In a recent publication, the authors report the following transformation when aggregating across three different scales:
Cognitive style level was used as a control variable and captured as
(sqrt(OV) + sqrt(SV) + sqrt(V))^2which is similar to, but a more robust measure than, the sum of the means of the three cognitive styles, particularly when team size varies. It captures both the level and the range of the member cognitive styles in the team (Matousek 2002, Rudin 1987).
The referenced publications are both lengthy real analysis textbooks, which don't provide a clear intuition for why this is valuable. What is the intuition behind this? Are there more robust transformations?