When score scales from different language versions are linked, choose an equating design and analysis procedures that suit the data. Check that any common items function similarly in both groups and consider dropping those that do not, using tools such as delta plots or plots of difficulty and discrimination estimates. Treat exchangeable scores as conditional on construct, method and item equivalence, and if a link is attempted, support the chosen design with evidence; alternatives include linking through items judged essentially equivalent or through bilingual test takers who sit both forms in counterbalanced order so that a conversion table can be built. Work through the full sequence of steps for score equivalence and report the evidence for each.
The graph holds this control, the 0 it maps to, and the evidence behind each claim, over MCP and REST.