Driving Anger Scale: How reliable are subscale scores? A bifactor model analysis

- This study examined the utility of creating and interpreting subscale scores for the Driving Anger Scale.
- A bifactor model approach was used.
- The results show that there is more common than specific variance in the DAS subscales.
- The explained common variance (ECV) for the general driving anger factor was large (63.81%).
- The only subscale that seemed to provide some specific variance was Illegal driving.

The utility of creating and interpreting subscale scores for the Driving Anger Scale (DAS; Deffenbacher, Oetting, & Lynch, 1994) was examined in this study. Using data from 414 drivers (built using the snowball sampling approach and consisting of university students and their peers), we verified whether subscale scores provide reliable information after accounting for a general driving anger factor, using a bifactor model analysis. The results show that the item pool can be considered unidimensional, as there is more common than specific variance in the DAS subscales. Also, the explained common variance (ECV) for the general driving anger factor was large (63.81%). The only subscale that seemed to provide some specific variance was Illegal driving. Overall, our results suggest that any interpretation of the DAS subscales should be done with caution, since their specific variance appears to be rather scarce.