Imagine that you are a member of the research firm described in this exercise. Imagine that you have not yet conducted the study, but are simply in the planning phases. Imagine that you will not measure all criterion variables, but will instead focus on just one: symptoms of stress. The university have told you that they predict that the on-campus students will score significantly higher on this variable compared to the off-campus students. They want you to design the study and analyse the data in such a way as to maximize the likelihood of finding a statistically significant difference between these two groups on this variable (if there is, in fact, such a difference in the population). You had initially planned to do the following:
- You would obtain scores from about 170 on-campus students and about 80 off-campus students.
- You would perform an independent-samples t test. It would be a two-tailed test and you would set alpha at α = .05.
Given the interest expressed by the University (as summarized above), describe two ways that you would modify either (a) the nature of your study or (b) the nature of your analyses so as to increase the power of your test. How could you change the study (or the analysis) so as to increase the likelihood of obtaining statistically significant results?