Deck 12: Advanced Data Analysis

Question 1

Marketing research is conducted to help reduce the uncertainty in decision making.

Answer 1

True

Question 2

Statistics is not a useful tool to help managers better understand the significance of marketplace characteristics.

Answer 2

False

Question 3

Statistically significant differences are significant enough to influence management decisions.

Answer 3

False

Question 4

Managerial significant differences occur when observed differences are considered significant enough to influence management decisions.

Question 5

We have hypotheses because we suspect our results will reveal some characteristic of the population under study, not because we know that they will do so.

Question 6

The null hypothesis is stated in terms that indicate you expect to find no difference or no effect.

Question 7

The alternative hypothesis states that you expect to find a difference or effect.

Question 8

We assume the null hypothesis is true until evidence suggests otherwise.

Question 9

The null and alternative hypotheses are competing hypotheses, therefore, only one can be true.

Question 10

A Type 1 error means accepting the null hypothesis when it is actually false.

Question 11

A Type 2 error means rejecting the null hypothesis when it is actually true.

Question 12

When testing a hypothesis in any particular study, we can make either a Type 1 error or a Type 2 error, but not both.

Question 13

The significance level is the maximum risk you will accept of making a Type 2 error.

Question 14

The greater the costs associated with a Type 1 error, the lower we should set our maximum risk.

Question 15

If costs associated with a Type 1 error are high and a Type 2 error is not costly, set the level of significance much higher, perhaps at .25 or higher.

Question 16

If costs associated with a Type 1 error are not high and a Type 2 error is costly, set the level of significance low, perhaps at .05 or lower.

Question 17

If both Type 1 and Type 2 errors are costly, set the level of significance very low and increase the sample size.

Question 18

An appropriate statistical test for hypotheses about frequency distributions is a chi-square.

Question 19

An appropriate statistical test for hypotheses about proportions is a chi-square.

Question 20

An appropriate statistical test for hypotheses about means with one small sample is a t-test.

Question 21

An appropriate statistical test for hypotheses about means with three or more subgroups or samples is a one-way ANOVA.

Question 22

The actual chance that we are making a Type 1 error is called the p-value.

Question 23

SPSS is a statistical software program that can be used to conduct statistical tests.

Question 24

Descriptive or causal research is used to generate hypotheses.

Question 25

Nonmonotonic pattern means that there is a general direction to the association between the two variables.

Question 26

Monotonic pattern means that there is no discernable direction to the association between the two variables.

Question 27

A curvilinear pattern exists when the association between the variables can be described by a curved line.

Question 28

A linear pattern exists when a straight line can be drawn to describe the relationship between the variables.

Question 29

Linear relationships are always positive in nature.

Question 30

If the association between two variables is strong, you can predict one variable by knowing the other one.

Question 31

The Pearson product moment correlation is used when both of the variables to be measured are metric.

Question 32

The correlation coefficient using the Pearson product moment correlation ranges from -1 to 0.

Question 33

The "r" measures the strength of an association on a -1 to +1 scale.

Question 34

The association between two variables is very strong when the correlation coefficient is greater than .80.

Question 35

For variables that are ordinal, association is measured using the Spearman rank order correlation.

Question 36

Chi-square tests are used when conducting hypothesis test for interval scaled variables.

Question 37

If the Chi-square test shows a .01 probability for the null hypothesis, the conclusion would be that you would get these results only 10 times in a hundred occasions if the null hypothesis were true.

Question 38

Multivariate analysis simultaneously analyzes multiple measurements of attitudes, attributes, or behavior.

Question 39

Multiple regression is an example of bivariate analysis.

Question 40

Multiple regression allows for the prediction of the level or magnitude of a phenomenon such as market size or market share.

Question 41

Canonical correlation has as its objective the identification of key descriptors on which various predefined events will vary.

Question 42

Canonical correlation allows for the establishment of a predictive model that can simultaneously forecast or explain several phenomena based on an understanding of their correlates.

Question 43

Factor analysis is one of the most popular multivariate techniques.

Question 44

Multidimensional scaling addresses the problem of identifying the dimensions upon which customers perceive or evaluate phenomena.

Question 45

Multidimensional scaling techniques result in perceptional maps that describe the positioning of companies or brands that are compared relative to the position they occupy in the minds of customers according to key attributes.

Question 46

Factor analysis is a method for determining the number and nature of underlying constructs or variables among a larger set of measures.

Question 47

Cluster analysis allows for the classification or segmentation of a larger group of variables into homogeneous subgroups, based on similarities on a profile of information.

Question 48

Hypotheses are speculations regarding specific findings of the research, therefore they are not useful.

Question 49

The t-test is a type of statistical test.

Question 50

Higher rs indicate higher degrees of association.

Question 51

When any observed difference in findings is considered significant enough to influence management decisions, the difference is

A) statistically significant
B) managerially significant
C) both statistically and managerially significant

Question 52

When any observed difference in findings is large enough that it is unlikely to have occurred due to mere chance, it is

A) statistically significant
B) managerially significant
C) both statistically and managerially significant

Question 53

Hypothesis testing involves all of the following steps except

A) compare the results to the alternative hypothesis but not to the null hypothesis
B) stating the hypotheses
C) determining the costs related to the two types of errors arising from the decisions
D) selecting the significance level

Question 54

Hypotheses are stated in which two forms?

A) null and void hypotheses
B) null and alternative hypotheses
C) alternative and original hypotheses
D) null and altered hypotheses

Question 55

The _____ hypothesis is stated in terms that indicate you expect to find no difference or no effect.

A) alternative
B) null

Question 56

The _____ hypothesis states that you expect to find a difference or effect.

A) alternative
B) null

Question 57

Which statement below is true?

A) only the null hypothesis can be true
B) only the alternative hypothesis can be true
C) both the null and alternative hypotheses can be true at the same time
D) only the null or the alternative hypothesis can be true

Question 58

When testing a hypothesis in any particular study, the researcher can make

A) only a type 1 error
B) only a type 2 error
C) both a type 1 and 2 error at the same time
D) only a type 1 or a type 2 error, but not both

Question 59

In setting the level of significance if costs associated with type 1 error are

A) high and a type 2 error is not costly set the level of significance very low
B) high and a type 2 error is not costly set the level of significance very high
C) not very high and a type 2 error is costly set the level of significance very low
D) not very high and a type 2 error is not costly set the level of significance much higher

Question 60

The greater the costs associated with Type 1 error, the _____ we should set our maximum risk.

A) higher
B) lower

Question 61

If both type 1 and type 2 errors are costly,

A) set the level of significance very high and increase the sample size
B) set the level of significance very high and decrease the sample size
C) set the level of significance very low and increase the sample size
D) set the level of significance very low and decrease the sample size

Question 62

The appropriate statistical test for hypotheses about frequency distributions with two or more subgroups and nominal data is

A) chi-square
B) Z-test
C) t-test
D) ANOVA

Question 63

The appropriate statistical test for hypotheses about frequency distributions with one sample and nominal data is

A) chi-square
B) Z-test
C) t-test
D) ANOVA

Question 64

The appropriate statistical test for hypotheses about proportions with one large sample is

A) chi-square
B) Z-test
C) t-test
D) ANOVA

Question 65

The appropriate statistical test for hypotheses about proportions with two large samples is

A) chi-square
B) Z-test
C) t-test
D) ANOVA

Question 66

The appropriate statistical test for hypotheses about means with two large samples is

A) chi-square
B) Z-test
C) t-test
D) ANOVA

Question 67

The appropriate statistical test for hypotheses about means with one small sample is

A) chi-square
B) Z-test
C) t-test
D) ANOVA

Question 68

The appropriate statistical test for hypotheses about means with three or more subgroups is

A) chi-square
B) Z-test
C) t-test
D) ANOVA

Question 69

The actual chance that we are making a type 1 error is called the

A) margin of error
B) null hypothesis
C) p value
D) t value

Question 70

If the p-value of a statistical test was .03, it would indicate the chance of making a _____ is 3 chances out of 100.

A) type 1 error
B) type 2 error
C) type 1 or type 2 error
D) type 1 and type 2 error

Question 71

The association between two variables can be graphically described in all of the following ways except

A) nonmonotonic
B) monotonic
C) curvilinear
D) geometric

Question 72

Curvilinear association means

A) there is no discernable direction to the association between the two variables
B) there is a general direction to the association between the two variables
C) the association between the variables can be described by a curved line
D) a straight line can be drawn to describe the association between the two variables

Question 73

Linear association means

A) there is no discernable direction to the association between the two variables
B) there is a general direction to the association between the two variables
C) the association between the variables can be described by a curved line
D) a straight line can be drawn to describe the association between the two variables

Question 74

Monotonic association means

A) there is no discernable direction to the association between the two variables
B) there is a general direction to the association between the two variables
C) the association between the variables can be described by a curved line
D) a straight line can be drawn to describe the association between the two variables

Question 75

Nonmonotonic association means

A) there is no discernable direction to the association between the two variables, but a relationship exists
B) there is a general direction to the association between the two variables
C) the association between the variables can be described by a curved line
D) a straight line can be drawn to describe the association between the two variables

Question 76

Bivariate measures of association include all of the following except

A) Pearson product moment correlation
B) Spearman's rank-order correlation
C) Chi-square
D) Multiple regression

Question 77

_____ is used to measure the degree to which two variables are correlated or associated with each other when both of the variables are metric.

A) Pearson product moment correlation
B) Spearman's rank-order correlation
C) Chi-square
D) Multiple regression

Question 78

In using the Pearson product moment correlation, a score of -0.90 would indicate the two variables are

A) highly positive correlated
B) highly inversely correlated
C) slightly positive correlated
D) slightly inversely correlated

Question 79

In using the Pearson product moment correlation, a score of 0.13 would indicate the two variables are

A) highly positive correlated
B) highly inversely correlated
C) slightly positive correlated
D) slightly inversely correlated

Question 80

The Pearson product moment correlation is used to measure the correlation of

A) nominal data
B) ordinal data
C) interval or ratio data
D) ordinal or interval