Chapter 23-Bivariate Statistical Analysis: Measures of Association
Chapter 23-Bivariate Statistical Analysis: Measures of Association
Chapter 23-Bivariate Statistical Analysis: Measures of Association
TRUE/FALSE
ANS: F
Measure of association is a general term that refers to a number of bivariate statistical techniques used
to measure the strength of a relationship between two variables, and causality is not always
hypothesized or determined.
3. Covariance is the extent to which a change in one variable corresponds systematically to a change in
another.
4. If r = -0.88, this indicates a weak relationship between the two variables under study.
ANS: F
This indicates a strong negative relationship.
5. The measure of association used in the textbook is the Pearson product-moment correlation coefficient
and its symbol is r.
6. A Spearman correlation is more appropriate for interval and ratio data than is the Pearson product-
moment correlation.
ANS: F
A Spearman correlation is more appropriate for ordinal level data.
7. The correlation coefficient, r, indicates the magnitude of the linear relationships but not the direction
of that relationship.
ANS: F
A correlation coefficient indicates both the magnitude of the linear relationship and the direction of
that relationship.
© 2010 Cengage Learning. All Rights Reserved. May not be scanned, copied or duplicated, or posted to a publicly accessible website, in
whole or in part.
PTS: 1 REF: p. 559 NAT: AACSB: Reflective Thinking
8. Correlation always means that a causal relationship exists between the two variables under study.
ANS: F
Systematic covariation does not in and of itself establish causality. The relationship would also need
to be nonspurious and that any hypothesized “cause” would have to occur before any subsequent
effect.
9. The square of the correlation coefficient indicates the part of the total variance of Y that can be
accounted for by X.
ANS: F
The main diagonal consists of correlations of 1.00 because it is the correlation of a variable with itself.
11. The statistical significance of a correlation can be tested using the t-test.
13. In a regression equation, the slope of the line, b, is the change in X that is due to a corresponding
change of one unit of Y.
ANS: F
In a regression equation, the slope of the line, b, is the change in Y that is due to a corresponding
change of one unit of X.
14. In most business research, the estimate of a in a regression equation is most important.
ANS: F
The estimate of b is most important because the explanatory power of regression rests with this
parameter because this is where the direction and strength of the relationship between the independent
and dependent variable is explained.
15. In simple regression, raw parameter estimates reflect the measurement scale range.
© 2010 Cengage Learning. All Rights Reserved. May not be scanned, copied or duplicated, or posted to a publicly accessible website, in
whole or in part.
ANS: T PTS: 1 REF: p. 566
NAT: AACSB: Reflective Thinking
ANS: F
It is always 0.
17. If the purpose of the regression analysis is forecasting, then standardized regression estimates must be
used.
ANS: F
Raw parameter estimates must be used for forecasting.
18. One way to determine the relationship between X and Y is to simply visually draw the best-fit straight
line through the points in the figure.
19. The ordinary least-squares method of regression analysis is based on the logic of how much better a
regression line can predict values of Y compared to simply using the mean as a prediction.
20. The least-squares regression line minimizes the sum of the squared deviations of the actual values
from the predicted values in the regression line.
ANS: F
An F-test provides the statistical significance of a regression model.
22. The coefficient of determination reflects the proportion of variance that can be explained by the
regression line.
23. A rule of thumb for R2 is that it must be greater than 0.80 for a regression model to be significant.
ANS: F
© 2010 Cengage Learning. All Rights Reserved. May not be scanned, copied or duplicated, or posted to a publicly accessible website, in
whole or in part.
Guidelines for R2 values are neither simple nor straightforward. Additionally, R2 does not provide a
test of significance, but rather reflects the proportion of variance explained by the regression line.
24. The first step in interpreting simple regression output is to interpret the overall significance of the
model.
MULTIPLE CHOICE
1. When the on-time performance of airlines is used to predict the number of customer complaints in a
regression equation, on-time performance is the _____ variable and the number of customer
complaints is the _____ variable.
a. independent; dependent
b. dependent; dependent
c. dependent; independent
d. independent; predictor
ANS: A PTS: 1 REF: p. 559
NAT: AACSB: Reflective Thinking
2. A bivariate statistical technique that is used to measure the strength of the relationship between two
variables is called the:
a. one-group t-test
b. two-group t-test
c. measure of association
d. correlation matrix
ANS: C PTS: 1 REF: p. 559
NAT: AACSB: Reflective Thinking
3. The extent to which a change in one variable corresponds systematically to a change in another is
called:
a. spurious association
b. significance
c. covariance
d. standardized coefficient
ANS: C PTS: 1 REF: p. 559
NAT: AACSB: Reflective Thinking
4. The Pearson product-moment correlation requires what type of scale for the data?
a. nominal
b. interval
c. ordinal
d. all of the above
© 2010 Cengage Learning. All Rights Reserved. May not be scanned, copied or duplicated, or posted to a publicly accessible website, in
whole or in part.
ANS: B PTS: 1 REF: p. 559
NAT: AACSB: Reflective Thinking
6. If there is no relationship between two variables, the correlation coefficient between them would be:
a. +1.0
b. -1.0
c. +0.50
d. 0
ANS: D PTS: 1 REF: p. 559
NAT: AACSB: Reflective Thinking
8. When the correlation between two variables is +0.92, this means that as one variable _____ , the other
variable _____.
a. decreases; increases
b. increases; decreases
c. increases; increases
d. decreases; stays the same
ANS: C PTS: 1 REF: p. 559
NAT: AACSB: Reflective Thinking
10. Which of the following represents the proportion of the total variance of a variable accounted for by
another value of another variable?
a. product-moment correlation
b. correlation coefficient
c. F-test
d. coefficient of determination
© 2010 Cengage Learning. All Rights Reserved. May not be scanned, copied or duplicated, or posted to a publicly accessible website, in
whole or in part.
ANS: D PTS: 1 REF: p. 562
NAT: AACSB: Reflective Thinking
12. If the correlation between X and Y is -0.72, approximately what percentage of the variance in Y can be
explained by X?
a. 52 percent
b. 72 percent
c. 85 percent
d. 28 percent
ANS: A PTS: 1 REF: p. 562
NAT: AACSB: Reflective Thinking| AACSB: Analytic
13. Which of the following is the standard form for reporting observed correlations among multiple
variables?
a. correlation matrix
b. contingency table
c. Pearson grid
d. inverse table
ANS: A PTS: 1 REF: p. 562
NAT: AACSB: Reflective Thinking
14. In a correlation matrix, the correlations in the main diagonal are all equal to:
a. -1.00
b. 0
c. +0.50
d. +1.00
ANS: D PTS: 1 REF: p. 564
NAT: AACSB: Reflective Thinking
15. If the t-test is used to test the significance of a correlation coefficient, the null hypothesis is:
a. r = +1.00
b. r = 0
c. r = -1.00
d. r = 100
ANS: B PTS: 1 REF: p. 564
NAT: AACSB: Reflective Thinking
16. In regression analysis, the symbol X is commonly used for the _____ variable, and the symbol Y is
commonly used for the _____ variable.
a. dependent; moderating
b. independent; dependent
c. dependent; independent
© 2010 Cengage Learning. All Rights Reserved. May not be scanned, copied or duplicated, or posted to a publicly accessible website, in
whole or in part.
d. independent; moderating
ANS: B PTS: 1 REF: p. 564
NAT: AACSB: Reflective Thinking
17. In regression analysis, the dependent variable is also called the _____ variable.
a. predictor
b. exogenous
c. criterion
d. internal
ANS: C PTS: 1 REF: p. 564
NAT: AACSB: Reflective Thinking
21. If the regression equation is: Y = 24.35 - 14.2 X , then 24.35 is the _____ , while -14.2 is the _____.
a. slope; y-intercept
b. independent variable; slope
c. dependent variable; y-intercept
d. Y-intercept; slope
ANS: D PTS: 1 REF: p. 567
NAT: AACSB: Reflective Thinking
22. If the regression equation is: Y = -4.2 + 3.6 X , then the expected score for Y when X is 4 would be:
a. -18.6
b. 10.2
c. 18.6
d. 4
© 2010 Cengage Learning. All Rights Reserved. May not be scanned, copied or duplicated, or posted to a publicly accessible website, in
whole or in part.
ANS: B PTS: 1 REF: p. 567
NAT: AACSB: Reflective Thinking|AACSB: Analytic
23. Which of the following provides a common metric allowing regression results to be compared to one
another no matter what the original scale range may have been?
a. standardized regression coefficient (b)
b. c2
c. coefficient of determination (R2)
d. raw parameter estimates
ANS: A PTS: 1 REF: p. 567
NAT: AACSB: Reflective Thinking
24. Which of the following is most appropriate if the purpose of the regression analysis is forecasting?
a. standardized regression coefficient (b)
b. Y-intercept a
c. coefficient of determination (R2)
d. raw parameter estimates
ANS: D PTS: 1 REF: p. 567
NAT: AACSB: Reflective Thinking
25. Lance is studying the relationship between sales training of the sales force and customer satisfaction
and loyalty. When researchers like Lance are focused on explanation rather than prediction, then
which of the following is most appropriate when using simple regression?
a. standardized regression coefficient (b)
b. Y-intercept a
c. c2
d. raw parameter estimates
ANS: A PTS: 1 REF: p. 567
NAT: AACSB: Reflective Thinking
26. All of the following are simple regression estimation techniques EXCEPT:
a. chi-square estimation
b. maximum likelihood
c. visual estimation
d. ordinary least squares (OLS)
ANS: A PTS: 1 REF: p. 567
NAT: AACSB: Reflective Thinking
27. Which regression estimation technique is based on the logic of how much better a regression line can
predict values of Y compared to simply using the mean as a prediction for all observations no matter
what the value of X may be?
a. visual estimation
b. maximum likelihood
c. coefficient of determination
d. ordinary least squares (OLS)
ANS: D PTS: 1 REF: p. 569
NAT: AACSB: Reflective Thinking
28. The statistical significance of a regression model is determined using which test?
a. t-test
© 2010 Cengage Learning. All Rights Reserved. May not be scanned, copied or duplicated, or posted to a publicly accessible website, in
whole or in part.
b. c2
c. F-test
d. Z-test
ANS: C PTS: 1 REF: p. 570
NAT: AACSB: Reflective Thinking
COMPLETION
1. If the number of coupons given out at shopping malls is used to predict the number of tickets that will
be sold to a country and western band's performance at a local club, the number of coupons is the
____________________ variable.
ANS: independent
ANS: interval
3. The extent to which a change in one variable corresponds systematically to a change in another is
referred to as ____________________.
ANS: covariance
ANS:
-1.0, +1.0
+1.0, -1.0
5. The statistical measure of the association between two variables is known as the
____________________ coefficient.
ANS: correlation
6. A correlation coefficient indicates both the ____________________ of the relationship between two
variables and the ____________________ of this relationship.
© 2010 Cengage Learning. All Rights Reserved. May not be scanned, copied or duplicated, or posted to a publicly accessible website, in
whole or in part.
7. Covariation in which the association between variables is in the opposite direction indicates a(n)
____________________.
ANS:
negative relationship
inverse relationship
9. If the correlation between two variables is -.55, the coefficient of determination is approximately
____________________.
ANS: + 0.30
10. The standard format for reporting the correlations between several variables is called the
____________________.
11. In a correlation matrix, the values in the main diagonal equal ____________________.
ANS: 1.00
12. In testing for the significance of a correlation, the null hypothesis is that the correlation is equal to
____________________.
ANS:
zero
0
13. The regression parameter that represents the height of the regression line relative to horizontal is
____________________.
ANS:
alpha
a
© 2010 Cengage Learning. All Rights Reserved. May not be scanned, copied or duplicated, or posted to a publicly accessible website, in
whole or in part.
14. In a regression equation, the slope of the regression line is denoted by the symbol
____________________.
ANS:
beta
b
15. The test for the statistical significance of the regression model is the ____________________ test.
ANS: F
ESSAY
1. Discuss the common procedures for testing association for data that are nominal, ordinal, and
interval/ratio.
ANS:
For nominal data, the statistical choice is cross-tabulation with c2 test. For ordinal data, the statistical
choice is the Spearman rank correlation or cross-tabulation with c2 test. For interval/ratio data, the
statistical choice is Pearson’s r or simple regression.
ANS:
Concomitant variation is one condition needed to establish a causal relationship between two
variables. When two variables covary, they display concomitant variation. This systematic
covariation does not in and of itself establish causality. The relationship would also need to be
nonspurious and that any hypothesized “cause” would have to occur before any subsequent effect.
ANS:
Regression analysis is a technique for measuring the linear association between a dependent and an
independent variable. Although simple regression and correlation are mathematically equivalent in
most respects, regression is a dependence technique where correlation is an interdependence
technique. A dependence technique draws a distinction between dependent and independent variables.
An interdependence technique does not make this distinction and simply is concerned with how
variables relate to one another.
© 2010 Cengage Learning. All Rights Reserved. May not be scanned, copied or duplicated, or posted to a publicly accessible website, in
whole or in part.
4. Discuss the pros and cons of raw regression estimates and standardized regression estimates and
discuss when each is appropriate.
ANS:
Regression parameter estimates can be presented in either raw or standardized form. One potential
problem with raw parameter estimates is due to the fact that they reflect the measurement scale
range. However, if the purpose of the regression analysis is forecasting, then raw parameter estimates
must be used. This is another way of saying that the researcher’s primary interest is prediction. A
standardized regression coefficient (b) provides a common metric allowing regression results to be
compared to one another no matter what the original scale range may have been. Standardized
regression estimates should be used when the researcher is testing explanatory hypotheses (i.e., when
the purpose of the research is more explanation than prediction).
ANS:
Interpreting simple regression output is a simple two-step process:
(1) Interpret the overall significance of the model. The output will include the “model F” and a
significance value. When the model F is significant (low p-value), the independent variable explains a
significant portion of the variation in the dependent variable. The coefficient of determination, or R2,
can be interpreted. This is the percentage of total variation in the dependent variable accounted for by
the independent variable.
(2) The individual parameter coefficient is interpreted. The t-value associated with the slope
coefficient can be interpreted.
© 2010 Cengage Learning. All Rights Reserved. May not be scanned, copied or duplicated, or posted to a publicly accessible website, in
whole or in part.