Skip to Main Content
We often hear that global software engineering teams are affected by time differences. While there is considerable research on the difficulties of distance, culture and other dimensions, there has been little research that isolated the impact of just time differences. The research question that guides us is whether there are gradual differences across time zones that impact team performance. In this study we conducted a laboratory experiment with 42 dyadic teams. The teams were randomly assigned into 4 time zone overlap conditions: full overlap, 2/3 overlap, 1/3 overlap and no overlap. Using a fictional map task, we found that participants' perceptions of process are unrelated to actual objective performance measures of speed and accuracy. Consistent with our expectations, we found that a small time separation has no effect on accuracy, but that more time separation has a significant effect on accuracy. Also consistent with our expectations, we found that a small amount of time separation has a significant effect on production speed. However, contrary to our expectations, we found that further increases in partial overlap have less significant effects on speed, and when there is no overlap speed actually increases, albeit not significantly - a "U-shaped" effect.
Date of Conference: 27-30 Aug. 2007