Practice Exam III
15. Computing a one-way between-subjects ANOVA is appropriate when
different participants are observed one time in each of two or more groups for one factor
92. A researcher computes a perfect negative correlation, in which each data point falls exactly on the regression line. In this example, the value of the standard error of estimate will be
equal to 0
117. An OR = 1.5 means:
exposure is associated with a higher risk of disease
2. The term "between-subjects" refers to
observing different participants one time in each group
138. A variable that is significant in a univariate logistic regression model, but not in a multivariate regression model, is likely only predictive in the univariate setting because:
of its association with the other predictor variables in the model
99. For a multiple regression analysis with 2 and 12 degrees of freedom, MS regression is 135 and MS residual is 15. What is the decision for this test?
reject the null hypothesis; the predictive variability of two predictor factors are significant
45. The correlation coefficient measures the extent to which changes in one factor are _______ in a second factor.
related to changes
111. Which measure of effect relies on the incidence rate?
relative risk
79. Using an analysis of regression, the variability in Y that is associated with error is measured by the
residual variation
38. The correlation coefficient is used to measure the ________ and ________ of the linear relationship between two factors.
strength; direction
1. For an analysis of variance, the term "one-way" refers to
the number of factors in the design
95. Multiple regression is a statistical method that includes ____ predictor variable(s) in the equation of the regression line.
two or more
83. The degrees of freedom associated with regression variation are equal to
the number of predictor variables
134. For a categorical variable with 6 levels, the number of dummy variables that must be created to include in a logistic regression model is
5
52. The normality assumption states that the population of X and Y scores form a bivariate ("two variable") normal distribution, such that
All of the above A) the population of X and Y scores are normally distributed B) for each X score, the distribution of Y scores is normally distributed C) for each Y score, the distribution of X scores is normally distributed D) all of the above
75. Which of the following is not needed to compute the y-intercept using the method of least squares?
Mxy
68. Linear regression describes the extent to which _______ predicts ________.
All of the above A) X; Y B) the predictor variable; the outcome variable C) the known variable; the to-be-predicted variable D) all of the above
26. Which of the following is not a post hoc test for a one-way between-subjects ANOVA?
F test for equal variance
6. What is the minimum number of groups that can be observed using the one-way between-subjects ANOVA design?
2
39. The correlation coefficient ranges from -1.0 to +1.0, with values closer to ±1.0 indicating
a stronger relationship between two factors
125. A confounder in the relationship between X and Y is:
a variable that is associated with both X and Y, and distorts the true relationship between X and Y.
56. A researcher observes a correlation of values from 2 to 10 points and draws conclusions about the full range of values in the population from 0 to 21 points. Which limitation for correctly interpreting a correlation coefficient did the researcher violate?
restriction of range
35. The null hypothesis for the Kruskal-Wallis test is
the sum of the ranks in each group do not differ
114. When does the OR best approximate RR?
when the incidence of disease is 1%.
4. The source of variability associated with error variance in the one-way between-subjects ANOVA is called
within-groups variability
113. If the proportion of smokers with lung cancer is .0443 and the proportion of non-smokers with lung cancer is .0012, then the RR is:
36.9
89. An estimate of the standard deviation or distance that data points fall from the regression line is measured by the
standard error of estimate
63. To summarize correlations, we report:
the p value
32. The Kruskal-Wallis test is the nonparametric analog to the
one-way ANOVA F test.
129. The dependent variable in a logistic regression model is:
Ln(Odds)
8. ANOVA stands for
analysis of variance
72. A researcher reports the following equation for a best-fitting straight line to a set of data points: Y = 0.48 X +12.03 . Which value is the slope?
0.48
47. A researcher measures the relationship between two variables, X and Y. If SSXY = 340 and SSXSSY = 320,000, then what is the value of the correlation coefficient?
0.60
93. A researcher computes an analysis of regression in which MSE = 0.82. What is the value of se in this example?
0.91
104. If F = 2.04 for the relative contribution of one factor, then what is this value when converted to a t statistic?
1.43
77. A researcher reports the following regression equation for two variables, X and Y: Y =−5.10 X 1.50 . If X = 2.30, then what is the value of Y-hat?
10.23
94. A researcher computes the following analysis of regression table. Based on the data given, what is the value of the standard error of estimate? (Note: Complete the table first.) Source of Variation SS Df MS F Regression 28 1 5.60 Residual 118 19 Total
2.24
133. If a logistic regression model generates an OR = 1.002 for the predictor X = "month", then the OR for a year increase in X increases the odds of Y by ______, adjusting for the other independent variables in the model.
2.4%
30. The following is a summary of a one-way between-subjects ANOVA: F(2, 37) = 3.42, p < .05. How many pairwise comparisons need to be made for this ANOVA result?
3
71. A researcher reports the following equation for a best-fitting straight line to a set of data points: Y =−+ 1.01 X 3.24 . Which value is the y-intercept?
3.24
13. A researcher assigned participants (n = 8 per group) to three dose groups. Different participants were assigned to each group and then assessed for a specific biomarker. What is the critical value for the one-way between-subjects ANOVA at a .05 level of significance?
3.47
132. If β1 = 0.003, then a 100 unit increase in X increases the odds of Y by ______% after adjusting for the other independent variables in the model.
35%
31. The following is a summary of a one-way between-subjects ANOVA: F(2, 37) = 3.42, p < .05. How many participants were observed in this study?
40
74. If b = -0.57, My = 2.75, and Mx = 5.25 for a set of data points, then what is the value of the y-intercept for the best-fitting linear equation?
5.74
115. In a case-control study, the OR is defined as:
the odds of exposure among cases/odds of exposure among controls
86. If the coefficient of determination is 0.30 and the sum of squares regression for an analysis of regression is 210, then what is the value of SSY?
700
33. The Kruskal-Wallis test can be used to:
All of the above A) compare a ranked outcome by race group (W, AA, Other). B) compare a skewed continuous variable by a categorical variable with four levels. C) compare an interval/ratio variable across more than two groups D) All of the above
57. Outliers can change the _____ of a correlation.
All of the above A) direction B) strength C) sign (+, -) D) all of the above
122. A RR = 3.6 means:
All of the above A) exposure is associated with an increased risk of disease B) risk in the exposed group is 3.6 times that in the unexposed C) compared to the unexposed, there is a 3.6 fold increase in risk among the exposed D) all of the above
108. What test is used to evaluate the association between exposure and disease in a 2x2 table?
B and C B) chi-square test of association C) Fisher's Exact test
20. A researcher conducts two studies on self-perception. In Study 1, 24 participants rate how positively they view themselves (on a 5-point scale) in one of three groups (n = 8 per group). In Study 2, the researcher conducts a similar study, except that k = 3 and n = 8. If SSB = 28 and SSE = 42 in both studies, then in which study will the decision be to reject the null hypothesis at a .05 level of significance for a one-way between-subjects ANOVA?
Both
131. The measure of effect estimated by a logistic regression model is:
Both A and B A) Odds Ratio. B) exp(β1)
123. An OR can be used as a measure of effect for a:
Both A and B A) cohort study. B) case-control study C) Both A and B.
140. An independent variable that is significant in both univariate and multivariate logistic regression models is:
Both A and B A) predictive independently of the other variables being modeled B) not due to its association with the other predictors
9. The degrees of freedom for the between-groups variability is called
Both A and C A) degrees of freedom numerator C) degrees of freedom between-groups
3. A lowercase k is used to denote
Both A and C A) the number of groups in a study C) the number of levels of the factor in a study
64. Which of the following would not be reported for a correlation?
C) the critical values for each test
25. Following a significant one-way between-subjects ANOVA in which k > 2, what is the next appropriate step?
Conduct post hoc tests.
37. A post-hoc test that can be used following a significant Kruskal-Wallis test is:
Dunns Q test
27. Which of the following post hoc tests is associated with the greatest power to detect an effect?
Fisher's LSD test
29. Post hoc tests are computed
Following a significant ANOVA test to make pairwise comparisons.
59. The Spearman rank-order correlation coefficient is a measure of the direction and strength of the linear relationship between two ________ variables.
Ordinal
126. Consider an analysis of a 2x2 table. The following is an example of confounding:
RR=1.0 for all subjects, RR=6.3 for males, RR=6.1 for females.
127. Consider an analysis of a 2x2 table. The following is an example of effect modification:
RR=2.0 for all subjects, RR=0.3 for males, RR=4.2 for females.
34. The Kruskal-Wallis test relies on:
Ranked data
18. A researcher randomly assigned 16 rodents to experience one of four levels of shock (n = 4 per group) following the illumination of a visual cue. If SSB = 24 and SSW = 48, then what was the decision at a .05 level of significance for a one-way between-subjects ANOVA?
Retain the null hypothesis.
43. A researcher measures the following correlation between cups of coffee consumed daily and daily work schedule. Which description best explains the relationship between these
The more a person works, the more coffee he or she tends to drink.
28. Which of the following post hoc tests is associated with the least power to detect an effect?
Tukey's HSD test
87. In a sample of 22 participants, suppose we conduct an analysis of regression with one predictor variable. If F = 4.07, then what is the decision for this test at a .05 level of significance?
X does not significantly predict Y.
106. The scores or data points for a regression analysis are typically reported in
a scatter plot
90. The standard error of estimate is used as a measure of the ________ in predictions using the equation of a regression line.
accuracy
118. An OR = 0.8 means:
cases are 0.8 times more likely to have been exposed as controls
110. A case-control study is characterized by:
choosing the cases and controls, identifying exposure prior to study.
109. A cohort study is characterized by:
choosing the study group, identifying exposure, following participants over time for disease.
121. A RR = 0.8 means:
compared to the unexposed, there is a 20% decrease in risk among the exposed
101. To standardize the beta coefficients, we first
convert the original data to standardized z scores
105. To summarize any type of regression analysis, we report each of the following except the
critical values
97. One key advantage for including multiple predictor variables in the equation of a regression line is that it allows you to
detect the extent to which two or more predictor variables interact
128. The dependent variable in logistic regression is:
dichotomous.
137. When given a dataset to analyze, the first step is to:
generate descriptive statistics and visual plots
50. The assumption that there is an equal variance or scatter of data points dispersed along the regression line is referred to as
homoscedasticity
5. Without changing the value of error variance, the ________ the between-groups variability, the more likely we are to reject the null hypothesis.
larger
53. Which of the following is the assumption that the best way to describe the pattern of data is using a straight line?
linearity
135. The coefficients in a logistic regression model are estimated using:
maximum likelihood estimation
69. Which of the following is used to determine the linear equation that "best fits" a set of data points?
method of least squares
130. A multiple logistic regression model has:
more than one independent variable
40. Which of the following indicates the strongest correlation?
r = -0.90
78. Using an analysis of regression, the variability in Y that is predicted by X is measured by the
regression variation
36. The test statistic used for the Kruskal-Wallis test follows which distribution?
the chi-square distribution with (k-1) degrees of freedom
112. Relative risk is computed as:
the proportion with disease among the exposed/proportion with disease among the non-exposed.
103. In addition to evaluating the significance of a multiple regression equation, we also should consider:
the relative contribution of each factor
67. A researcher measures the extent to which the speed at which people eat (in minutes) predicts calorie intake (in kilocalories). Which factor is the predictor variable in this example?
the speed at which people eat
91. What is the computation for the standard error of estimate?
the square root of the mean square residual
100. The value of b1 and b2 are referred to as
unstandardized beta coefficients
44. The denominator of the correlation coefficient measures the extent to which two variables
vary independently
49. A researcher measures the following correlation: r = -0.21. What is the value of the coefficient of determination?
0.04
10. The degrees of freedom for error is called
All of the Above A) degrees of freedom error B) degrees of freedom denominator C) degrees of freedom within-groups D) all of the above
54. Which of the following is a limitation for interpreting a correlation?
All of the above A) Correlations do not demonstrate cause-and-effect. B) Outliers can change the direction and/or strength of the correlation. C) Conclusions should not be drawn beyond the range of scores measured. D) All of the above
136. An appealing feature of logistic regression modeling is that:
B and C B) it allows for the consideration of multiple independent variables C) it allows for the estimation of ORs that are adjusted for the other predictors in the model
81. The more that the variability in ____ is associated with regression variation, the more likely it is that X predicts Y.
Y
120. A RR = 1.0 means:
All of the above A) exposure is not associated with risk of disease B) risk in the exposed group is the same as that in the unexposed C) compared to the unexposed, there is no increase in risk among the exposed D) all of the above
116. An OR = 1 means
All of the above A) exposure is not associated with the risk of disease B) the odds of exposure in cases is the same as the odds of exposure in controls C) cases are just as likely to have been exposed as controls. D) all of the above
65. Select the description below that identifies the following correlation: r = .28, p < .01.
All of the above A) the correlation is positive B) the correlation is statistically significant C) the coefficient of determination is .08 D) all of the above
119. An OR = 3.6 means:
All of the above A) exposure is associated with an increased risk of disease B) the odds of exposure in cases is 3.6 fold the odds of exposure in controls C) cases are 3.6 times more likely to have been exposed as controls D) all of the above
82. Which of the following statements is true regarding the sources of variation present in an analysis of regression?
The closer that data points fall to the regression line, the more the variance in Y will be attributed to regression variation.
16. A researcher divides participants into groups that will engage in low, moderate, or intense levels of exercise. The total calories consumed by participants following the exercise are then recorded. What type of statistical design is appropriate for this study?
a one-way between-subjects ANOVA
70. Which of the following is used to determine the significance of predictions made by a best fitting linear equation?
analysis of regression
96. A statistical method that includes two or more predictor variables in the equation of a regression line to predict changes in a criterion variable is called
multiple regression
84. The degrees of freedom associated with residual variation are equal to
n - 2
88. A researcher computes the following analysis of regression table. Based on the data given, what is the decision for this test at a .05 level of significance? (Note: Complete the table first.) Source of Variation SS df MS F Regression 1 28 Residual 118 19
significantly predicts Y.
17. Homogeneity of variance is an assumption for the one-way between-subjects ANOVA. What does this assumption mean?
that the variance is equal in each population from which samples are selected
60. The appropriate correlation coefficient for measuring the direction and strength of the linear relationship between two ranked or ordinal variables is
the Spearman correlation coefficient
11. A researcher compares differences in creatinine between participants in a three treatment groups. If she observes 15 participants in each group, then what are the degrees of freedom for the one-way between-subjects ANOVA?
(2, 42)
46. A researcher measures the relationship between narcissism and willingness to help. If SSXY = 240, SSX = 320, and SSY = 410, then what is the value of the correlation coefficient?
0.66
139. A limitation of logistic regression is:
All of the above A) the coding of categorical independent variables can be difficult to interpret B) the choice of independent variables is not always straight forward C) it requires a large sample size D) All of the above
42. The numerator of the correlation coefficient measures the extent to which two variables
Both A and C A) vary together C) covary
124. A RR can be used as a measure of effect for a:
Cohort Study
58. A correlation coefficient can ______ demonstrate cause.
Never
22. A researcher computes the following one-way between-subjects ANOVA table for a study where k = 3 and n = 12. State the decision at a .05 level of significance. (Hint: Complete the table first.) Source of Variation SS df MS F Between groups 120 Within groups (error) 780 Total
Retain the null hypothesis.
76. Which of the following is not needed to compute the slope using the method of least squares?
SSY
48. Suppose a correlation is computed in each of two samples. If the value of SSXY is the same in each sample, and √SSXSSY is larger in Sample 1, then in which sample will the value of the correlation coefficient be larger?
Sample 2
51. What is the problem with the following data for computing a correlation? Factor 1 Factor 2 3 3 3 3 3 3 3 3 3 3
The correlation coefficient will equal 0 because it violates the assumption of normality.
98. Which of the following equations is appropriate for a linear regression with three predictor variables?
Y ' b1 X 1 b2 X 2 b3 X 3 a
80. Both sources of variation in an analysis of regression measure the variability in A) X and Y B) X only C) Y only
Y only
55. An unanticipated variable not accounted for in a research study that could be causing or associated with observed changes in one or more measured variables is called
a confound variable
7. A researcher notes that the variability attributed to difference between group means is quite large. Which source of variation is the researcher referring to?
between-groups
66. A researcher measures the extent to which time spent watching educational preschool television programming predicts success in school. Which variable is the outcome variable in this example?
success in school
102. The equation for the standardized regression equation is
zY β1 ( zX1 ) β2 ( zX 2 )
73. If SSXY = -16.32 and SSX = 40.00 for a set of data points, then what is the value of the slope for the best-fitting linear equation?
-0.41
61. A researcher measures the correlation in rankings for a sample of restaurants and consumers' rankings of their favorite restaurants. If ΣD2 = 96 and n = 12, then what is the value of the correlation coefficient?
0.66
85. If the coefficient of determination is 0.32 and SSY = 150, then what is the sum of squares residual for an analysis of regression?
102
23. In a study with four groups and 10 participants in each group, the sum of squares for the between-groups source of variation is 60. What is the value for the mean square between-groups in this study?
20
12. A researcher conducts a study in which k = 5 and N = 80. What are the degrees of freedom between-groups for the one-way between-subjects ANOVA?
4
107. A case-control study is performed to study the relationship between esophageal cancer and an exposure (exposure A). Esophageal cancer is a very rare disease (prevalence <<< 1%) in the general population. The cases were 100 persons with the cancer of whom 30 had exposure A. The cases were a random sample of individuals with cancer in the population. The controls were 200 persons without the cancer of whom 18 had exposure A The controls were a random sample of individuals without cancer in the population. Fisher's exact test gave a p-value of .01. What is the estimated odds ratio of cancer for those who had exposure A relative to those who did not have exposure A?
4.3
14. Which of the following is an assumption for computing a one-way between-subjects ANOVA?
All of the above A) The population being sampled from is normally distributed. B) Participants were selected to participate using a random procedure. C) One observation has no effect on the likelihood of another observation. D) all of the above
24. When the variability attributed to between-groups is equal to the variability attributed to error, then the value of the test statistic for a one-way between-subjects ANOVA is,
Equal to 1.
19. A researcher assigns 21 subjects to 3 treatment groups. An equal number of participants are assigned to each group. If F = 4.08 for this study, then what was the decision at a .05 level of significance for a one-way between-subjects ANOVA?
Reject the null hypothesis.
21. A researcher computes the following one-way between-subjects ANOVA table. State the decision at a .05 level of significance. (Hint: Complete the table first.) Source of Variation SS df MS F Between groups 32 4 Within groups (error) 122 45 Total
Reject the null hypothesis.
62. A researcher measures the correlation of the time it take participants to complete two tasks purported to measure the same cognitive skill. Participant times are converted to ranks from fastest to slowest. If ΣD2 = 165 and n = 20, then what is the decision for this correlation test?
Reject the null hypothesis.