In order to justify aggregation at the group level of dependent and independent
variables we used a measure of within group agreement (rWG) developed by James et al.
(1984). rWG has been suggested as an indicator of the extent to which group members
agree in their evaluations of the variables considered in the study. The aim of this
analysis is to provide justification to the aggregation of measures within each team.
We used a uniform distribution to represent random responses from team members.
For all teams, it was observed that rWG ..80, suggesting strong agreement. Thus,
based on the rWG estimates, aggregating the individual data to the team level can be
justified.
Exploratory Factor analysis was conducted on the independent variables to explore
the multidimensionality of the data. Results justified the use of the two constructs (goal commitment and conflict management). Due to the limited size of the sample, only
exploratory factor analysis was conducted. The extraction method used was maximum
likelihood with oblique rotation. Oblique rotation derives factor loadings based on the
assumption that the factors are likely to be correlated, which is the case in this study.
Table II shows the loadings that resulted from the factor analysis, which support a
two-factor structure for the independent variables consistent with the conceptual
model in Figure 1.
The scree plot in Figure 2 further supports the two-factor structure as indicated by
the elbow on the graph seen after the second factor. The red line just before the elbow
shows components with eigenvalues greater than 1. The first factor accounted for
69 percent of the variance, whereas the second accounted for 20 percent.