```xml <result> <column_name>Perceptions of corruption</column_name> <family>additive</family> <subtype>intensive_additive</subtype> <confidence>0.92</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 an intensity/perception score is meaningless), product (not a ratio)</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> Gate prior: DIMENSION table with no sequence axis makes multiplicative highly implausible. The column is a perception/corruption index score bounded [0, 0.453], centered at mean=0.11, median=0.085 - far from both 0.0 and 1.0. This is an intensive measure (a level/score) where summing across entities is meaningless; averaging is the natural aggregation. No naming tokens suggest ratio/rate. The presence of zeros and the bounded positive range confirm this is an absolute intensity measure, not a ratio. </reasoning> <table_type_conflict>false</table_type_conflict> </result> ```