Before class, please read these two journal articles and try to
understand their statistical analyses. Come prepared to discuss
the following review questions.
Myers & Hayes, 2006
Myers, D., & Hayes, J. A. (2006).
Effects of therapist general self-disclosure and countertransference
disclosure on ratings of the therapist and session.
Psychothearpy: Theory, Research, Practice, Training,
43(2), 173-185.
- How many 'cells' in this study?
What are each of the predictors and their levels?
- "Pearson's correlation coefficients among the dependent measures
ranged from .13 to .69" (p. 179)
What assumption of parametric data are they addressing with this
information?
- Using the graph of the interaction between Working Alliance
and Self-Disclosure on Expertness ratings as a guide, draw
another graph that presents the same data in a different way.
Horowitz, et al., 2007
Horowitz, J. L., Garber, J., Ciesla, J. A., Young, J. F., & Mufson, L.
(2007).
Prevention of depressive symptoms in adolescents: A randomized trial
of cognitive-behavioral and interpersonal prevention programs.
Journal of Consulting and Clinical Psychology,
75(5), 693-706.
- Given the core assumptions for ANCOVA (both statistical and
conceptual), do you think these authors make a good case for including
pre-treatment depression scores as a covariate? Why or why not?
(See lecture06, 16Oct, on ANCOVA.)
- The author's third moderation hypothesis (c) expected to see
a stronger effect for the interpersonal program on depressive
symptoms if the participant had higher sociotropy scores.
Does the graph of the interaction (Figure 3, p. 700) appear to
support that hypothesis?