```xml <result> <column_name>Region: Midwest</column_name> <family>additive</family> <subtype>intensive_additive</subtype> <confidence>0.85</confidence> <requires_user_confirmation>false</requires_user_confirmation> <sum_meaningful>false</sum_meaningful> <signed>false</signed> <reducers> <primary_reducer>mean</primary_reducer> <valid_reducers>mean, min, max, median, distribution</valid_reducers> <excluded_reducers>sum (summing percentages across entities is meaningless), product (not a ratio measure)</excluded_reducers> <average_semantics>arithmetic_mean</average_semantics> </reducers> <inner_outer_measure> <inner_measure_reducer>min</inner_measure_reducer> <outer_measure_reducer>max</outer_measure_reducer> </inner_outer_measure> <multiplicative_partner_derivable>false</multiplicative_partner_derivable> <reasoning> DIMENSION table with no sequence axis strongly gates against multiplicative. The column contains percentage values (0.79% to 10.40%) representing regional proportions or shares - these are intensive levels (snapshots of allocation) rather than growth rates or ratios. The dtype is object (stored as strings with % symbols), and the values represent absolute allocation levels at a point in time. Summing regional percentages across entities is meaningless; the natural aggregation is mean to find average regional representation. This is an intensive_additive measure: an absolute level that should be averaged, not summed. </reasoning> <table_type_conflict>false</table_type_conflict> </result> ```