Team Faultline Measurement

Ordinary diversity asks how much members vary on a single attribute, like the spread of ages or the mix of genders. A faultline asks whether several attributes line up to create a clean split into homogeneous subgroups. Two teams can have identical diversity on every separate attribute yet have completely different faultlines: in one the attributes reinforce each other to form a sharp us-versus-them divide, while in the other they cut across each other and blur any coalition. Lau and Murnighan argued that this alignment — captured by faultline strength — is what drives subgroup conflict, which is why measuring faultlines explains team outcomes that single-attribute diversity indices cannot.

Fau, introduced by Thatcher, Jehn, and Zanutto, is faultline strength: the proportion of total attribute variation explained by the best split into subgroups, ranging from zero (no clean split) to one (a perfectly aligned split). Faultline distance is the Euclidean separation between the resulting subgroup centroids on the splitting attributes. Strength tells you how cleanly the group can fracture; distance tells you how far apart the subgroups are once it does. Two groups can share the same Fau but differ in distance, so reporting both gives a fuller description of the cleavage than either alone.

The classic Fau index was built for a two-subgroup split, which is a limitation when real teams contain several potential coalitions. Meyer and Glenz addressed this by comparing the available measures and proposing an approach based on average silhouette width, a clustering criterion that evaluates how well members fit their own subgroup relative to others and can therefore identify the optimal number and composition of subgroups. This extends faultline measurement beyond the two-subgroup case, though it adds algorithmic choices — such as the distance metric and attribute weighting — that the researcher must justify.