run incorrectly
E) none of the above

Question 22

A correlation coefficient is an index number constrained to fall between the range of:

A) 0 and 1.00
B) 0 and 100
C) -1.00 and +1.00
D) -1.00 and 0
E) -100 and 0

Question 23

The chi-square test is useful for determining:

A) if a nonmonotonic relationship exists between two nominal-scaled variables
B) if a monotonic relationship exists between two nominal-scaled variables
C) if a nonmonotonic relationship exists between two interval-scaled variables
D) if a duotonic relationship exists between two variables
E) if a duotonic relationship exists between three variables

Question 24

In the textbook you were given an example of running a chi-square test using SPSS. The output shows a "Pearson Chi-Square" value of 23.272, df = 7 and the Asymp. Sig. = .002. This means:

A) there is a significant association
B) there is no significant association
C) the difference is associative
D) the test was run incorrectly
E) none of the above

Question 25

The logic of the chi-square test would argue that, for a significant relationship to exist:

A) there should be large differences between the observed and expected frequencies
B) there should be few differences between the observed and expected frequencies
C) there should be no differences between the observed and expected frequencies
D) there should be negative differences between the observed and expected frequencies
E) there should be only one difference between the observed and expected frequencies

Question 26

In chi-square analysis, the greater the number of cells, the larger the degrees of freedom. The greater the number of cells, the more opportunity exists to calculate a large chi-square value. In other words, the chi-square value can be "inflated" not due to a real association but simply due to the fact that there are more cells in the analysis. This is why degrees of freedom are used to:

A) determine how many cells you should analyze
B) determine whether or not the computed chi-square value should be used for a post hoc test
C) determine whether or not the chi-square value has a probability high enough to support, or not support, the null hypothesis
D) all of the above
E) None of the above; degrees of freedom is not calculated with the chi-square test.

Question 27

In the textbook you were given an example of running a chi-square test using SPSS. The output shows a "Pearson Chi-Square" value of 23.272. This value alone means:

A) there is a significant difference
B) there is no significant difference
C) the difference is associative
D) the difference must immediately be rounded up
E) none of the above

Question 28

Let's assume we find in a study that the correlation coefficient between number of years of education and cigarette smoking is -.89. This means that as education level increases:

A) smoking tends to increase
B) smoking tends to decrease
C) smoking changes 89 percent
D) smoking is nonexistent
E) only 89 out of every 100 people in the study would not smoke

Question 29

What is used to determine whether a nonmonotonic relationship exists between two nominal-scaled variables?

A) tabulation analysis and t tests
B) cross-tabulation and chi-square tests
C) cross-tabulation and t tests
D) tabulation analysis and chi-square tests
E) only t tests

Question 30

If we were to graph two variables, let's say, height (in inches)and GPA, and the graph showed points scattered about in a formless shape, we could say there is:

A) likely no significant relationship between height and GPA
B) likely a positive relationship between height and GPA
C) likely a negative relationship between height and GPA
D) a need to re-graph the data
E) likely a curvilinear relationship between height and GPA

Question 31

Which of the following correlation coefficients would indicate a "moderate" association?

A) .85
B) -.85
C) -.75
D) .35
E) .60

Question 32

In order to run a chi-square test using SPSS, the proper command sequence is:

A) ANALYZE; CHI-SQUARE; GO
B) ANALYZE; SUMMARIZE; CROSSTABS; CHI-SQUARE
C) ASSOCIATIONS; NONMONO; CHI-SQUARE
D) ANALYZE; DESCRIPTIVE STATISTICS; CROSSTABS
E) ASSOCIATIONS; STATISTICS; CROSSTABS

Question 33

If you were to find a significant association between two nominally scaled variables, a good way to present the findings in your cross-tabulation table would be to use:

A) a p value for each nonmonotonic relationship found
B) a Sig. value for each nonmonotonic relationship found
C) graphical presentations
D) numerical presentations that clearly indicate the direction and strength of the relationship
E) numerical presentations that clearly indicate the linear values in the relationship

Question 34

If you wished to compute a Pearson Product Moment correlation coefficient using SPSS, which command sequence would you use?

A) CORRELATE; PEARSON; GO
B) ANALYZE; CORRELATE; PEARSON
C) ANALYZE; CORRELATE; PEARSON; R
D) ANALYZE; CORRELATE; PEARSON; GO
E) ANALYZE; CORRELATE; BIVARIATE

Question 35

The Pearson Product Moment correlation measures the linear relationship between two:

A) nominal- or ordinal-scaled variables
B) nominal- or interval-scaled variables
C) interval- or ratio-scaled variables
D) nominal- or ratio-scaled variables
E) ordinal- or interval-scaled variables

Question 36

Which of the following is NOT a number that can be found in each cross-classification table cell?

A) frequency
B) raw percentage
C) column percentage
D) overall percentage
E) row percentage

Question 37

A cross-tabulation table is sometimes referred to as a:

A) tabular table
B) nonmonotonic display table
C) r x c table
D) t x n table
E) c x x table

Question 38

If we run the chi-square test and we get a .02 level of significance to support the null hypothesis, this means:

A) there is adequate support for the null hypothesis
B) there is no association between the two nominally scaled variables
C) there is only a 2 percent chance that the two nominally scaled variables are systematically related
D) there is a significant association between the two nominally scaled variables, and this information alone is adequate to explain the nature of the association
E) there is a significant association between the two nominally scaled variables, but this information alone is a "flag" that we need to look more closely at the variables to discern the nature of the relationship

Question 39

Assume that a researcher and client determine that a p value of .05 or less determines significance. Listed below are several correlation coefficients and their respective significance levels. Which correlation coefficient demonstrates an association not likely due to chance; that is, is it significant?

A) .22, .06
B) .75, .05
C) .32, .15
D) .76, .95
E) .05, 1.00

Question 40

The intersection of a row and column in a cross-tabulation table is called:

A) a cross-tabulation cell
B) a dangerous intersection
C) a chi-square
D) a cross-cell interaction
E) a row box

Question 41

When the descriptors on a scale measure levels or amounts, that is, the level of sales dollars, the scale is metric (interval or ratio).

Question 42

Which of the following are caveats of correlation?

A) It's use is limited to metric variables (interval or ratio scaled).
B) It examines the relationship between only two variables.
C) Do not assume cause and effect.
D) It is limited to linear relationships.
E) all of the above

Question 43

The manager of the city's professional hockey team conducted a large survey. He wanted to know if there was an association between fans being "season ticket holders" versus "nonseason ticket holders" and whether they "bought" versus "didn't buy" hockey team merchandise at the game. Because his survey included these measurements, he used SPSS to run a Pearson Product Moment correlation coefficient that turned out to be .88 with a Sig. value of .001. This meant:

A) there is no significant relationship
B) there is a significant, strong, positive relationship
C) there is a significant, strong, negative relationship
D) there is a significant, moderate, positive relationship
E) None of the above; Pearson's Product Moment correlation is not the appropriate measure of association here because both variables are nominally scaled.

Question 44

The characteristic of direction, in describing the relationship between two variables, is applicable only when the relationship is either monotonic or linear.

Question 45

When the descriptors on a scale measure levels or amounts, that is, the level of sales dollars, the scale is categorical (nominal or ordinal).

Question 46

If we determine a precise linear relationship between two variables, then by knowing the value of one variable we should be able to predict the other variable.

Question 47

We would use cross-tabulation if we wanted to visualize the nonmonotonic relationship between two nominal scaled variables.

Question 48

The fact that most tourists at sunny beach resorts use sun block is a curvilinear relationship.

Question 49

Cross-tabulation tables allow us to look at two variables simultaneously and are arranged in a row and column format.

Question 50

The key to establishing a nonmonotonic relationship between two variables is determining if one is present (or absent)when the other is present (or absent).

Question 51

Frequencies tables, part of cross-tabulation, contain raw data.

Question 52

A researcher runs a correlation analysis between two variables that she is certain are associated but the analysis indicates the two variables are not associated. The researcher may then want to:

A) run another association test and add three variables
B) adopt a lower standard for determining significance, that is, a p value of .20
C) do nothing; if the association is deemed insignificant it is inappropriate to run further analyses
D) run a scatter plot in search of a curvilinear relationship
E) run a scatter plot in search of a linear relationship

Question 53

Coffee orders at a restaurant are present at breakfast and soft drink orders are present at lunch. This is an example of a monotonic relationship.

Question 54

Associative analyses determine whether stable relationships exist between two variables.

Question 55

Both low-level and high-level scales can incorporate very precise linear relationships.

Question 56

Presence refers to the strength of a relationship between three or more variables.

Question 57

Significant association relationships are always causal.

Question 58

Let's assume we use SPSS to run a correlation analysis, and we get a Pearson correlation coefficient of .941 and a Sig. value of .000. These figures mean:

A) there is 94.1 percent probability for supporting the null hypothesis
B) there is little or no probability for supporting the null hypothesis
C) there is little or no probability for supporting the null hypothesis, and there is a strong association
D) there is little or no probability for supporting the null hypothesis, and there is a strong, positive association
E) that more information is needed

Question 59

The key to establishing a monotonic relationship between two variables is determining if there is a general direction (increasing or decreasing)between the two variables.

Question 60

Associative analysis procedures identify if there are consistent and systematic relationships between variables.

Question 61

With the information used to construct a cross-tabulation table, a column percentages table can also be constructed, but a row percentages table cannot.

Question 62

The degrees of freedom in chi-square are calculated by multiplying the rows, minus one times the columns, minus one.

Question 63

Chi-square analysis is a form of a "goodness of fit" test.

Question 64

To use a correlation, you must first establish that it is statistically significant from one.

Question 65

When we present findings on cross-tabulations with chi-square analysis, graphical presentations such as bar or stacked bar charts are often useful.

Question 66

When it comes to determining the statistical significance of the correlation coefficient, there are rules of thumb. For example, .81 to 1.00 is considered to be "strong."

Question 67

Row cell percentage is calculated by dividing a cell frequency by the cell row total.

Question 68

It is more difficult for a chi-square value with high degrees of freedom (i.e., there are more cells)to achieve "significance" than for a chi-square value with fewer degrees of freedom (i.e., there are fewer cells).

Question 69

Regardless of the size of the correlation value, if it is not significant, it is meaningless.

Question 70

Covariation is defined as the amount of change in one variable systematically associated with a change in another variable.

Question 71

In the chi-square analysis, the greater the differences between the observed frequencies and the expected frequencies, the less likely it is that there will be a statistically significant relationship.

Question 72

To run chi-square analysis in SPSS, you should go to the Crosstabs command, click Statistics, then select Chi-Square.

Question 73

When we present findings on cross-tabulations with chi-square analysis, we cannot use the characteristics of direction or strength because we are dealing with categorical, or nominal scaled, variables.

Question 74

In chi-square, the null hypothesis states that there is no association. When the "Asymp. Sig." on the SPSS output is low, this means there is low support for the null hypothesis.

Question 75

There is no connection between scatter diagrams and correlation coefficients.

Question 76

The manager of the city's professional hockey team conducted a large survey. He wanted to know if there was an association between fans being "season ticket holders" versus "non-season ticket holders" and whether they "bought" versus "didn't buy" hockey team merchandise at the game. Because his survey included these measurements, he used SPSS to run a cross-tabulation analysis and selected the chi-square test to test for the significance of any association. The Pearson Chi-Square turned out to be 81.6 with an Asymp. Sig. value of .001. This meant there is a significant relationship.

Question 77

Chi-square analysis always begins with the assumption that no association exists between the two nominal scaled variables under analysis.

Question 78

The chi-square analysis would be appropriate to determine if there is an association between the number of dollars spent on books and the number of years of education.

Question 79

A correlation coefficient is an index number constrained to fall between the range of -1.0 and +1.0 that communicates both the strength and the direction of association between three or more variables.

Question 80

Generally speaking, the chi-square value is calculated by dividing the sum of the squared differences between observed and expected frequencies by the expected frequencies.