Home > Advanced Correlational Strategies > Practice Questions >

1 .       Regression analyses are only appropriate when there is a ______ relationship between variables. [Hint]

 A. Nonlinear B. Linear C. Curvilinear D. Positive

2 .       Assume a researcher constructs a regression equation, which uses SAT score to predict high school GPA. In this example, SAT score is the __________________, and college GPA is the _______________. [Hint]

 A. dependent variable; predictor variable B. predictor variable; dependent variable C. dependent variable; criterion variable D. criterion variable; dependent variable

3 .       In ___________, several predictor variables are entered into a regression equation, all at the same time. [Hint]

 A. step-wise regression B. linear regression C. hierarchical regression D. simultaneous regression

4 .       In ___________, several predictor variables are entered into a regression equation, in an order that is predetermined by the researchers, based hypotheses they want to test. [Hint]

 A. step-wise regression B. linear regression C. hierarchical regression D. simultaneous regression

5 .       In ___________, several predictor variables are entered into a regression equation, starting with the predictor variable that by itself best predicts the criterion variable. [Hint]

 A. step-wise regression B. linear regression C. hierarchical regression D. simultaneous regression

6 .       In step-wise regression, the regression equation stops adding predictor variables [Hint]

 A. when remaining predictor variables no longer enhance our ability to predict unique variance in the criterion, over those already entered into the equation. B. when remaining predictor variables begin to enhance our ability to predict unique variance in the criterion, over those already entered into the equation. C. when predictor variables reach a point where they begin to significantly correlate with one another. D. when five (5) predictor variables have already been entered into the equation, regardless of what the remaining predictors can contribute.

7 .       In a study of the relationship between 6 measures of social confidence and romantic relationship stability, R2 is calculated to be .40. This means that [Hint]

 A. 4% of the variance in relationship stability is predicted by the 6 measures of social confidence. B. 16% of the variance in relationship stability is predicted by the 6 measures of social confidence. C. 40% of the variance in relationship stability is predicted by the 6 measures of social confidence. D. 60% of the variance in relationship stability is predicted by the 6 measures of social confidence.

8 .       Eron, Huesmann, Lefkowitz, and Walder (1972) conducted a ____________ to examine the direction of the relationship between watching television violence and aggressive behavior. [Hint]

 A. structural equation modeling B. cross-lagged panel design C. factor analysis D. hierarchical multiple regression

9 .       Although correlational studies prevent researchers from drawing cause-and-effect conclusions, the findings from the Eron, Huesmann, Lefkowitz, and Walder (1972) study were consistent with the idea that watching television causes people to be aggressive. This is because [Hint]

 A. children who were asked to watch violent television programs were more likely to act aggressively than children who were asked to watch non-violent television programs. B. the correlation between TV violence at Time 1 and Aggression at Time 1 was positive and significant. C. the correlation between TV violence at Time 1 and Aggression at Time 2, was greater than the correlation between Aggression at Time 1 and TV violence at Time 2. D. the correlation between Aggression at Time 1 and TV violence at Time 2, was greater than the correlation between TV violence at Time 1 and Aggression at Time 2.

10 .       The procedure that mathematically compares the correlation matrix implied by a particular hypothesized model to the real correlation matrix based on actual data collected, is called [Hint]

 A. structural equation modeling B. cross-lagged panel design C. factor analysis D. hierarchical multiple regression

11 .       __________ refers to a class of statistical techniques that are used to determine the minimum number of dimensions that do a reasonably good job in accounting for the observed relationships among a large number of variables. [Hint]

 A. structural equation modeling B. cross-lagged panel design C. factor analysis D. hierarchical multiple regression

12 .       There are hundreds of traits/dispositions on which people differ from one another. However, McCrae and Costa (1987) suggest that there are only _____ basic personality factors, and that most traits actually "load" (i.e., there is a significant correlation between item and factor) on those factors. [Hint]

 A. 3 B. 4 C. 5 D. 10

 Copyright © 1995-2010, Pearson Education, Inc., publishing as Pearson Allyn & Bacon Legal and Privacy Terms