Topic 6: Correlation and Simple Linear Regression
Consider the following news headline, “Cigarette Smokers Make Lower College Grades Than Nonsmokers.” The article goes on to say that researchers at a university collected information on the rate of cigarette smoking (total number of cigarettes smoked in a one-month period) and the college student’s current GPA for completed college courses. They found that those with higher levels of cigarette use had lower GPAs, and vice versa. What can you rightfully conclude from this? Why? Be specific about what this kind of information may indicate and mean.
A pharmaceutical company conducts a study to evaluate the relationship between length of time taking drug X and size of tumor. They enlist 2,400 patients with the same type of benign tumor who have been on drug X for varying lengths of time (measured in days). They find that the longer the patient is on the drug, the smaller the patient’s tumor size (measured in volume). In fact, they report the results as r(2398) = -.21, with significance of p < .0001. What is your initial perception of this information? What are other ways, beyond looking at the p-value alone, you could use to try to interpret what this outcome of r(2398) = -.21 tells us about the relationship between length of time taking drug X and the size of the tumor? Support your responses.
Topic 7: Multiple Regression
Describe a study you might design that could use multiple regression. What are the variables? What are the null and alternative hypotheses? Why would multiple regression be appropriate for this design? What would multiple regression indicate in relation to your research question? What would it not indicate? Be specific and use your sample study to illustrate your answers. This response does not need to be supported with external research.
Considering your personal research interest, which of the methods described in this course do you believe will be most useful to your research? Why? This response does not need to be supported with external research.
Topic 8: Multivariate Comparison of Means: The One-Way and Factorial MANOVA
Consider your dissertation research interests. Identify one categorical/nominal scale IV with more than 2 categories, and three DVs that are measured on continuous scales. Think of DV measures that probably are moderately correlated with each other because they are measuring different components of the same or similar concepts (e.g., three different measures of academic performance). What information would a one-way MANOVA provide you? What more would you want to know if you get significant results in the MANOVA? Why would this be significant to your research? (Research support is not required for this question.)
Using the DC Network, locate information about the 10 Strategic Points, the Prospectus template, the Proposal template, and the Dissertation template. What is the purpose of each of these documents? How do you anticipate interacting with these documents? Explain.