- Is there a sufficient correlation between the dependent variables to justify the use of MANOVA?
- Was the assumption of Equality of Covariance Matrices violated? Explain.
The assumption of Equality of Covariance Matrices was not violated. Based on the table titled Box's Test of Equality of Covariance Matrices indicated that (p = .463) thus there was no violation of the assumption.
- Is there a statistically significant multivariate interaction effect? Identify the dependent variable(s) of this interaction effect.
Yes, there a statistically significant multivariate interaction effect and the dependent variable that has been identified of this interaction effect is emotion.
- What would be the proper follow-up tests for a statistically significant interaction effect?
The proper follow up test for a statistically significant interaction effect would be to carry out a one way ANOVA for males with treatment as the independent variable. Another one way ANOVA for females with treatment as the independent variable would also be carried out. In addition, post hoc tests should also be carried out on the data.
- Identify the proper post hoc analyses for any statistically significant univariate effects. Explain your answer.
Based on the answer provided in question 2 it is clear that the assumption of Equality of Covariance Matrices was not violated thus the most appropriate post hoc test to be carried out in this case for any statistically significant univariate effects is the Least Significant Difference (LSD) test.
- Is there a statistically significant multivariate gender effect on the dependent variate?
Yes, there is a statistically significant multivariate gender effect on the dependent variate.
- Why would a researcher conduct a MANOVA instead of several ANOVAs?
A researcher would prefer to conduct a MANOVA instead of several ANOVAs mainly because a MANOVA is a more powerful tool as compared to the ANOVA. This implies that the MANOVA will clearly indicate any variance that may occur.
- Write a Results section for this research
The results of this research indicate that there is a sufficient correlation between the dependent variables (emotion and worry). This indicates that a MANOVA could be used for the research. The study also indicates that gender has a significant effect on the dependent variate. This indicates that the symptomatic improvement (measured by indexes of both worry and general emotion) of three different anxiety treatments for females and males is greatly affected by gender. This thus indicates that gender moderates the effectiveness of the treatments. Based on the general linear model Treatment * Gender it is clear that males record higher means for all treatments as well as for both dependent variables.
References:
Leech, N. L., Barrett, K. C., & Morgan, G. A. (2005). SPSS for intermediate statistics: Use and interpretation. Mahwah, N.J: Lawrence Erlbaum.
Hinton, P. R. (2004). SPSS explained. London: Routledge.
