chapter 4
78. If X is found to be highly and positively correlated to Y, all of the following are true except a. X causes Y b. X is related to Y c. X is associated with Y d. X can predict Y
A
80. The Y-intercept of a regression line is the value of Y when X is equal to a. zero b. the mean of Y c. the mean of X d. the slope
A
82. Simple linear regression refers to ______ X variable(s) and ______ Y variable(s). a. one; one b. one; two c. two; two d. many; one
A
83. In PASW, to run a straight-line regression analysis, go to Analyze, Regression, then a. Linear b. Logistic c. Bivariate d. Correlation
A
18. The percentage of variance in one variable accounted for or shared by another variable is denoted by what statistic? a. correlation coefficient b. r2 c. Spearman's rho d. multiple correlation
B
19. If a correlation coefficient of 1.15 is found, this would indicate a. an extremely high relationship b. a computational error c. an extremely high inverse relationship d. a negative relationship
B
2. Which of the following correlation coefficients reflects the greatest linear relationship? a. .12 b. .98 c. .58 d. .93 e. 1.12
B
20. If two variables have a correlation coefficient of .85, it can be inferred that a. causality exists b. a relationship exists c. both causality and a relationship exist d. nothing can be inferred
B
26. The relationship between the coordinate points (1,2), (3,3), (4,5), and (5,5) is a. indirect b. direct c. nonlinear d. no relationship
B
12. Which of these correlation coefficients has the least predictive value? a. .91 b. .50 c. .17 d. .23
C
16. The shape of a scatterplot indicating the relationship between shoe size and performance in a graduate statistics class would resemble a. a banana b. a box c. an orange d. a pear
C
17. r2xy indicates the a. true score variance of X b. total variance of Y that is reliable c. percentage of variance common to X and Y d. covariance of Y associated with X
C
37. If predicting Y from X and r = .95 which of the following is most likely to be true? a. The scores are more variable in X than in Y. b. The relationship indicates that one cannot predict very well. c. If you scored high on one variable, you probably scored high on the other variable. d. If you scored high on one variable, you probably scored low on the other variable.
C
49. What does bivariate mean? a. correlation b. X and Y c. two variables d. multiple regression
C
57. The correlation between test A and test B is .34. The correlation between test A and test C is .68. Which of the following statements is correct? a. The relationship between tests A and C is twice as strong as that between tests A and B. b. If computed, the relationship between tests B and C would be approximately .34. c. The relationship between tests A and C is higher than that between tests A and B. d. A student receiving a high score on test A will receive a higher score on test C than on test B.
C
65. The most appropriate graph for displaying an X,Y relationship is a a. bar graph b. dot plot c. histogram d. scatterplot
D
100. Four different variables were separately examined to investigate their predictive relationship to Y. Which of the following models shows a better fit to the data? a. model 1: sum of squared residuals = 200 b. model 2: sum of squared residuals = 350 c. model 3: sum of squared residuals = 500 d. model 4: sum of squared residuals = 900
A
101. Four different variables were separately examined to investigate their predictive relationship to Y. Which of the following models shows a better fit to the data? a. model 1: 1 − R2 = .09 b. model 2: 1 − R2 = .15 c. model 3: 1 − R2 = .60 d. model 4: 1 − R2 = .88
A
106. If a correlation coefficient is negative, the relationship is indirect. a. true b. false
A
108. A small SEE would indicate a relatively good fit. a. true b. false
A
109. Subtracting the coefficient of determination from 1 is helpful if you want to better understand how much lack of fit is in your model. a. true b. false
A
112. The sum of the residuals is always equal to zero. a. true b. false
A
113. A scatterplot is useful for identifying unusual points in the data. a. true b. false
A
14. Which of the following statements is true? a. A correlation of +.80 is just as strong as a correlation of -.80. b. Two things being highly correlated means that one causes the other. c. In skewed distributions the mean is usually greater than the median. d. The standard deviation is not the square root of the variance.
A
15. Which of the following is true if a correlation of -1.00 exists between two variables? a. A high score on one variable is associated with a low score on the other variable. b. All persons with the same score on one variable will also have equal scores on the other variable. c. The score on one variable is caused by the score on the other variable. d. The coefficient of determination between the two variables would be 0.00.
A
21. If a correlation of -.70 were found between body weight and the number of push-ups one can do, which of the following statements would be correct? a. Generally, heavy people can do few push-ups. b. Excess body weight makes doing push-ups difficult. c. Generally, underweight people can do few push-ups. d. Low body weight is associated with low push-up scores.
A
22. What is the general relationship between skinfold measures and hydrostatically determined percent body fat? a. high positive b. low positive c. zero d. low negative
A
23. The correlation between percent body fat by skinfolds and underwater weighing is .90. A scatterplot of this relationship would look like a. a pickle b. a box c. an orange d. a pear
A
24. A positive correlation between two variables indicates that a. students scoring low on one variable score low on the other variable b. students scoring high on one variable score low on the other variable c. students scoring low on one variable score high on the other variable d. students scoring high on one variable have no relation to their score on the other variable
A
27. A correlation of .8 is considered a. a high correlation b. a low correlation c. to indicate a cause-and-effect relationship d. to indicate that two variables have a shared variance of 80%
A
3. Which of the following best describes the correlation between the 12-minute run and VO2max? a. high and positive b. low and positive c. high and negative d. low and negative e. zero or near zero
A
31. Multiple regression indicates a. multiple predictor variables b. multiple dependent variables c. nonlinear relationships d. that high correlations have been found
A
32. Which of the following describes the general relationship between the SEE and the coefficient of determination? a. They are inversely (negatively) related. b. They are directly (positively) related. c. There is no relationship between them. d. The relationship can be high or it can be low, depending on the situation. e. It depends on the correlation and the variables measured.
A
35. Mary is predicted to score 150 on an achievement test in which the mean is 130 and the standard deviation is 20. Approximately how likely is it that her actual score will be above 158 if the coefficient of determination between the predictor and the achievement test is .84? a. 2.5% b. 16% c. 34% d. unable to determine
A
4. Maureen scored 140 on an achievement test in which the mean was 80 and the standard deviation was 20. What might you predict she would score on another test if the correlation between achievement and the other test were .91? a. well above the mean b. slightly above the mean c. about the mean d. slightly below the mean e. well below the mean
A
41. Which symbol is commonly used to indicate a Pearson correlation coefficient? a. r b. a c. s d. p
A
60. A test serves a prediction function well if it a. has a strong statistical relationship with a criterion b. contains items that appeal to common sense c. is stable over time and situations d. contains items that tend to correlate highly with one another
A
67. The sign (plus or minus) of a correlation coefficient indicates a. the direction of the relationship b. the importance of the relationship c. the magnitude of the relationship d. the statistical significance of the relationship
A
68. The square of the correlation coefficient is called the a. coefficient of determination b. coefficient of reliability c. coefficient of validity d. coefficient of variation
A
71. If 30 students take a pretest on the first day of class and 20 of those students take the posttest on the last day of class, what is the correlation sample size (n)? a. 20 b. 30 c. 50 d. 60 e. A correlation cannot be computed in this case.
A
72. If there are 10 values of X and 10 paired values of Y, what is the sample size (n) of the correlation coefficient? a. 10 b. 18 c. 20 d. 38 e. 40
A
74. In PASW, to calculate a correlation coefficient, go to Analyze, Correlate, then a. Bivariate b. Multivariate c. Univariate d. Moderate
A
92. A regression analysis yields the equation Y = 3.0 + 20(X) − 30(Z). What is or are the dependent variable(s)? a. Y b. X c. Z d. X and Z
A
95. A regression analysis yields the equation Y = 3.0 − 20(X) − 30(Z). Considering an X value of 5 and a Z value of 10, what is the Y-intercept? a. 3 b. 20 c. 30 d. 100 e. 300
A
97. A regression analysis yields the equation Y = 3.0 + 20(X) − 30(Z). What would the predicted value of Y be if X and Z were both zero? a. 3 b. 53 c. 403 d. zero
A
98. Four different variables were separately examined to investigate their predictive relationship to Y. Which of the following models shows a better fit to the data? a. model 1: SEE = .009 b. model 2: SEE = .012 c. model 3: SEE = .762 d. model 4: SEE = .988
A
30. In simple linear prediction, the residual scores are a. strongly correlated with the predictor variable (X) b. found by subtracting the predicted Y from the measured Y c. moderately correlated with the predictor variable (X) d. inversely related to the measured Y
B
34. You want to predict success on your final examination based on your midterm score. The prediction equation is Y' = 5(X) + 20. What is your predicted score on the final if you scored 10 on the midterm? The mean on the final is 70, and the standard deviation of the final is 10. a. 60 b. 70 c. 80 d. unable to determine
B
39. Which of the following statistics is a measure of the amount of variance accounted for between two variables? a. standard deviation b. r2 c. r d. median
B
40. What does a correlation coefficient describe about the relationship between two variables? a. the mean and strength b. the direction and strength c. the strength and distribution d. the mean and variability
B
50. What does a scatterplot tell you? a. the correlation between variables b. the general nature of the relationship between variables c. the limitations of the correlation coefficient d. the shared variance between two variables
B
59. The residual variance best represents a. true score variance b. experimental variance c. nonpredicted variance d. common variance
C
61. If no relationship exists between two variables, a plot of matched points would look like a a. slanted straight line b. football c. circle d. slightly curved line
C
70. If values of X tend to be random as their paired values of Y tend to decrease, the relationship between X and Y will be a. direct b. indirect c. nearly zero d. positive
C
73. A correlation coefficient describes the ___________ and the ___________ of a linear relationship. a. association; scale b. association; direction c. magnitude; direction d. magnitude; scale e. none of these
C
75. The coefficient of determination can be interpreted as a(n) a. average b. median c. percent d. standard deviation
C
76. The PPM correlation coefficient cannot a. be negative b. be squared c. be interpreted for curvilinear relationships d. be interpreted as a measure of strength
C
8. Skinfolds are often used in multiple regression equations to estimate body fatness. What does multiple regression equation refer to? a. Many people can use the equation. b. Many subjects were used to develop the equation. c. More than one predictor variable is used. d. There are multiple ways to determine reliability and validity.
C
86. The standard error of estimate is also called the a. standard error of X b. standard error of mean c. standard error of prediction d. standard error of variance e. standard error of measurement
C
9. The square of the correlation coefficient indicates a. the amount of residual variance b. the total variance of the two variables c. the percentage of variance in common with the two variables d. the covariance of the two variables
C
103. A regression analysis yields the equation Y = 125.0 − 50(X). Considering an X value of −10, what is the Y-intercept? a. −10 b. 50 c. −50 d. 125
D
104. A regression analysis yields the equation Y = 125.0 − 50(X). Considering an X value of −10, what is the predicted value of Y? a. −500 b. 500 c. −375 d. 625
D
11. What statistic is computed when correlating more than two variables with the criterion? a. slope b. Pearson product moment c. t test d. multiple R
D
25. An indirect relationship would indicate a. a zero correlation b. a positive correlation c. a relationship not suitable for prediction d. a negative correlation
D
36. In general, which of the following types of relationship is best to obtain? a. a direct (positive) relationship b. an inverse (negative) relationship c. a relationship with a low coefficient of determination d. a relationship with a low standard error of estimate e. a relationship with low variability
D
38. Which of the following is the primary purpose of correlational research? a. to study individual situations in depth b. to study disease rates in the population c. to combine and quantify past research reports d. to study the relationships among variables
D
48. What can we conclude if the correlation between length of marriage and marital happiness is .82? a. There is no relationship between length of marriage and happiness. b. Long marriages cause happy marriages. c. Happy marriages cause long marriages. d. Long marriages tend to be happy ones.
D
51. What does the coefficient of determination tell you? a. the correlation between variables b. the general nature of the relationship between variables c. the limitations of the correlation coefficient d. the shared variance between two variables
D
63. The PPM correlation coefficient is an index of the a. direct relationship between two variables b. indirect relationship between two variables c. inverse relationship between two variables d. linear relationship between two variables
D
64. An indirect relationship is the same as a a. causal relationship b. direct relationship c. positive relationship d. negative relationship
D
87. The standard error of estimate for the regression line of X and Y can be calculated with the standard deviation of Y and the a. standard deviation of X b. standard error of X c. coefficient of determination of Y and Z d. correlation coefficient of X and Y
D
93. A regression analysis yields the equation Y = 3.0 + 20(X) − 30(Z). What is or are the independent variable(s)? a. Y b. X c. Z d. X and Z
D
96. A regression analysis yields the equation Y = 3.0 + 20(X) − 30(Z). Considering an X value of 5 and a Z value of 10, what is the predicted value of Y? a. 53 b. 400 c. 403 d. −197
D
99. Four different variables were separately examined to investigate their predictive relationship to Y. Which of the following models shows a better fit to the data? a. model 1: R2 = .08 b. model 2: R2 = .25 c. model 3: R2 = .75 d. model 4: R2 = .98
D
10. If there is a negative relationship between two physical measures, a. the relationship is quite small b. the relationship does not always exist c. the two physical measures are unrelated d. the value of one measure decreases as the value of the other measure decreases e. the value of one measure increases as the value of the other measure decreases
E
29. In a prediction or regression equation, the standard error of estimate is a function of a. the standard deviation of the predictor variable (X) b. the standard deviation of the dependent variable (Y) c. the correlation between X and Y d. the sample size e. b and c
E
33. Which of the following correlation coefficients provides the most predictive power? a. 1.09 b. .92 c. .01 d. .64 e. -.94
E
62. Which of the following correlations displays the strongest relationship? a. .25 b. .95 c. −.25 d. −.93 e. −.99
E
66. If values of X are all positive and values of Y are all negative, the relationship between X and Y will be a. direct b. indirect c. linear d. positive e. unapparent until a correlation is computed
E
79. With correlation, a. X is the dependent variable b. Y is the dependent variable c. the more important variable is the dependent variable d. the outcome variable is the dependent variable e. it does not matter which variable is the dependent variable
E
85. The correlation between regression residual and Y should be a. indirect b. direct c. positive d. linear e. zero
E
90. If a regression analysis uses age, sum of skinfolds (SS), SS2, and gender to better understand body density (BD), the dependent variable(s) is or are a. age b. SS c. SS2 d. gender e. BD
E
56. A negative correlation coefficient indicates low relationship. a. true b. false
B
46. Which of the following relationships would be most precisely described with a correlation statistic? a. height and weight b. height and competitiveness c. weight and anxiety d. competitiveness and anxiety
A
5. Scotty is predicted to have a score of 75 on a test. The correlation between the predictor variable (X) and the predicted variable (Y) is .80. Both variables have standard deviations of 10. How likely is it that Scotty's actual score on the Y variable is above 81? a. 16% b. 34% c. 68% d. 84%
A
52. What can you do if two variables are correlated? a. Estimate one from the other. b. Describe the causative relationship between the variables. c. Determine which variable is more important. d. Disregard both variables.
A
55. What does the standard error of estimate tell you? a. how much error there is in prediction b. how high the correlation is c. the coefficient of determination d. the multiple correlation
A
58. The Dallas Cowboys computed correlations between players' ratings (determined by the coaches) and four specific performance tests. The correlations with the ratings were as follows: test A = -.87; test B = .75; test C = .62; test D = .57. In selecting potential players, the Cowboys would do best to use which test? a. test A b. test B c. test C d. test D
A
6. The correlation between X and Y is .90. What can be said with certainty about Y in relation to X? a. They are related. b. High scores on X are associated with low scores on Y. c. A high score on X causes one's score to be high on Y. d. More than one of these can be stated with certainty. e. None of these can be stated with certainty.
A
102. A regression analysis yields the equation Y = 125.0 − 50(X). What is the value of the correlation coefficient? a. impossible to determine b. negative c. positive d. nearly zero e. very high in absolute value
B
105. A regression analysis yields the equation Y = 125.0 − 50(X). What is the relationship between X and Y? a. direct b. indirect c. positive d. nonlinear
B
107. A correlation coefficient of .96 depicts a strong indirect relationship. a. true b. false
B
110. An adjusted R square is useful when interpreting a simple linear regression analysis. a. true b. false
B
111. Multiple correlation is when you analyze more than 1 correlation coefficient. a. true b. false
B
13. Between what limits can the Pearson product-moment correlation coefficient vary? a. 0.0 and +1.0 b. -1.0 and +1.0 c. -2.0 and +2.0 d. 0 to 100
B
42. What does a negative value of a Pearson correlation coefficient mean? a. that there is no relationship between two variables b. that high scores on one variable are associated with low scores on a second variable c. that high scores on one variable are associated with high scores on a second variable d. that there was a mistake in the calculation of the coefficient
B
53. What does the Pearson product-moment correlation coefficient tell you? a. the relationship between two variables b. the linear relationship between two variables c. the relationship among several variables d. the amount of variation explained
B
69. If values of X tend to increase as their paired values of Y tend to decrease, the relationship between X and Y will be a. direct b. indirect c. nonlinear d. positive e. unapparent until a correlation is computed
B
88. The following can be used for determining regression fit except a. SEE b. SEM c. R2 d. residuals
B
89. If a regression analysis uses age, sum of skinfolds (SS), SS2, and gender to better understand body density, the analysis is called a. simple regression b. multiple regression c. simple correlation d. logistic regression
B
91. If a regression analysis uses age, sum of skinfolds (SS), SS2, and gender to better understand body density, how many predictors are there? a. 1 b. 2 c. 3 d. 4 e. 5
C
94. A regression analysis yields the equation Y = 3.0 + 20(X) − 30(Z). What is the correlation between the predictor and predicted variable? a. very high b. very low c. positive d. negative e. it is impossible to determine from what is given
C
43. Which conclusion should be drawn if the correlation between a physical fitness test and a rating of basketball playing ability is 0.90? a. People who are physically fit will be successful in playing basketball. b. Playing basketball develops physical fitness. c. Physical fitness is necessary for success in playing basketball. d. Basketball playing ability and physical fitness are strongly associated.
D
44. What does a correlation coefficient of 1.0 mean? a. that no relationship exists between two sets of scores b. that those who did the best on the first test did the worst on the second test c. that those who did the best on the first test were average on the second test d. that those who did the best on the first test also did the best on the second test
D
45. Which correlation coefficient best describes the relationship between height and weight in healthy individuals across ages 12 to 80? a. -.87 b. .00 c. .25 d. .80
D
47. How well does a test of athletic ability with a .12 correlation with batting average in baseball players predict batting success? a. extremely well b. very well c. moderately well d. not very well
D
54. What is the difference between the actual and the estimated variable called? a. overprediction b. underprediction c. PPM d. error
D
7. What is the major distinction between simple correlation and multiple correlation? a. one uses r and the other uses R b. the size of the SEE (it is larger with multiple correlation) c. the size of the SEE (it is smaller with multiple correlation) d. the number of predictors used for the correlation e. the number of people used for the correlation
D
77. Estimating one variable from another variable is called a. correlation b. estimation c. generalization d. prediction
D
81. The amount of change in Y for a unit change in X is evaluated with the ___________ of a regression line. a. X value b. Y value c. Y-intercept d. slope
D
84. The amount of inaccuracy of regression formula is not represented by a. E b. Y − Y′ c. residuals d. Y-intercept
D
1. What is the major difference between simple regression and multiple regression? a. R is used b. r is used c. R2 is used d. number of subjects needed e. number of predictors used
E
28. In detecting a strong curvilinear relationship, a. a correlation near -1.0 would be expected b. a correlation near +1.0 would be expected c. a correlation near -.5 would be expected d. a correlation near +.5 would be expected e. a correlation near 0 would be expected
E
