Research Methods in Psychology

Evaluating a World of Information 1st

Edition Morling Test Bank
Visit to download the full and correct content document: https://testbankdeal.com/dow
nload/research-methods-in-psychology-evaluating-a-world-of-information-1st-edition-
morling-test-bank/
CHAPTER 7: BIVARIATE CORRELATIONAL RESEARCH

MULTIPLE CHOICE

1. A study finds a correlation coefficient of r = .22. According to Cohen’s benchmarks, the magnitude of
this effect is:
a. Negative
b. Large
c. Multiply determined
d. Categorical
e. None of the above
ANS: E DIF: Easy
REF: Introduction to Bivariate Correlations: Review: Describing Associations Between Two
Quantitative Variables KEY: Learning Objective 2
NOT: Factual

2. Which of the following graph formats is the best way to examine an association claim between a
categorical variable and a quantitative variable?
a. A scatterplot
b. A line graph
c. A bar graph
d. A pie chart
e. All of the above
ANS: C DIF: Easy
REF: Introduction to Bivariate Correlations: Describing Associations with Categorical Data
KEY: Learning Objective 3 NOT: Factual

3. When examining an association claim using a bar graph, an association is indicated by which of the
following?
a. A difference in the height between the bars
b. The number of bars in the graph
c. The number of observations that make each bar
d. The direction of the bars
e. None of the above
ANS: A DIF: Medium
REF: Introduction to Bivariate Correlations: Describing Associations with Categorical Data
KEY: Learning Objective 3 NOT: Factual

4. When examining an association in which one variable is categorical and one is quantitative, which of
the following statistical tests is likely to be used?
a. A t test
b. A phi coefficient
c. Point-biserial correlation
d. Either a or c
e. All of the above
ANS: D DIF: Difficult
REF: Introduction to Bivariate Correlations: Describing Associations with Categorical Data
KEY: Learning Objective 3 NOT: Factual
5. Which of the following tells you that an association claim is being made?
a. The graph used to interpret the results
b. The fact that the two variables are measured
c. The statistic used to interpret the results
d. The validities that can be established
e. None of the above
ANS: B DIF: Medium
REF: Introduction to Bivariate Correlations: Two Measured Variables Make a Study Correlational
KEY: Learning Objective 2 | Learning Objective 3 NOT: Conceptual

RESEARCH STUDY 7.1

Dr. Oswald conducts a study examining the relationship between the number of friends one has and
the experience of daily stress and life satisfaction. She randomly samples 1,500 elderly men and
women in the Memphis, Tennessee, area in the southern United States. Below are her findings.
 Life satisfaction and experience of daily stress: r = -.57 (p = .01)
 Number of friends one has and experience of daily stress: r = .09, not sig.
 Number of friends one has and life satisfaction: r = .36 (p = .04)

6. Refer to Research Study 7.1 above to answer the following question.

In evaluating Dr. Oswald’s study, you question the construct validity of the study. Which of the
following questions would you be asking?
a. How did Dr. Oswald recruit her participants?
b. Which statistic did Dr. Oswald compute?
c. How reliable is the measure of daily stress?
d. Does the number of friends cause people to experience less stress?
e. How long did the study take to conduct?
ANS: C DIF: Easy
REF: Interrogating Association Claims: Construct Validity: How Well Was Each Variable
Measured? KEY: Learning Objective 4 NOT: Applied

7. Refer to Research Study 7.1 above to answer the following question.

According to the benchmarks established by Cohen, what type of effect size has Dr. Oswald found for
the association between number of friends people have and life satisfaction?
a. Very small
b. Small
c. Medium
d. Large
e. Very large
ANS: C DIF: Easy
REF: Interrogating Association Claims: Statistical Validity Question 1: What Is the Effect Size?
KEY: Learning Objective 5 NOT: Applied

8. Refer to Research Study 7.1 above to answer the following question.

Comparing all three correlations, Dr. Oswald will be most able to accurately predict life satisfaction
from the experience of daily stress because:
 a. The relationship is negative
b. The relationship has the largest effect size
c. The relationship was reported first
d. The relationship was statistically significant
e. None of the above—this relationship will not lead to the most accurate predictions.
ANS: B DIF: Easy
REF: Interrogating Association Claims: Statistical Validity Question 1: What Is the Effect Size?
KEY: Learning Objective 5 NOT: Conceptual

9. Refer to Research Study 7.1 above to answer the following question.

Which of the following conclusions can Dr. Oswald draw about the number of friends one has and life
satisfaction, based on her statistical analyses?
a. The probability of her sample coming from a zero association population is about 4%.
b. The probability of her sample coming from a zero association population is about 96%.
c. The relationship is not statistically significant.
d. The strong correlation means that the number of friends one has causes increases in life
satisfaction.
e. None of the above
ANS: A DIF: Difficult
REF: Interrogating Association Claims: Statistical Validity Question 2: Is the Correlation Statistically
Significant? KEY: Learning Objective 5
NOT: Applied

10. Refer to Research Study 7.1 above to answer the following question.
The “not sig.” in Dr. Oswald’s findings indicates all of the following EXCEPT:
a. It is likely that the association between number of friends one has and experience of daily
stress is from a zero association population
b. Effect size could not be calculated
c. There is not a statistically significant association between the two variables
d. She cannot reliably predict the experience of daily stress from the number of friends one
has
e. All of the above are true
ANS: B DIF: Difficult
REF: Interrogating Association Claims: Statistical Validity Question 2: Is the Correlation Statistically
Significant? KEY: Learning Objective 5
NOT: Applied

11. Refer to Research Study 7.1 above to answer the following question.
In determining whether the relationship between two of her variables was statistically significant,
which of the following must be considered?
a. Sample size and number of variables analyzed
b. Direction of the association and strength of the association
c. Sample size and effect size
d. The number of outliers and the direction of the association
e. None of the above
ANS: C DIF: Medium
REF: Interrogating Association Claims: Statistical Validity Question 2: Is the Correlation Statistically
Significant? KEY: Learning Objective 5
 NOT: Applied

12. Refer to Research Study 7.1 above to answer the following question.
Dr. Oswald realizes that the women in her study have more friends than the men in her study. This
might result in which of the following?
a. Outliers due to subgroups
b. Larger effect sizes
c. More measured variables
d. Spurious associations due to subgroups
e. None of the above
ANS: D DIF: Medium
REF: Interrogating Association Claims: Statistical Validity Question 3: Are There Subgroups?
KEY: Learning Objective 5 NOT: Applied

13. Refer to Research Study 7.1 above to answer the following question.
Dr. Oswald creates a scatterplot of the relationship between the experience of daily stress and life
satisfaction. In doing so, she realizes there are three scores that seem to be very extreme and are
nowhere near the other points on the scatterplot. Specifically, it appears that three people report very
high levels of daily stress and very low levels of life satisfaction. Dr. Oswald should probably consider
these scores _________________.
a. Random
b. Moderators
c. Outliers
d. Curvilinear scores
e. None of the above
ANS: C DIF: Medium
REF: Interrogating Association Claims: Statistical Validity Question 4: Could Outliers Be Affecting
the Relationship? KEY: Learning Objective 5 NOT: Applied

14. Refer to Research Study 7.1 above to answer the following question.
Dr. Oswald creates a scatterplot of the relationship between the experience of daily stress and life
satisfaction. In doing so, she realizes there are three scores that seem to be very extreme and are
nowhere near the other points on the scatterplot. Specifically, it appears that three people report very
high levels of daily stress and very low levels of life satisfaction. Which of the following statements is
true?
a. These scores may have strengthened the correlation between these two variables.
b. These scores are more likely to have an effect because of the large sample size.
c. These scores are more likely to have an effect because they are extreme on both variables.
d. Both a and c are true.
e. All of the above are true.
ANS: D DIF: Difficult
REF: Interrogating Association Claims: Statistical Validity Question 4: Could Outliers Be Affecting
the Relationship? KEY: Learning Objective 5 NOT: Conceptual

15. Refer to Research Study 7.1 above to answer the following question.
Dr. Oswald has decided to examine one of her relationships with a scatterplot to double-check for a
curvilinear relationship. Which relationship will be most important for her to examine?
a. Life satisfaction and experience of daily stress
 b. Number of friends one has and experience of daily stress
c. Number of friends one has and life satisfaction
d. Both a and b
e. Both a and c
ANS: B DIF: Medium
REF: Interrogating Association Claims: Statistical Validity Question 5: Is the Relationship
Curvilinear? KEY: Learning Objective 5 NOT: Applied

16. Refer to Research Study 7.1 above to answer the following question.
Matt, Dr. Oswald’s research assistant, is discussing the findings of the study with some other students.
He claims that the experience of more daily stress causes people to have lower life satisfaction. Which
of the following keeps Matt from being correct?
a. The study did not establish the covariance of cause and effect
b. The study did not establish temporal precedence
c. The study did not establish internal validity
d. Both b and c
e. All of the above
ANS: D DIF: Medium
REF: Interrogating Association Claims: Internal Validity: Can We Make a Causal Inference from an
Association? KEY: Learning Objective 7 NOT: Applied

17. Refer to Research Study 7.1 above to answer the following question.
Dr. Oswald submits her study for publication in a scientific journal. If one of the peer reviewers is
concerned about the external validity of her study, which of the following is the most important aspect
of Dr. Oswald’s study to consider?
a. The random sampling technique used to recruit the participants
b. The 1,500 people in the sample
c. The use of three measured variables
d. The number of significant findings
e. All of the above
ANS: A DIF: Medium
REF: Interrogating Association Claims: External Validity: To Whom Can the Association Be
Generalized? KEY: Learning Objective 6 NOT: Applied

18. Refer to Research Study 7.1 above to answer the following question.
Considering Dr. Oswald’s study, her results could most safely be generalized to which of the groups
below?
a. People in the southern United States
b. Men
c. Elderly people
d. People with a high number of friends
e. People with high life satisfaction
ANS: C DIF: Medium
REF: Interrogating Association Claims: External Validity: To Whom Can the Association Be
Generalized? KEY: Learning Objective 6 NOT: Applied

19. Refer to Research Study 7.1 above to answer the following question.
 Considering Dr. Oswald’s sample, which of the following statements is true?
a. The association found in her study could probably generalize to teenagers.
b. The association found in her study could probably generalize to people in other large cities
in Tennessee.
c. The association found in her study could probably generalize to people in New York City.
d. The association found in her study could probably generalize to elderly persons living in
nursing homes.
e. All of the above are true.
ANS: B DIF: Medium
REF: Interrogating Association Claims: Just How Important Is External Validity?
KEY: Learning Objective 6 NOT: Applied

20. Refer to Research Study 7.1 above to answer the following question.
Dr. Oswald finds that the relationship between the number of friends one has and life satisfaction is
stronger for men than for women. In this study, sex (male or female) is considered a/an:
a. Outlier
b. Cause
c. Moderator
d. Effect
e. Spurious variable
ANS: C DIF: Medium
REF: Interrogating Association Claims: Moderating Variables KEY: Learning Objective 6
NOT: Applied

21. Refer to Research Study 7.1 above to answer the following question.
Dr. Oswald finds that the relationship between the number of friends one has and life satisfaction is
stronger for men than for women. Why might Dr. Oswald have looked for this difference?
a. To examine her study’s external validity
b. To examine her study’s internal validity
c. To determine whether the association was statistically significant
d. To determine whether the association was curvilinear
e. To determine whether the association was spurious
ANS: A DIF: Medium
REF: Interrogating Association Claims: Moderating Variables KEY: Learning Objective 6
NOT: Conceptual

22. Which of the following is true of statistical significance testing?

a. It is a process of inference.
b. It can lead to an incorrect conclusion about the population.
c. It is synonymous with measuring effect size.
d. Both a and b are true.
e. All of the above are true.
ANS: D DIF: Medium
REF: Interrogating Association Claims: Statistical Validity Question 2: Is the Correlation Statistically
Significant? KEY: Learning Objective 5
NOT: Conceptual

23. Stronger effect sizes allow for which of the following?

 a. Greater likelihood of finding a statistically significant relationship
b. Greater accuracy in predicting one variable as opposed to another
c. Greater likelihood of a finding being important in the real world
d. Both a and c
e. All of the above
ANS: E DIF: Difficult
REF: Interrogating Association Claims: Statistical Validity Question 1: What Is the Effect Size?
KEY: Learning Objective 5 NOT: Applied

24. In which of the following cases might a small effect still be important?
a. When the sample is very large
b. When the study has life-or-death implications
c. When the finding is also statistically significant
d. When external validity is high
e. None of the above
ANS: B DIF: Medium
REF: Interrogating Association Claims: Statistical Validity Question 1: What Is the Effect Size?
KEY: Learning Objective 5 NOT: Applied

25. Which of the following is true of the relationship between effect size and statistical significance?
a. Effect size alone is sufficient to indicate statistical significance.
b. Statistical significance alone is sufficient to indicate effect size.
c. An association’s effect size has no effect on statistical significance.
d. Effect size and statistical significance are synonymous terms.
e. None of the above are true.
ANS: E DIF: Difficult
REF: Interrogating Association Claims: Statistical Validity Question 2: Is the Correlation Statistically
Significant? KEY: Learning Objective 5
NOT: Conceptual

26. Statistical significance depends on which of the following?

a. Sample size and number of variables analyzed
b. Direction of the association and strength of the association
c. Sample size and effect size
d. Number of outliers and direction of the association
e. None of the above
ANS: C DIF: Medium
REF: Interrogating Association Claims: Statistical Validity Question 2: Is the Correlation Statistically
Significant? KEY: Learning Objective 5
NOT: Factual

27. Martin has found a correlation of r ? .18 between the two variables of caffeine consumption and
frontal lobe activity. This correlation is more likely to be statistically significant if:
a. The study can be applied to the real world
b. Martin used a larger number of subjects
c. Martin measured frontal lobe activity extremely accurately
d. Martin’s measure of caffeine consumption is categorical
e. Martin’s study is based on a random sample
ANS: B DIF: Medium
 REF: Interrogating Association Claims: Statistical Validity Question 3: Are There Subgroups?
KEY: Learning Objective 5 NOT: Applied

28. All of the following are true of outliers EXCEPT:

a. They have the biggest effect when dealing with large sample sizes
b. They can affect the direction of an association
c. They can affect the strength of an association
d. They can weaken a relationship
e. They are especially problematic when there are outliers on both variables
ANS: A DIF: Medium
REF: Interrogating Association Claims: Statistical Validity Question 4: Could Outliers Be Affecting
the Relationship? KEY: Learning Objective 5 NOT: Factual

29. Why are curvilinear relationships hard to detect with correlation coefficients (r)?
a. Because curvilinear relationships require a large amount of scores
b. Because curvilinear relationships have small effect sizes
c. Because r looks for the best straight line to fit the data
d. Because r assumes a zero association
e. Because r assumes a negative relationship
ANS: C DIF: Difficult
REF: Interrogating Association Claims: Statistical Validity Question 5: Is the Relationship
Curvilinear? KEY: Learning Objective 5 NOT: Conceptual

30. Which of the following is NOT a question you should ask about the statistical validity of an
association claim?
a. What is the effect size?
b. Is the correlation statistically significant?
c. Are there subgroups?
d. Is random assignment affecting the findings?
e. Could outliers be affecting the relationship?
ANS: D DIF: Easy
REF: Interrogating Association Claims: Statistical Validity: How Well Do the Data Support the
Conclusion? KEY: Learning Objective 6 NOT: Factual

31. For a third variable to be plausible as the explanation in an established association, which of the
following must also be true?
a. The third variable must be related to both of the measured variables in the original
association.
b. The third variable must be measured on the same scale as the original measured variables.
c. The third variable must be a categorical variable.
d. The third variable must have a positive relationship with the two measured variables in the
original association.
e. The third variable must be spurious.
ANS: A DIF: Difficult
REF: Interrogating Association Claims: Internal Validity: Can We Make a Causal Inference from an
Association? KEY: Learning Objective 7 NOT: Factual

32. When evaluating the external validity of an association claim, which of the following is the most
important issue to consider?
 a. The way the sample was selected from the population
b. The size of the sample
c. The number of subgroups
d. The presence of outliers
e. The size of the original population
ANS: A DIF: Medium
REF: Interrogating Association Claims: External Validity: To Whom Can the Association Be
Generalized? KEY: Learning Objective 6 NOT: Factual

33. If an association study did not select people for the study by using random sampling, which of the
following statements is true?
a. The association should be rejected as inconclusive.
b. The variables should be measured more than once.
c. The effect size should be considered, but tests of statistical significance should not.
d. Moderators should be looked for.
e. None of the above statements are true.
ANS: E DIF: Difficult
REF: Interrogating Association Claims: Just How Important Is External Validity?
KEY: Learning Objective 6 NOT: Factual

34. Which of the following is true of moderators?

a. They help establish a cause and effect relationship.
b. They decrease effect size.
c. They can inform external validity.
d. They weaken statistical significance.
e. None of the above are true.
ANS: C DIF: Medium
REF: Interrogating Association Claims: Moderating Variables KEY: Learning Objective 6
NOT: Factual

35. What is the relationship between moderators and external validity?

a. Moderators suggest that associations may be spurious.
b. Moderators suggest that associations may not generalize to all subgroups of people.
c. Moderators are necessary for external validity to be established.
d. Moderators suggest that an association between two variables will extend to another
variable.
e. Moderators tell us nothing about external validity.
ANS: B DIF: Difficult
REF: Interrogating Association Claims: Moderating Variables KEY: Learning Objective 6
NOT: Conceptual

SHORT ANSWER

RESEARCH STUDY 7.2

Dr. Moynihan conducts a study examining the relationship between the number of panic attacks a
person experienced in the past month, the number of nightmares experienced in the past month, and
people’s levels of neuroticism (a personality trait characterized by instability, anxiety, and worry). She
collects a convenience sample of 50 community college students (25 males, 25 females). Below are
her findings.
  A – number of panic attacks in the past month and neuroticism: r = -.03, not sig.
 B – number of panic attacks in the past month and number of nightmares in the past month: r = -.14
(p = .05)
 C – number of nightmares in the past month and neuroticism: r = .48 (p = .003)

1. Refer to Research Study 7.2 above to answer the following question.

Draw a scatterplot for the three associations found by Dr. Moynihan.

ANS:
Students should draw three scatterplots. Scatterplot A should show a flat vertical line with dots
showing no discernable pattern OR it could be drawn to show a curvilinear relationship. Scatterplot B
should show a weak negative relationship (a line pointing down, with dots loosely grouped around the
line). Scatterplot C should show a strong positive relationship (a line pointing up, with dots closely
grouped around the line).

DIF: Easy
REF: Introduction to Bivariate Correlations: Review: Describing Associations Between Two
Quantitative Variables KEY: Learning Objective 1
NOT: Applied

2. Refer to Research Study 7.2 above to answer the following question.

Explain the three associations from Dr. Moynihan’s study in terms of direction of each association,
statistical significance, and effect size (using Cohen’s benchmarks).

ANS:
Students should explain all three findings in three ways:
a. Direction of the association – A is negative, B is negative, and C is positive
b. Statistical significance – B and C are statistically significant, A is not statistically significant
c. Effect size – A has a zero association, B has a small effect size, C has a large effect size

DIF: Easy
REF: Introduction to Bivariate Correlations: Review: Describing Associations Between Two
Quantitative Variables KEY: Learning Objective 2
NOT: Applied

3. Refer to Research Study 7.2 above to answer the following question.

State two questions that you might ask if you were to interrogate the construct validity of Dr.
Moynihan’s study.

ANS:
Students should pose at least two questions that relate to the reliability of the measures or the validity
of the measures. For example, one question could be: “Is the measure of neuroticism really measuring
neuroticism?” Another question could be: “Are people able to reliably report on the number of
nightmares reported in the past month?”

DIF: Medium
REF: Interrogating Association Claims: Construct Validity: How Well Was Each Variable
Measured? KEY: Learning Objective 4 NOT: Applied
4. Refer to Research Study 7.2 above to answer the following question.
Of the three findings in Dr. Moynihan’s study, which finding has the largest effect size? Name two
things that having a large effect size means.

ANS:
Students should state that the finding that has the largest effect size is C (number of nightmares in the
past month and neuroticism). Students should state that large effect sizes do two of the following: (a)
allow us to more accurately predict the number of nightmares in the past month from people’s
neuroticism scores; (b) indicate how important a finding is; (c) increase the likelihood of finding a
statistically significant association between the two variables.

DIF: Medium
REF: Interrogating Association Claims: Statistical Validity Question 1: What Is the Effect Size?
KEY: Learning Objective 4 NOT: Applied

5. Refer to Research Study 7.2 above to answer the following question.

Of the three findings in Dr. Moynihan’s study, which associations are statistically significant? What
does it mean that an association is statistically significant?

ANS:
Students should state that there are two statistically significant findings: B (number of panic attacks in
the past month and number of nightmares in the past month) and C (number of nightmares in the past
month and neuroticism). Statistical significance means that the association is likely to have been found
by chance. Students may be specific that there is a less than 5% chance that Dr. Moynihan’s sample
came from a zero association population.

DIF: Difficult
REF: Interrogating Association Claims: Statistical Validity Question 2: Is the Correlation Statistically
Significant? KEY: Learning Objective 5
NOT: Conceptual

6. Refer to Research Study 7.2 above to answer the following question.

What is an outlier? What aspect of Dr. Moynihan’s study makes his findings especially susceptible to
outliers? Why is this a problem?

ANS:
Students should state that outliers are extreme scores. Dr. Moynihan should be especially worried
about outliers because he is working with a fairly small sample size. Small sample sizes are a problem
because the outliers can have more of an effect on the association.

DIF: Difficult
REF: Interrogating Association Claims: Statistical Validity Question 2: Is the Correlation Statistically
Significant? KEY: Learning Objective 5
NOT: Conceptual

7. Refer to Research Study 7.2 above to answer the following question.

Imagine that Dr. Moynihan finds two outliers. These two people report a very high number of
nightmares in the past month and have very high scores on neuroticism. Why should Dr. Moynihan be
worried about the presence of these two outliers?
 ANS:
Students should state that Dr. Moynihan should be worried about the presence of outliers because they
may be affecting the association between these two variables. In this case, it is possible that these two
scores are causing there to be a significant relationship when there is none.

DIF: Difficult
REF: Interrogating Association Claims: Statistical Validity Question 2: Is the Correlation Statistically
Significant? KEY: Learning Objective 5
NOT: Conceptual

8. Refer to Research Study 7.2 above to answer the following question.

Imagine that Dr. Moynihan is concerned about curvilinear relationships. Which of the three
relationships is most likely to be affected by a curvilinear relationship? How should Dr. Moynihan
check for this type of relationship? Why would this be effective?

ANS:
Students should state that relationship A (number of panic attacks in the past month and neuroticism)
is most likely to be affected by a curvilinear relationship (students may also state that this is because
curvilinear relationships yield nonsignificant r coefficients, but they do not have to). Students should
state that Dr. Moynihan should check for this type of relationship by creating a scatterplot of that
association. This would be effective because it would allow him to see the curvilinear relationship,
whereas an r coefficient only looks for straight lines.

DIF: Medium
REF: Interrogating Association Claims: Statistical Validity Question 5: Is the Relationship
Curvilinear? KEY: Learning Objective 5 NOT: Conceptual and Applied

9. Refer to Research Study 7.2 above to answer the following question.

Examining Dr. Moynihan’s study and using the three criteria of causation, why can he not say that the
number of panic attacks in the past month causes one to have fewer nightmares in the same month?

ANS:
Students should state that Dr. Moynihan is unable to establish two of the three criteria. He can
establish criterion 1—the two variables do covary. He cannot establish temporal precedence, as both
variables were measured in the past month (so they are co-occurring). He also cannot establish internal
validity because (a) neither variable was manipulated and (b) a third variable could be causing both
variables.

DIF: Medium
REF: Interrogating Association Claims: Internal Validity: Can We Make a Causal Inference from an
Association? KEY: Learning Objective 7 NOT: Conceptual

10. Refer to Research Study 7.2 above to answer the following question.
How strong is the external validity of Dr. Moynihan’s study? How does the fact that he finds no
differences in the associations in men and women also speak to the study’s external validity?

ANS:
 Students should question the external validity of his study because of the convenience sampling
method used to collect his sample. Students may also question the sample size; however, if this is the
only issue they state, it is not sufficient, as sample size is less important that sampling method. The
lack of a moderating effect of sex suggests that his findings may apply equally to men and women.

DIF: Medium
REF: Interrogating Association Claims: External Validity: To Whom Can the Association Be
Generalized? KEY: Learning Objective 6 NOT: Applied

11. Draw a scatterplot for the following three associations:

a. A curvilinear association
b. A positive association with a large effect size
c. A zero association

ANS:
Students should draw three scatterplots. Scatterplot A should show a curvilinear relationship (either a
U relationship or an inverted U relationship). Scatterplot B should show a steep line pointed upward,
with dots closely grouped around the line. Scatterplot C should show a flat vertical line with dots
showing no discernable pattern OR it could be drawn to show a curvilinear relationship.

DIF: Easy
REF: Introduction to Bivariate Correlations: Review: Describing Associations Between Two
Quantitative Variables KEY: Learning Objective 1
NOT: Applied

12. What does it mean that an association is “spurious”? What can cause spurious associations?

ANS:
Students should state that an association is spurious when it is attributable only to the mean differences
in subgroups within the sample. Students may also state that spurious correlations make it look like
there is an association between two variables when there is none. Spurious associations can be caused
by subgroups.

DIF: Medium
REF: Interrogating Association Claims: Statistical Validity Question 3: Are There Subgroups?
KEY: Learning Objective 5 NOT: Conceptual

13. Name two ways in which the analysis of an association between two quantitative variables differs from
the analysis of an association between a categorical variable and a quantitative variable.

ANS:
Students should state that a scatterplot should be used when the association is between two quantitative
variables and that a bar graph should be used when one of the variables is categorical and one variable
is quantitative. Students should also state the statistical analyses are different. Students may
specifically state that quantitative variable associations should be analyzed with r and
categorical/quantitative associations should be analyzed with t tests or point biserial correlations,
although they do not need to state this.

DIF: Easy
REF: Introduction to Bivariate Correlations: Review: Analyzing Associations When One Variable Is
Categorical KEY: Learning Objective 3 NOT: Factual
14. Why is the size of a sample not as important to external validity as the way a sample was collected?

ANS:
Students should state that although large sample sizes are good, findings of research studies can be
generalized to others only if the sample closely matches the population of interest. A large sample that
is poorly sampled is not as good as a small sample that is well sampled in terms of external validity.

DIF: Medium
REF: Interrogating Association Claims: External Validity: To Whom Can the Association Be
Generalized? KEY: Learning Objective 6 NOT: Conceptual

15. What are the five questions that consumers of research should ask to evaluate the statistical validity of
an association claim?

ANS:
Students should state the following five questions:
a. What is the effect size?
b. Is the correlation statistically significant?
c. Are there subgroups?
d. Could outliers be affecting the relationship?
e. Is the relationship curvilinear?

DIF: Medium
REF: Interrogating Association Claims: Statistical Validity: How Well Do the Data Support the
Conclusion? KEY: Learning Objective 5 NOT: Factual