<result> <column_name>material_delta</column_name> <family>additive</family> <subtype>intensive_additive</subtype> <confidence>0.65</confidence> <requires_user_confirmation>true</requires_user_confirmation> <sum_meaningful>false</sum_meaningful> <signed>false</signed> <reducers> <primary_reducer>mean</primary_reducer> <valid_reducers>mean, min, max, median</valid_reducers> <excluded_reducers>sum (summing a delta/change level 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 name "material_delta" suggests a change/difference quantity, which could be either an additive delta or a multiplicative ratio. The values [0.005, 0.02] are small positive numbers centered far from both 0.0 and 1.0, ruling out ratio_multiplicative (which would cluster near 1.0). This appears to be an absolute change level or intensity measure. On a DIMENSION table, this is most likely an intensive_additive property (a characteristic delta/threshold per entity), where averaging across entities is meaningful but summing is not. Confidence is moderate due to limited sample size and ambiguity in whether "delta" refers to a stored change level versus a ratio. </reasoning> <table_type_conflict>false</table_type_conflict> </result>