Compare scores between populations only where the reporting scale has been shown to reach the required level of invariance. Linking or equating versions onto one scale demands substantial samples and evidence that bias in the construct, in the method and in individual items has been ruled out. Unit equivalence alone is not enough for direct comparison; that needs scalar or full score equivalence, where both the measurement unit and the origin match across groups. Interpretation should stay within the strength of the evidence, so avoid comparative claims when invariance has not been demonstrated.
The graph holds this control, the 0 it maps to, and the evidence behind each claim, over MCP and REST.