3.4 Analysis
The main unit of analysis in this study was the team; therefore, data were not analyzed
at the individual level. Team level variables were calculated as the averages of the
individual team members’ values for all variables. We used a test for direct and
mediation effects with bootstrapping (Preacher and Hayes, 2004) using PASW/SPSS
version 20 to conduct the analysis. Simulation research shows that bootstrapping is
more powerful than the Sobel test and the causal steps approach to testing intervening
variable effects (MacKinnon et al., 2004; Williams and MacKinnon, 2008). The analysis
tested a simple mediation model with commitment to goals as independent variable
and performance and satisfaction as dependents. The mediator variable was conflict
management.
Table III shows data on descriptive statistics and correlations of the variables
evaluated in this study.