1. A researcher was interested in investigating the impact of Food ...

1. A researcher was interested in investigating the impact of Food and Drink in a Computer Classroom. The amount of food and drink consumed was correlated with alertness in class. The researcher also wanted to control for the variable of interest in subject matter when examining this relationship. Researcher A wanted to use bivariate correlations to examine this problem. S/he divided the sample into two groups that indicated different levels of interest (those taking the course as a requirement and those taking as an elective) and ran a bivariate correlation for each group. Researcher B used partial correlation to examine this issue. S/he used an interest scale (scaled 1 to 5) to measure interest in the subject matter and used this score as a control variable before examining the relationship food/drink consumed and alertness. Which researcher is addressing the problem most appropriately. In your response, provide at least two convincing reasons for your selection indicating why this approach is better than the other one. 2. Examine the following Venn diagram carefully. 2a. Which proportion of this diagram represents the partial correlation between X1 and Y, controlling for the influence of X2? A. (B+D) divided by (A+B+C+D) B. (B+C) divided by (A+B+C+D) C. C divided by (A+C) D. B divided by (A+B) 2b. Which part(s) of this diagram represent(s) the portion of the dependent variable explained by semi-partial correlation with X2, controlling for the influence of X1? A. B B. B & D C. C D. C & D 2c. Which part(s) of the diagram represent(s) represent the overall R2 using both X1 and X2 to predict Y? Y A B C D E F G X1 X2 3. Examine the following results from a multiple regression analysis. There are three predictor variables that have been used to predict alertness in this class. X1 - Amount of Caffeine (prior to class, of course) X2 - Amount of time spent talking about college sports or political candidates X3 - Hours of sleep the night before class B .325 3.24 4.2 95% C.I. of B .013 to .525 1.5 to 4.97 -2.22 to 6.32 t test (Sig.) .467 (.823) 3.50 (.003) 2.74 (.032) Beta .114 .800 .695 Zero-order Correlation .225 .562 .499 Partial .022 .332 .233 Semi-Partial .011 .257 .125 VIF 2.50 2.75 8.90 Tolerance .825 .752 .325 Constant (Intercept) = 7.25 a. Select the best predictor and indicate at least three reasons for this selection. b. Identify the variable that appears to exhibit the most multicollinearity and indicate at least two reasons why. c. Write the regression equation that would be used to predict alertness, based on the three predictor variables above. 4. Describe a situation in which Analysis of Covariance (ANCOVA) would be used in your area of specialization. a, Describe the DV and how it will be measured. b. Describe the covariate(s) and why it(they) is(are) appropriate. c. Describe the conditions of the IV (treatment variable) and why this variable would impact the DV after accounting for the covariate(s). d. Describe what evidence will allow you to conclude that the IV had an impact on the DV, after controlling for the covariate(s). e. Explain how regression is used to adjust the group means prior to conducting a oneway ANOVA.