Performing Univariate Analysis For Given Data

Question

Data: A MANOVA is performed to study the effect of demand stress and training in behavioral desensitization on communication apprehension. The sample has 2 dependent variables (DVs) and two independent variables (IVs). The first IV has 3 levels, the second IV has 2. These two IVs in factorial arrangement produce 6 cells; 3 participants are assigned to each cell, so there are 3 X 6 = 18 participants in the study. The SPSS data file (stress.sav) is available in Blackboard.

Independent variables:

1. LMH refers to level of demand stress

• three groups = Low, Moderate, and High (1, 2 and 3);

1. TREAT refers to the use of response desensitization vs. instruction in Latin

• two groups = Treatment vs. No Treatment (0 and 1).

Dependent variables (both are reverse coded so that the higher the score the LESS apprehension):

1. DV1 writing apprehension and
2. DV2 public speaking apprehension

Procedure:

Open the data file in SPSS and follow the following instructions:

• Select ANALYZE/GENERAL LINEAR MODEL/ MULTIVARIATE from the menu;
• Move DV1 and DV2 in the Dependent Variables area and LMH and TREAT in the Fixed Factors area;
• Select Options, then check "Descriptive statistics" and "Effect size estimates" under Display; click Continue;
• In the GLM Multivariate window, click OK.
  

1. EXERCISE. Using the MANOVA described above, answer the following questions:

1. What are the Wilk's Lambda coefficients for each main effect and the interaction? (Report the appropriate value, F, df1, df2, p from the Multivariate Tests table)
2. How would you interpret these effects?
3. Why is the multivariate procedure better in this case than two univariate (ANOVA) tests?
4. Given the effects detected, is the sample size in this multivariate ANOVA example adequate?

Summary

The question belongs to Statistics and it discusses about performing univariate analysis for the data given.

Total Word Count 884