A new metric for assessing group level participation in fluid teams

Equality of participation is an important factor in the success of multidisciplinary science teams. The typical measure, standard deviation, fails to provide unbiased estimates across groups of different sizes or within groups that change size over time. We propose a new metric of participation equality that takes into account real-world teams that have members come and go naturally over the course of a meeting. This new metric ranges from zero (entirely equal participation) to one (entirely dominated by a single person). This metric is at the group level and for whatever period of time the researcher specifies. Using 11 hours of transcribed utterances from informal, fluid, co-located meetings during the Mars Exploration Rover (MER) mission, we computed this metric for 549 blocks of time. We found that this metric had good convergent validity via having strong positive correlations with both a standard deviation metric of words spoken and participation equality as assessed by two independent coders. It also had good discriminant validity by being uncorrelated with positive and negative affect words, including anxiety and sadness words. Furthermore, when only fluid groups were examined, it maintained a strong correlation with coder-assessed participation. Future research can take advantage of this metric in other settings where team membership is fluid and equality of participation is of interest.