12.45 College students, anxiety, depression, and a MANOVA: Jason Nelson and Noel Gregg (2012) conducted a study of college students with disabilities. They reported that: “A 3 × 3 × 2 MANOVA was conducted to determine the effect of level of educational attainment (transitioning adolescents and college underclassmen—that is, freshmen and sophomores—and college upperclassmen—that is, juniors and seniors), disability status (dyslexia, ADHD, and ADHD/dyslexia), and gender on BDI-II and BAI scores” (p. 249). (BDI is the Beck Depression Inventory, which assesses symptoms of depression; BAI is Beck Anxiety Inventory, which assesses symptoms of anxiety.)

a. What are the independent variables in this study and what are their levels?

b. What are the dependent variables in this study?

c. Based on these variables, explain why the researchers conducted a MANOVA.

d. The researchers report that there was only one significant effect in this MANOVA. Specifically, “Tukey’s HSD post hoc tests indicated college underclassmen with disabilities had higher BAI (M = 9.10, SD = 8.37) and BDI-II scores (M = 9.30, SD = 8.92) than did transitioning adolescents with disabilities (BAI: M = 5.18, SD = 6.18; BDI-II: M = 4.50, SD = 5.40). The effect sizes for these group differences on the BAI (d = 0.50) and BDI-II (d = 0.59) were medium in magnitude.” i. Put these findings into plain English and in your own words. ii. Explain why the researchers conducted Tukey HSD tests. iii. Why did the researchers report Cohen’s d when they conducted a form of ANOVA (i.e., MANOVA)? (Hint: Think about the number of means in their comparisons.)