```xml <result> <column_name>table_concentration</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 concentration levels 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 "concentration" indicates an intensive physical property (a level/density measure). While the median is 1.0 and mean is 0.87 (near 1.0), the range [0.20, 1.0] and the semantic meaning of concentration as a proportion/density strongly indicate this is an absolute level measure, not a ratio that compounds. Concentration values represent states at fixed points, where averaging (not summing or multiplying) is the natural aggregation. The negative log_symmetry_gain confirms the distribution is not log-normal, further supporting additive classification. </reasoning> <table_type_conflict>false</table_type_conflict> </result> ```